1bitLLM
/

bitnet_b1_58-3B

Text Generation

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

Resources

View closed (1)

can you provide wikitest ppl and c4 ppl separately?

#11 opened 7 months ago by

Can you provide more details on the training?

#10 opened 7 months ago by

Any plans to use MQA (multi-query attention) or GQA (grouped-query attention) in the future?

#9 opened 7 months ago by

Efficient Inference Kernel Support for 1.58bit.

#8 opened 7 months ago by

This code from BitLinear doesn't make sense

#7 opened 8 months ago by

Is it bitnet {-1,0,1}?

#6 opened 8 months ago by

ValueError: Tokenizer class BitnetTokenizer does not exist or is not currently imported.

#5 opened 8 months ago by

Longer inference time

#4 opened 8 months ago by

Why are these models fp32?

#2 opened 8 months ago by

Is there a chat/instruct model in plans?

#1 opened 8 months ago by