eduagarcia commited on
Commit
6da7311
1 Parent(s): fbd2a73

Add new column: Main Language

Browse files
.gitignore CHANGED
@@ -17,3 +17,4 @@ downloads/
17
  tasks_config/legal_config.yaml
18
 
19
  src/assets/model_counts.html
 
 
17
  tasks_config/legal_config.yaml
18
 
19
  src/assets/model_counts.html
20
+ languages.jsonl
app.py CHANGED
@@ -29,7 +29,8 @@ from src.display.utils import (
29
  fields,
30
  WeightType,
31
  Precision,
32
- Tasks
 
33
  )
34
  from src.envs import (
35
  API,
@@ -125,10 +126,11 @@ def update_table(
125
  type_query: list,
126
  precision_query: str,
127
  size_query: list,
 
128
  hide_models: list,
129
  query: str,
130
  ):
131
- filtered_df = filter_models(df=hidden_df, type_query=type_query, size_query=size_query, precision_query=precision_query, hide_models=hide_models)
132
  filtered_df = filter_queries(query, filtered_df)
133
  filtered_df = update_leaderboard_avg_scores(filtered_df, columns)
134
  df = select_columns(filtered_df, columns)
@@ -177,7 +179,7 @@ def filter_queries(query: str, filtered_df: pd.DataFrame):
177
 
178
 
179
  def filter_models(
180
- df: pd.DataFrame, type_query: list, size_query: list, precision_query: list, hide_models: list
181
  ) -> pd.DataFrame:
182
  # Show all models
183
  if "Private or deleted" in hide_models:
@@ -197,6 +199,7 @@ def filter_models(
197
  type_emoji = [t[0] for t in type_query]
198
  filtered_df = filtered_df.loc[df[AutoEvalColumn.model_type_symbol.name].isin(type_emoji)]
199
  filtered_df = filtered_df.loc[df[AutoEvalColumn.precision.name].isin(precision_query + ["None"])]
 
200
 
201
  numeric_interval = pd.IntervalIndex(sorted([NUMERIC_INTERVALS[s] for s in size_query]))
202
  params_column = pd.to_numeric(df[AutoEvalColumn.params.name], errors="coerce")
@@ -225,6 +228,7 @@ leaderboard_df = filter_models(
225
  type_query=[t.to_str(" : ") for t in ModelType],
226
  size_query=list(NUMERIC_INTERVALS.keys()),
227
  precision_query=[i.value.name for i in Precision],
 
228
  hide_models=["Contains a merge/moerge", "Flagged"], # "Private or deleted", "Contains a merge/moerge", "Flagged"
229
  )
230
 
@@ -289,6 +293,13 @@ with demo:
289
  interactive=True,
290
  elem_id="filter-columns-size",
291
  )
 
 
 
 
 
 
 
292
 
293
  leaderboard_table = gr.components.Dataframe(
294
  value=leaderboard_df[
@@ -319,6 +330,7 @@ with demo:
319
  filter_columns_type,
320
  filter_columns_precision,
321
  filter_columns_size,
 
322
  hide_models,
323
  search_bar,
324
  ],
@@ -335,6 +347,7 @@ with demo:
335
  filter_columns_type,
336
  filter_columns_precision,
337
  filter_columns_size,
 
338
  hide_models,
339
  search_bar,
340
  ],
@@ -343,7 +356,7 @@ with demo:
343
  # Check query parameter once at startup and update search bar + hidden component
344
  demo.load(load_query, inputs=[], outputs=[search_bar, hidden_search_bar])
345
 
346
- for selector in [shown_columns, filter_columns_type, filter_columns_precision, filter_columns_size, hide_models]:
347
  selector.change(
348
  update_table,
349
  [
@@ -352,6 +365,7 @@ with demo:
352
  filter_columns_type,
353
  filter_columns_precision,
354
  filter_columns_size,
 
355
  hide_models,
356
  search_bar,
357
  ],
@@ -455,6 +469,13 @@ with demo:
455
  value=ModelType.FT.to_str(" : "),
456
  interactive=True,
457
  )
 
 
 
 
 
 
 
458
 
459
  with gr.Column():
460
  precision = gr.Dropdown(
@@ -472,7 +493,6 @@ with demo:
472
  interactive=True,
473
  )
474
  base_model_name_textbox = gr.Textbox(label="Base model (for delta or adapter weights)")
475
-
476
  submit_button = gr.Button("Submit Eval")
477
  submission_result = gr.Markdown()
478
  submit_button.click(
@@ -485,6 +505,7 @@ with demo:
485
  private,
486
  weight_type,
487
  model_type,
 
488
  ],
489
  submission_result,
490
  )
 
29
  fields,
30
  WeightType,
31
  Precision,
32
+ Tasks,
33
+ Language
34
  )
35
  from src.envs import (
36
  API,
 
126
  type_query: list,
127
  precision_query: str,
128
  size_query: list,
129
+ language_query: list,
130
  hide_models: list,
131
  query: str,
132
  ):
133
+ filtered_df = filter_models(df=hidden_df, type_query=type_query, size_query=size_query, language_query=language_query, precision_query=precision_query, hide_models=hide_models)
134
  filtered_df = filter_queries(query, filtered_df)
135
  filtered_df = update_leaderboard_avg_scores(filtered_df, columns)
136
  df = select_columns(filtered_df, columns)
 
179
 
180
 
181
  def filter_models(
182
+ df: pd.DataFrame, type_query: list, size_query: list, language_query: list, precision_query: list, hide_models: list
183
  ) -> pd.DataFrame:
184
  # Show all models
185
  if "Private or deleted" in hide_models:
 
