FastChat
FastChat copied to clipboard
lm-sys
An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.
# Due to the removal of the "fp16" attribute from Qwen2's config, the following conditional statement needs to be added if a Qwen2 model is to be used: if( "qwen2"...
When I use training code to finetune mistral-7b,I have met the exception like "Could not find the transformer layer class to wrap in the model.". How can I fix it?
when call "/v1/chat/completions", there will call the function [check_length](https://github.com/lm-sys/FastChat/blob/main/fastchat/serve/openai_api_server.py#L437) to compute `max_new_tokens` use [min(max_tokens, context_len - token_num)](https://github.com/lm-sys/FastChat/blob/main/fastchat/serve/openai_api_server.py#L169) where [token_num](https://github.com/lm-sys/FastChat/blob/main/fastchat/serve/base_model_worker.py#L152) is len(tokinzer(pormot).input_ids)， but when compute `max_src_len` use `max_src_len = context_len -...
## Why are these changes needed? support korean language ## Related issue number (if applicable) ## Checks - [ ] I've run `format.sh` to lint the changes in this PR....
Hi, I plan to make a lighter Docker image when only API-based inference is needed. But torch is imported when getting conversation templates. Do you have any opinion on how...
I am able to run the Vicuna 7B model just fine normally, by providing it with either "messages" (Q&A format) or the "prompt" (Sentence completion capability). However when I follow...
It seems current model_chatglm.py does not work with GLM-4 models. Anything to do like chatglm3 to fix it? Btw, is it possible to use Fastchat for multi-modes scenarios (.pdf, .png)?
I have noticed that the Korean language is missing in the monitor of FastChat. Here's the link: https://github.com/lm-sys/FastChat/blob/c5223e34babd24c3f9b08205e6751ea6e42c9684/fastchat/serve/monitor/monitor.py#L391 I believe there are many Korean users, so are there any plans...
https://github.com/bin123apple/AutoCoder - This model beats all AIs including GPT 4 April 2024 and even GPT 4o, and is a free and open source AI model
Hello, Character.AI does not offer an API, so can you please collaborate with Character.AI to provide unlimited access to Character 1.0, Character 1.1, and Character 1.2 for the purpose of...