change to qwen2 7b
Browse files
app.py
CHANGED
@@ -1,16 +1,16 @@
|
|
1 |
import gradio as gr
|
2 |
from llama_cpp import Llama
|
3 |
|
4 |
-
model = "
|
5 |
llm = Llama.from_pretrained(
|
6 |
repo_id=model,
|
7 |
-
filename="
|
8 |
-
verbose=
|
9 |
use_mmap=False,
|
10 |
use_mlock=True,
|
11 |
n_threads=2,
|
12 |
n_threads_batch=2,
|
13 |
-
n_ctx=
|
14 |
)
|
15 |
|
16 |
|
|
|
1 |
import gradio as gr
|
2 |
from llama_cpp import Llama
|
3 |
|
4 |
+
model = "Qwen/Qwen2-7B-Instruct-GGUF"
|
5 |
llm = Llama.from_pretrained(
|
6 |
repo_id=model,
|
7 |
+
filename="qwen2-7b-instruct-q4_k_m.gguf",
|
8 |
+
verbose=False,
|
9 |
use_mmap=False,
|
10 |
use_mlock=True,
|
11 |
n_threads=2,
|
12 |
n_threads_batch=2,
|
13 |
+
n_ctx=40000,
|
14 |
)
|
15 |
|
16 |
|