199
  type_emoji = [t[0] for t in type_query]
200
  filtered_df = filtered_df.loc[df[AutoEvalColumn.model_type_symbol.name].isin(type_emoji)]
201
  filtered_df = filtered_df.loc[df[AutoEvalColumn.precision.name].isin(precision_query + ["None"])]
202
+ filtered_df = filtered_df.loc[df[AutoEvalColumn.main_language.name].isin(language_query + ["None"])]
203
 
204
  numeric_interval = pd.IntervalIndex(sorted([NUMERIC_INTERVALS[s] for s in size_query]))
205
  params_column = pd.to_numeric(df[AutoEvalColumn.params.name], errors="coerce")
 
228
  type_query=[t.to_str(" : ") for t in ModelType],
229
  size_query=list(NUMERIC_INTERVALS.keys()),
230
  precision_query=[i.value.name for i in Precision],
231
+ language_query=[i.value.name for i in Language],
232
  hide_models=["Contains a merge/moerge", "Flagged"], # "Private or deleted", "Contains a merge/moerge", "Flagged"
233
  )
234
 
 
293
  interactive=True,
294
  elem_id="filter-columns-size",
295
  )
296
+ filter_columns_language = gr.CheckboxGroup(
297
+ label="Model Main Language",
298
+ choices=[i.value.name for i in Language],
299
+ value=[i.value.name for i in Language],
300
+ interactive=True,
301
+ elem_id="filter-columns-language",
302
+ )
303
 
304
  leaderboard_table = gr.components.Dataframe(
305
  value=leaderboard_df[
 
330
  filter_columns_type,
331
  filter_columns_precision,
332
  filter_columns_size,
333
+ filter_columns_language,
334
  hide_models,
335
  search_bar,
336
  ],
 
347
  filter_columns_type,
348
  filter_columns_precision,
349
  filter_columns_size,
350
+ filter_columns_language,
351
  hide_models,
352
  search_bar,
353
  ],
 
356
  # Check query parameter once at startup and update search bar + hidden component
357
  demo.load(load_query, inputs=[], outputs=[search_bar, hidden_search_bar])
358
 
359
+ for selector in [shown_columns, filter_columns_type, filter_columns_precision, filter_columns_size, filter_columns_language, hide_models]:
360
  selector.change(
361
  update_table,
362
  [
 
365
  filter_columns_type,
366
  filter_columns_precision,
367
  filter_columns_size,
368
+ filter_columns_language,
369
  hide_models,
370
  search_bar,
371
  ],
 
469
  value=ModelType.FT.to_str(" : "),
470
  interactive=True,
471
  )
472
+ main_language = gr.Dropdown(
473
+ choices=[i.value.name for i in Language if i != Language.Unknown],
474
+ label="Main Language",
475
+ multiselect=False,
476
+ value="English",
477
+ interactive=True,
478
+ )
479
 
480
  with gr.Column():
481
  precision = gr.Dropdown(
 
493
  interactive=True,
494
  )
495
  base_model_name_textbox = gr.Textbox(label="Base model (for delta or adapter weights)")
 
496
  submit_button = gr.Button("Submit Eval")
497
  submission_result = gr.Markdown()
498
  submit_button.click(
 
505
  private,
506
  weight_type,
507
  model_type,
508
+ main_language
509
  ],
510
  submission_result,
511
  )
initial_queue.jsonl CHANGED
@@ -1,215 +1,215 @@
1
  // 1- base models <=7B
2
- {"model": "TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
3
- {"model": "meta-llama/Llama-2-7b-hf", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
4
- {"model": "mistralai/Mistral-7B-v0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
5
- {"model": "huggyllama/llama-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
6
- {"model": "openlm-research/open_llama_3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
7
- {"model": "openlm-research/open_llama_3b_v2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
8
- {"model": "openlm-research/open_llama_7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
9
- {"model": "openlm-research/open_llama_7b_v2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
10
  // 2 - Larger base models <= 13B
11
- {"model": "meta-llama/Llama-2-13b-hf", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
12
- {"model": "huggyllama/llama-13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
13
- {"model": "openlm-research/open_llama_13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
14
- {"model": "upstage/SOLAR-10.7B-v1.0", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
15
  // 3 - portuguese models
16
- {"model": "maritaca-ai/sabia-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
17
- {"model": "dominguesm/canarim-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
18
- {"model": "22h/open-cabrita3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
19
- {"model": "recogna-nlp/bode-7b-alpaca-pt-br", "base_model": "meta-llama/Llama-2-7b-chat-hf", "revision": "main", "precision": "float16", "weight_type": "Adapter", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)"}
20
- {"model": "recogna-nlp/bode-13b-alpaca-pt-br", "base_model": "meta-llama/Llama-2-13b-chat-hf", "revision": "main", "precision": "float16", "weight_type": "Adapter", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)"}
21
- {"model": "22h/cabrita_7b_pt_850000", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
22
- {"model": "22h/cabrita-lora-v0-1", "base_model": "huggyllama/llama-7b", "revision": "main", "precision": "float16", "weight_type": "Adapter", "model_type": "🔶 : fine-tuned"}
23
- {"model": "wandgibaut/periquito-3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
24
- {"model": "nicolasdec/Cabra", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)"}
25
- {"model": "nicolasdec/cabra13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)"}
26
- {"model": "lrds-code/samba-1.1B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)"}
27
- {"model": "lrds-code/boana-7b-instruct", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)"}
28
- {"model": "nicholasKluge/Aira-2-portuguese-124M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)"}
29
- {"model": "nicholasKluge/Aira-2-portuguese-560M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)"}
30
- {"model": "nicholasKluge/Aira-2-portuguese-1B7", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)"}
31
  // other must-have <=7B
32
- {"model": "dynamofl/dynamo-8B-v0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
33
- {"model": "01-ai/Yi-6B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
34
- {"model": "Unbabel/TowerBase-7B-v0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
35
- {"model": "tiiuae/falcon-7b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
36
- {"model": "bigscience/bloom-560m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
37
- {"model": "bigscience/bloom-1b7", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
38
- {"model": "bigscience/bloom-3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
39
- {"model": "bigscience/bloom-7b1", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
40
- {"model": "stabilityai/stablelm-2-1_6b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
41
- {"model": "stabilityai/stablelm-3b-4e1t", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
42
  // Larger base models >13B
43
- {"model": "mistralai/Mixtral-8x7B-v0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
44
- {"model": "huggyllama/llama-30b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
45
- {"model": "01-ai/Yi-34B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
46
- {"model": "meta-llama/Llama-2-70b-hf", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
47
- {"model": "huggyllama/llama-65b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
48
  // minors must
49
- {"model": "togethercomputer/RedPajama-INCITE-Base-3B-v1", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
50
- {"model": "togethercomputer/RedPajama-INCITE-7B-Base", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
51
- {"model": "DAMO-NLP-MT/polylm-1.7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
52
- {"model": "DAMO-NLP-MT/polylm-13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
53
- {"model": "Deci/DeciLM-6b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🔶 : fine-tuned"}
54
- {"model": "Deci/DeciLM-7B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🔶 : fine-tuned"}
55
  // multiple (ch-jp)/en bi/multi lingual models
56
- {"model": "internlm/internlm2-7b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🔶 : fine-tuned"}
57
- {"model": "internlm/internlm2-base-7b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
58
- {"model": "internlm/internlm-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
59
- {"model": "internlm/internlm2-20b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🔶 : fine-tuned"}
60
- {"model": "internlm/internlm2-base-20b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
61
- {"model": "internlm/internlm-20b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
62
- {"model": "Qwen/Qwen-1_8B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
63
- {"model": "Qwen/Qwen-7B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
64
- {"model": "Qwen/Qwen-14B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
65
- {"model": "xverse/XVERSE-7B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
66
- {"model": "xverse/XVERSE-13B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
67
- {"model": "xverse/XVERSE-13B-256K", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
68
- {"model": "Skywork/Skywork-13B-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
69
- {"model": "baichuan-inc/Baichuan-7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
70
- {"model": "baichuan-inc/Baichuan-13B-Base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
71
- {"model": "baichuan-inc/Baichuan2-7B-Base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
72
- {"model": "baichuan-inc/Baichuan2-13B-Base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
73
- {"model": "OrionStarAI/Orion-14B-Base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
74
- {"model": "deepseek-ai/deepseek-llm-7b-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
75
- {"model": "deepseek-ai/deepseek-moe-16b-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
76
- {"model": "BAAI/Aquila-7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
77
- {"model": "BAAI/Aquila2-7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
78
- {"model": "THUDM/chatglm3-6b-base", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
79
- {"model": "THUDM/glm-2b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
80
- {"model": "THUDM/glm-10b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
81
- {"model": "fnlp/moss-moon-003-base", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
82
- {"model": "fnlp/moss-base-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
83
  // multiple chinese/jp large
84
- {"model": "Qwen/Qwen-72B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
85
- {"model": "xverse/XVERSE-65B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
86
- {"model": "xverse/XVERSE-65B-2", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
87
- {"model": "deepseek-ai/deepseek-llm-67b-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
88
- {"model": "BAAI/Aquila2-34B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
89
- {"model": "BAAI/Aquila2-70B-Expr", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
90
  // minors must 2
91
- {"model": "gpt2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
92
- {"model": "t5-small", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
93
- {"model": "t5-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
94
- {"model": "t5-large", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
95
- {"model": "google/mt5-small", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
96
- {"model": "google/mt5-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
97
- {"model": "google/mt5-large", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
98
  //others
99
- {"model": "NucleusAI/nucleus-22B-token-500B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
100
- {"model": "EleutherAI/pythia-14m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
101
- {"model": "EleutherAI/pythia-70m-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
102
- {"model": "EleutherAI/pythia-160m-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
103
- {"model": "EleutherAI/pythia-410m-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
104
- {"model": "EleutherAI/pythia-1b-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
105
- {"model": "EleutherAI/pythia-2.8b-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
106
- {"model": "EleutherAI/pythia-6.9b-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
107
- {"model": "EleutherAI/pythia-12b-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
108
- {"model": "EleutherAI/gpt-neo-125m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
109
- {"model": "EleutherAI/gpt-neo-1.3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
110
- {"model": "EleutherAI/gpt-neo-2.7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
111
- {"model": "EleutherAI/gpt-j-6b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
112
- {"model": "EleutherAI/gpt-neox-20b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
113
- {"model": "facebook/opt-125m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
114
- {"model": "facebook/opt-350m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
115
- {"model": "facebook/opt-1.3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
116
- {"model": "facebook/opt-2.7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
117
- {"model": "facebook/opt-6.7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
118
- {"model": "facebook/opt-13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
119
- {"model": "facebook/opt-30b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
120
  //other large
121
- {"model": "facebook/opt-66b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
122
- {"model": "tiiuae/falcon-40b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
123
  // minors portuguese
124
- {"model": "pierreguillou/gpt2-small-portuguese", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
125
- {"model": "pucpr/gpt2-bio-pt", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
126
- {"model": "unicamp-dl/ptt5-small-portuguese-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
127
- {"model": "unicamp-dl/ptt5-base-portuguese-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
128
- {"model": "unicamp-dl/ptt5-large-portuguese-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
129
- {"model": "unicamp-dl/ptt5-small-t5-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
130
- {"model": "unicamp-dl/ptt5-base-t5-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
131
- {"model": "unicamp-dl/ptt5-large-t5-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
132
- {"model": "josu/gpt-neo-pt-br", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
133
- {"model": "josu/gpt-neo-pt-1.3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
134
- {"model": "monilouise/opt125M_portuguese", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
135
- {"model": "HeyLucasLeao/gpt-neo-small-portuguese", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
136
  // other langs (es/Ko/Jp/nordic)
137
- {"model": "projecte-aina/FLOR-760M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
138
- {"model": "projecte-aina/FLOR-1.3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
139
- {"model": "projecte-aina/FLOR-6.3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
140
- {"model": "projecte-aina/aguila-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
141
- {"model": "EleutherAI/polyglot-ko-12.8b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
142
- {"model": "matsuo-lab/weblab-10b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
143
- {"model": "pfnet/plamo-13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
144
- {"model": "AI-Sweden-Models/gpt-sw3-6.7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
145
- {"model": "AI-Sweden-Models/gpt-sw3-6.7b-v2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
146
- {"model": "AI-Sweden-Models/gpt-sw3-20b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
147
- {"model": "AI-Sweden-Models/gpt-sw3-40b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
148
- {"model": "OpenLLM-France/Claire-Mistral-7B-0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
149
- {"model": "OpenLLM-France/Claire-7B-0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)"}
150
  // huge models:
151
  //{"model": "bigscience/bloom", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
152
  //{"model": "tiiuae/falcon-180B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
153
  //{"model": "facebook/galactica-120b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
154
  //random chat models
155
- {"model": "openchat/openchat-3.5-0106", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)"}
156
  //other 2
157
- {"model": "stabilityai/stablelm-base-alpha-3b-v2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
158
- {"model": "stabilityai/stablelm-base-alpha-7b-v2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
159
- {"model": "stabilityai/stablelm-base-alpha-3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
160
- {"model": "stabilityai/stablelm-base-alpha-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
161
- {"model": "openai-community/openai-gpt", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
162
- {"model": "openai-community/gpt2-medium", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
163
- {"model": "openai-community/gpt2-large", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
164
- {"model": "openai-community/gpt2-xl", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
165
- {"model": "microsoft/phi-1", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
166
- {"model": "microsoft/phi-1_5", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
167
- {"model": "microsoft/phi-2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
168
- {"model": "mosaicml/mpt-7b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
169
- {"model": "mosaicml/mpt-30b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
170
- {"model": "mosaicml/mpt-7b-8k", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
171
- {"model": "01-ai/Yi-6B-200K", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
172
- {"model": "01-ai/Yi-34B-200K", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
173
- {"model": "google/t5-v1_1-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
174
- {"model": "google/t5-v1_1-small", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
175
- {"model": "google/t5-v1_1-large", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
176
- {"model": "google/t5-v1_1-xl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
177
- {"model": "google/t5-v1_1-xxl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
178
- {"model": "google/mt5-xl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
179
- {"model": "google/mt5-xxl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
180
- {"model": "google/umt5-small", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
181
- {"model": "google/umt5-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
182
- {"model": "google/umt5-xl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
183
- {"model": "google/umt5-xxl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
184
- {"model": "AdaptLLM/law-LLM", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned"}
185
- {"model": "AdaptLLM/medicine-LLM", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned"}
186
- {"model": "AdaptLLM/finance-LLM", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned"}
187
- {"model": "AdaptLLM/law-LLM-13B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned"}
188
- {"model": "AdaptLLM/medicine-LLM-13B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned"}
189
- {"model": "AdaptLLM/finance-LLM-13B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned"}
190
- {"model": "cerebras/Cerebras-GPT-111M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
191
- {"model": "cerebras/Cerebras-GPT-256M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
192
- {"model": "cerebras/Cerebras-GPT-590M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
193
- {"model": "cerebras/Cerebras-GPT-1.3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
194
- {"model": "cerebras/Cerebras-GPT-2.7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
195
- {"model": "cerebras/Cerebras-GPT-6.7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
196
- {"model": "cerebras/Cerebras-GPT-13B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
197
- {"model": "cerebras/btlm-3b-8k-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
198
- {"model": "ai-forever/mGPT-13B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
199
- {"model": "ai-forever/mGPT", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
200
- {"model": "EleutherAI/pythia-70m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
201
- {"model": "EleutherAI/pythia-160m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
202
- {"model": "EleutherAI/pythia-410m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
203
- {"model": "EleutherAI/pythia-1b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
204
- {"model": "EleutherAI/pythia-2.8b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
205
- {"model": "EleutherAI/pythia-6.9b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
206
- {"model": "EleutherAI/pythia-12b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
207
- {"model": "facebook/galactica-125m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
208
- {"model": "facebook/galactica-1.3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
209
- {"model": "facebook/galactica-6.7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
210
- {"model": "facebook/galactica-30b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
211
- {"model": "facebook/xglm-564M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
212
- {"model": "facebook/xglm-1.7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
213
- {"model": "facebook/xglm-2.9B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
214
- {"model": "facebook/xglm-4.5B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
215
- {"model": "facebook/xglm-7.5B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
 
1
  // 1- base models <=7B
2
+ {"model": "TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
3
+ {"model": "meta-llama/Llama-2-7b-hf", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
4
+ {"model": "mistralai/Mistral-7B-v0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
5
+ {"model": "huggyllama/llama-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
6
+ {"model": "openlm-research/open_llama_3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
7
+ {"model": "openlm-research/open_llama_3b_v2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
8
+ {"model": "openlm-research/open_llama_7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
9
+ {"model": "openlm-research/open_llama_7b_v2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
10
  // 2 - Larger base models <= 13B
11
+ {"model": "meta-llama/Llama-2-13b-hf", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
12
+ {"model": "huggyllama/llama-13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
13
+ {"model": "openlm-research/open_llama_13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
14
+ {"model": "upstage/SOLAR-10.7B-v1.0", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
15
  // 3 - portuguese models
16
+ {"model": "maritaca-ai/sabia-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
17
+ {"model": "dominguesm/canarim-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
18
+ {"model": "22h/open-cabrita3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
19
+ {"model": "recogna-nlp/bode-7b-alpaca-pt-br", "base_model": "meta-llama/Llama-2-7b-chat-hf", "revision": "main", "precision": "float16", "weight_type": "Adapter", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)", "main_language": "Portuguese"}
20
+ {"model": "recogna-nlp/bode-13b-alpaca-pt-br", "base_model": "meta-llama/Llama-2-13b-chat-hf", "revision": "main", "precision": "float16", "weight_type": "Adapter", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)", "main_language": "Portuguese"}
21
+ {"model": "22h/cabrita_7b_pt_850000", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
22
+ {"model": "22h/cabrita-lora-v0-1", "base_model": "huggyllama/llama-7b", "revision": "main", "precision": "float16", "weight_type": "Adapter", "model_type": "🔶 : fine-tuned", "main_language": "Portuguese"}
23
+ {"model": "wandgibaut/periquito-3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
24
+ {"model": "nicolasdec/Cabra", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)", "main_language": "Portuguese"}
25
+ {"model": "nicolasdec/cabra13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)", "main_language": "Portuguese"}
26
+ {"model": "lrds-code/samba-1.1B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)", "main_language": "Portuguese"}
27
+ {"model": "lrds-code/boana-7b-instruct", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)", "main_language": "Portuguese"}
28
+ {"model": "nicholasKluge/Aira-2-portuguese-124M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)", "main_language": "Portuguese"}
29
+ {"model": "nicholasKluge/Aira-2-portuguese-560M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)", "main_language": "Portuguese"}
30
+ {"model": "nicholasKluge/Aira-2-portuguese-1B7", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)", "main_language": "Portuguese"}
31
  // other must-have <=7B
32
+ {"model": "dynamofl/dynamo-8B-v0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "English"}
33
+ {"model": "01-ai/Yi-6B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
34
+ {"model": "Unbabel/TowerBase-7B-v0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "English"}
35
+ {"model": "tiiuae/falcon-7b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
36
+ {"model": "bigscience/bloom-560m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
37
+ {"model": "bigscience/bloom-1b7", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
38
+ {"model": "bigscience/bloom-3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
39
+ {"model": "bigscience/bloom-7b1", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
40
+ {"model": "stabilityai/stablelm-2-1_6b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
41
+ {"model": "stabilityai/stablelm-3b-4e1t", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
42
  // Larger base models >13B
43
+ {"model": "mistralai/Mixtral-8x7B-v0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
44
+ {"model": "huggyllama/llama-30b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
45
+ {"model": "01-ai/Yi-34B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
46
+ {"model": "meta-llama/Llama-2-70b-hf", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
47
+ {"model": "huggyllama/llama-65b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
48
  // minors must
49
+ {"model": "togethercomputer/RedPajama-INCITE-Base-3B-v1", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
50
+ {"model": "togethercomputer/RedPajama-INCITE-7B-Base", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
51
+ {"model": "DAMO-NLP-MT/polylm-1.7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
52
+ {"model": "DAMO-NLP-MT/polylm-13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
53
+ {"model": "Deci/DeciLM-6b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🔶 : fine-tuned", "main_language": "English"}
54
+ {"model": "Deci/DeciLM-7B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🔶 : fine-tuned", "main_language": "English"}
55
  // multiple (ch-jp)/en bi/multi lingual models
56
+ {"model": "internlm/internlm2-7b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🔶 : fine-tuned", "main_language": "?"}
57
+ {"model": "internlm/internlm2-base-7b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
58
+ {"model": "internlm/internlm-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
59
+ {"model": "internlm/internlm2-20b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🔶 : fine-tuned", "main_language": "?"}
60
+ {"model": "internlm/internlm2-base-20b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
61
+ {"model": "internlm/internlm-20b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
62
+ {"model": "Qwen/Qwen-1_8B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Chinese"}
63
+ {"model": "Qwen/Qwen-7B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Chinese"}
64
+ {"model": "Qwen/Qwen-14B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Chinese"}
65
+ {"model": "xverse/XVERSE-7B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
66
+ {"model": "xverse/XVERSE-13B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
67
+ {"model": "xverse/XVERSE-13B-256K", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
68
+ {"model": "Skywork/Skywork-13B-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
69
+ {"model": "baichuan-inc/Baichuan-7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Chinese"}
70
+ {"model": "baichuan-inc/Baichuan-13B-Base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
71
+ {"model": "baichuan-inc/Baichuan2-7B-Base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Chinese"}
72
+ {"model": "baichuan-inc/Baichuan2-13B-Base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Chinese"}
73
+ {"model": "OrionStarAI/Orion-14B-Base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Chinese"}
74
+ {"model": "deepseek-ai/deepseek-llm-7b-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
75
+ {"model": "deepseek-ai/deepseek-moe-16b-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
76
+ {"model": "BAAI/Aquila-7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
77
+ {"model": "BAAI/Aquila2-7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
78
+ {"model": "THUDM/chatglm3-6b-base", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Chinese"}
79
+ {"model": "THUDM/glm-2b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
80
+ {"model": "THUDM/glm-10b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
81
+ {"model": "fnlp/moss-moon-003-base", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
82
+ {"model": "fnlp/moss-base-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
83
  // multiple chinese/jp large
84
+ {"model": "Qwen/Qwen-72B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Chinese"}
85
+ {"model": "xverse/XVERSE-65B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
86
+ {"model": "xverse/XVERSE-65B-2", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
87
+ {"model": "deepseek-ai/deepseek-llm-67b-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
88
+ {"model": "BAAI/Aquila2-34B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
89
+ {"model": "BAAI/Aquila2-70B-Expr", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "?"}
90
  // minors must 2
91
+ {"model": "gpt2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
92
+ {"model": "t5-small", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
93
+ {"model": "t5-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
94
+ {"model": "t5-large", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
95
+ {"model": "google/mt5-small", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
96
+ {"model": "google/mt5-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
97
+ {"model": "google/mt5-large", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
98
  //others
99
+ {"model": "NucleusAI/nucleus-22B-token-500B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
100
+ {"model": "EleutherAI/pythia-14m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
101
+ {"model": "EleutherAI/pythia-70m-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
102
+ {"model": "EleutherAI/pythia-160m-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
103
+ {"model": "EleutherAI/pythia-410m-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
104
+ {"model": "EleutherAI/pythia-1b-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
105
+ {"model": "EleutherAI/pythia-2.8b-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
106
+ {"model": "EleutherAI/pythia-6.9b-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
107
+ {"model": "EleutherAI/pythia-12b-deduped", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
108
+ {"model": "EleutherAI/gpt-neo-125m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
109
+ {"model": "EleutherAI/gpt-neo-1.3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
110
+ {"model": "EleutherAI/gpt-neo-2.7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
111
+ {"model": "EleutherAI/gpt-j-6b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
112
+ {"model": "EleutherAI/gpt-neox-20b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
113
+ {"model": "facebook/opt-125m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
114
+ {"model": "facebook/opt-350m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
115
+ {"model": "facebook/opt-1.3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
116
+ {"model": "facebook/opt-2.7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
117
+ {"model": "facebook/opt-6.7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
118
+ {"model": "facebook/opt-13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
119
+ {"model": "facebook/opt-30b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
120
  //other large
121
+ {"model": "facebook/opt-66b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
122
+ {"model": "tiiuae/falcon-40b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
123
  // minors portuguese
124
+ {"model": "pierreguillou/gpt2-small-portuguese", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
125
+ {"model": "pucpr/gpt2-bio-pt", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
126
+ {"model": "unicamp-dl/ptt5-small-portuguese-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
127
+ {"model": "unicamp-dl/ptt5-base-portuguese-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
128
+ {"model": "unicamp-dl/ptt5-large-portuguese-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
129
+ {"model": "unicamp-dl/ptt5-small-t5-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
130
+ {"model": "unicamp-dl/ptt5-base-t5-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
131
+ {"model": "unicamp-dl/ptt5-large-t5-vocab", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
132
+ {"model": "josu/gpt-neo-pt-br", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
133
+ {"model": "josu/gpt-neo-pt-1.3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
134
+ {"model": "monilouise/opt125M_portuguese", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
135
+ {"model": "HeyLucasLeao/gpt-neo-small-portuguese", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Portuguese"}
136
  // other langs (es/Ko/Jp/nordic)
137
+ {"model": "projecte-aina/FLOR-760M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Spanish"}
138
+ {"model": "projecte-aina/FLOR-1.3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Spanish"}
139
+ {"model": "projecte-aina/FLOR-6.3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Spanish"}
140
+ {"model": "projecte-aina/aguila-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Spanish"}
141
+ {"model": "EleutherAI/polyglot-ko-12.8b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Other"}
142
+ {"model": "matsuo-lab/weblab-10b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "Other"}
143
+ {"model": "pfnet/plamo-13b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
144
+ {"model": "AI-Sweden-Models/gpt-sw3-6.7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
145
+ {"model": "AI-Sweden-Models/gpt-sw3-6.7b-v2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
146
+ {"model": "AI-Sweden-Models/gpt-sw3-20b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
147
+ {"model": "AI-Sweden-Models/gpt-sw3-40b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
148
+ {"model": "OpenLLM-France/Claire-Mistral-7B-0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Other"}
149
+ {"model": "OpenLLM-France/Claire-7B-0.1", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🆎 : language adapted models (FP, FT, ...)", "main_language": "Other"}
150
  // huge models:
151
  //{"model": "bigscience/bloom", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
152
  //{"model": "tiiuae/falcon-180B", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
153
  //{"model": "facebook/galactica-120b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained"}
154
  //random chat models
155
+ {"model": "openchat/openchat-3.5-0106", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "💬 : chat models (RLHF, DPO, IFT, ...)", "main_language": "English"}
156
  //other 2
157
+ {"model": "stabilityai/stablelm-base-alpha-3b-v2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
158
+ {"model": "stabilityai/stablelm-base-alpha-7b-v2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
159
+ {"model": "stabilityai/stablelm-base-alpha-3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
160
+ {"model": "stabilityai/stablelm-base-alpha-7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
161
+ {"model": "openai-community/openai-gpt", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
162
+ {"model": "openai-community/gpt2-medium", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
163
+ {"model": "openai-community/gpt2-large", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
164
+ {"model": "openai-community/gpt2-xl", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
165
+ {"model": "microsoft/phi-1", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
166
+ {"model": "microsoft/phi-1_5", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
167
+ {"model": "microsoft/phi-2", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
168
+ {"model": "mosaicml/mpt-7b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
169
+ {"model": "mosaicml/mpt-30b", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
170
+ {"model": "mosaicml/mpt-7b-8k", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
171
+ {"model": "01-ai/Yi-6B-200K", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
172
+ {"model": "01-ai/Yi-34B-200K", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
173
+ {"model": "google/t5-v1_1-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
174
+ {"model": "google/t5-v1_1-small", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
175
+ {"model": "google/t5-v1_1-large", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
176
+ {"model": "google/t5-v1_1-xl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
177
+ {"model": "google/t5-v1_1-xxl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
178
+ {"model": "google/mt5-xl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
179
+ {"model": "google/mt5-xxl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
180
+ {"model": "google/umt5-small", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
181
+ {"model": "google/umt5-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
182
+ {"model": "google/umt5-xl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
183
+ {"model": "google/umt5-xxl", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
184
+ {"model": "AdaptLLM/law-LLM", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned", "main_language": "English"}
185
+ {"model": "AdaptLLM/medicine-LLM", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned", "main_language": "English"}
186
+ {"model": "AdaptLLM/finance-LLM", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned", "main_language": "English"}
187
+ {"model": "AdaptLLM/law-LLM-13B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned", "main_language": "English"}
188
+ {"model": "AdaptLLM/medicine-LLM-13B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned", "main_language": "English"}
189
+ {"model": "AdaptLLM/finance-LLM-13B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🔶 : fine-tuned", "main_language": "English"}
190
+ {"model": "cerebras/Cerebras-GPT-111M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
191
+ {"model": "cerebras/Cerebras-GPT-256M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
192
+ {"model": "cerebras/Cerebras-GPT-590M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
193
+ {"model": "cerebras/Cerebras-GPT-1.3B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
194
+ {"model": "cerebras/Cerebras-GPT-2.7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
195
+ {"model": "cerebras/Cerebras-GPT-6.7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
196
+ {"model": "cerebras/Cerebras-GPT-13B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
197
+ {"model": "cerebras/btlm-3b-8k-base", "base_model": "", "revision": "main", "precision": "bfloat16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
198
+ {"model": "ai-forever/mGPT-13B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
199
+ {"model": "ai-forever/mGPT", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
200
+ {"model": "EleutherAI/pythia-70m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
201
+ {"model": "EleutherAI/pythia-160m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
202
+ {"model": "EleutherAI/pythia-410m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
203
+ {"model": "EleutherAI/pythia-1b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
204
+ {"model": "EleutherAI/pythia-2.8b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
205
+ {"model": "EleutherAI/pythia-6.9b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
206
+ {"model": "EleutherAI/pythia-12b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
207
+ {"model": "facebook/galactica-125m", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
208
+ {"model": "facebook/galactica-1.3b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
209
+ {"model": "facebook/galactica-6.7b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
210
+ {"model": "facebook/galactica-30b", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
211
+ {"model": "facebook/xglm-564M", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
212
+ {"model": "facebook/xglm-1.7B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
213
+ {"model": "facebook/xglm-2.9B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
214
+ {"model": "facebook/xglm-4.5B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
215
+ {"model": "facebook/xglm-7.5B", "base_model": "", "revision": "main", "precision": "float16", "weight_type": "Original", "model_type": "🟢 : pretrained", "main_language": "English"}
src/display/utils.py CHANGED
@@ -66,6 +66,7 @@ auto_eval_column_dict.append(["dummy", ColumnContent, ColumnContent("Model Name"
66
  if GET_ORIGINAL_HF_LEADERBOARD_EVAL_RESULTS:
67
  auto_eval_column_dict.append(["original_benchmark_average", ColumnContent, ColumnContent("🤗 Leaderboard Average", "number", False)])
68
  auto_eval_column_dict.append(["npm", ColumnContent, ColumnContent("NPM (Average) ⬆️", "number", False)])
 
69
 
70
  # We use make dataclass to dynamically fill the scores from Tasks
71
  AutoEvalColumn = make_dataclass("AutoEvalColumn", auto_eval_column_dict, frozen=True)
@@ -103,7 +104,8 @@ baseline_row = {
103
  AutoEvalColumn.license.name: "",
104
  AutoEvalColumn.still_on_hub.name: False,
105
  AutoEvalColumn.moe.name: False,
106
- AutoEvalColumn.eval_time.name: 0.0
 
107
  }
108
 
109
  baseline_list = []
@@ -152,6 +154,7 @@ human_baseline_row = {
152
  AutoEvalColumn.still_on_hub.name: False,
153
  AutoEvalColumn.moe.name: False,
154
  AutoEvalColumn.eval_time.name: 0.0,
 
155
  }
156
 
157
  baseline_list = []
@@ -225,7 +228,27 @@ class Precision(Enum):
225
  return Precision.qt_GPTQ
226
  return Precision.Unknown
227
 
 
 
 
 
 
 
 
228
 
 
 
 
 
 
 
 
 
 
 
 
 
 
229
 
230
 
231
  # Column selection
 
66
  if GET_ORIGINAL_HF_LEADERBOARD_EVAL_RESULTS:
67
  auto_eval_column_dict.append(["original_benchmark_average", ColumnContent, ColumnContent("🤗 Leaderboard Average", "number", False)])
68
  auto_eval_column_dict.append(["npm", ColumnContent, ColumnContent("NPM (Average) ⬆️", "number", False)])
69
+ auto_eval_column_dict.append(["main_language", ColumnContent, ColumnContent("Main Language", "str", False)])
70
 
71
  # We use make dataclass to dynamically fill the scores from Tasks
72
  AutoEvalColumn = make_dataclass("AutoEvalColumn", auto_eval_column_dict, frozen=True)
 
104
  AutoEvalColumn.license.name: "",
105
  AutoEvalColumn.still_on_hub.name: False,
106
  AutoEvalColumn.moe.name: False,
107
+ AutoEvalColumn.eval_time.name: 0.0,
108
+ AutoEvalColumn.main_language.name: "?"
109
  }
110
 
111
  baseline_list = []
 
154
  AutoEvalColumn.still_on_hub.name: False,
155
  AutoEvalColumn.moe.name: False,
156
  AutoEvalColumn.eval_time.name: 0.0,
157
+ AutoEvalColumn.main_language.name: "?",
158
  }
159
 
160
  baseline_list = []
 
228
  return Precision.qt_GPTQ
229
  return Precision.Unknown
230
 
231
+ class Language(Enum):
232
+ English = ModelDetails("English")
233
+ Portuguese = ModelDetails("Portuguese")
234
+ Spanish = ModelDetails("Spanish")
235
+ Chinese = ModelDetails("Chinese")
236
+ Other = ModelDetails("Other")
237
+ Unknown = ModelDetails("?")
238
 
239
+ def from_str(language):
240
+ language = language.lower().replace('-', '').replace('_', '')
241
+ if language in ["pt", "ptpt", "ptbr", "portuguese"]:
242
+ return Language.Portuguese
243
+ if language in ["en", "enus", "engb", "english", ]:
244
+ return Language.English
245
+ if language in ["es", "spanish"]:
246
+ return Language.Spanish
247
+ if language in ["zh", "chinese"]:
248
+ return Language.Chinese
249
+ if language in ["other", "multi", "multilingual"]:
250
+ return Language.Other
251
+ return Language.Unknown
252
 
253
 
254
  # Column selection
src/leaderboard/read_evals.py CHANGED
@@ -4,6 +4,7 @@ import math
4
  import os
5
  from dataclasses import dataclass
6
  from typing import List
 
7
 
8
  import dateutil
9
  import numpy as np
@@ -11,7 +12,7 @@ import numpy as np
11
  from huggingface_hub import ModelCard
12
 
13
  from src.display.formatting import make_clickable_model
14
- from src.display.utils import AutoEvalColumn, ModelType, Tasks, Precision, WeightType, ORIGINAL_TASKS
15
  from src.envs import GET_ORIGINAL_HF_LEADERBOARD_EVAL_RESULTS, SHOW_INCOMPLETE_EVALS
16
 
17
  @dataclass
@@ -26,6 +27,7 @@ class EvalResult:
26
  precision: Precision = Precision.Unknown
27
  model_type: ModelType = ModelType.Unknown # Pretrained, fine tuned, ...
28
  weight_type: WeightType = WeightType.Original # Original or Adapter
 
29
  architecture: str = "Unknown" # From config file
30
  license: str = "?"
31
  likes: int = 0
@@ -137,6 +139,7 @@ class EvalResult:
137
  self.architecture = request.get("architectures", "Unknown")
138
  self.status = request.get("status", "FAILED")
139
  self.hidden = request.get("hidden", False)
 
140
  except Exception as e:
141
  self.status = "FAILED"
142
  print(f"Could not find request file for {self.org}/{self.model}")
@@ -188,7 +191,8 @@ class EvalResult:
188
  AutoEvalColumn.moe.name: ("moe" in self.tags if self.tags else False) or "moe" in self.full_model.lower(),
189
  AutoEvalColumn.flagged.name: self.flagged,
190
  AutoEvalColumn.eval_time.name: self.eval_time,
191
- AutoEvalColumn.npm.name: npm
 
192
  }
193
 
194
  for task in Tasks:
 
4
  import os
5
  from dataclasses import dataclass
6
  from typing import List
7
+ import traceback
8
 
9
  import dateutil
10
  import numpy as np
 
12
  from huggingface_hub import ModelCard
13
 
14
  from src.display.formatting import make_clickable_model
15
+ from src.display.utils import AutoEvalColumn, ModelType, Tasks, Precision, Language, WeightType, ORIGINAL_TASKS
16
  from src.envs import GET_ORIGINAL_HF_LEADERBOARD_EVAL_RESULTS, SHOW_INCOMPLETE_EVALS
17
 
18
  @dataclass
 
27
  precision: Precision = Precision.Unknown
28
  model_type: ModelType = ModelType.Unknown # Pretrained, fine tuned, ...
29
  weight_type: WeightType = WeightType.Original # Original or Adapter
30
+ main_language: Language = Language.Unknown
31
  architecture: str = "Unknown" # From config file
32
  license: str = "?"
33
  likes: int = 0
 
139
  self.architecture = request.get("architectures", "Unknown")
140
  self.status = request.get("status", "FAILED")
141
  self.hidden = request.get("hidden", False)
142
+ self.main_language = request.get("main_language", "?")
143
  except Exception as e:
144
  self.status = "FAILED"
145
  print(f"Could not find request file for {self.org}/{self.model}")
 
191
  AutoEvalColumn.moe.name: ("moe" in self.tags if self.tags else False) or "moe" in self.full_model.lower(),
192
  AutoEvalColumn.flagged.name: self.flagged,
193
  AutoEvalColumn.eval_time.name: self.eval_time,
194
+ AutoEvalColumn.npm.name: npm,
195
+ AutoEvalColumn.main_language.name: self.main_language
196
  }
197
 
198
  for task in Tasks:
src/submission/submit.py CHANGED
@@ -27,7 +27,8 @@ def add_new_eval(
27
  private: bool,
28
  weight_type: str,
29
  model_type: str,
30
- source="leaderboard"
 
31
  ):
32
  global REQUESTED_MODELS
33
  global USERS_TO_SUBMISSION_DATES
@@ -119,6 +120,7 @@ def add_new_eval(
119
  "params": model_size,
120
  "architectures": architecture,
121
  "weight_type": weight_type,
 
122
  "status": "PENDING",
123
  "submitted_time": current_time,
124
  "model_type": model_type,
 
27
  private: bool,
28
  weight_type: str,
29
  model_type: str,
30
+ main_language: str,
31
+ source="leaderboard",
32
  ):
33
  global REQUESTED_MODELS
34
  global USERS_TO_SUBMISSION_DATES
 
120
  "params": model_size,
121
  "architectures": architecture,
122
  "weight_type": weight_type,
123
+ "main_language": main_language,
124
  "status": "PENDING",
125
  "submitted_time": current_time,
126
  "model_type": model_type,