github.com/lm-sys/FastChat issues | Ecosyste.ms: OpenCollective

Make Table non editable: Set interactive=False in Gradio

zakuru opened this pull request 11 months ago

fix: inconsistent tokenization by llama tokenizer

congchan opened this pull request 11 months ago

Feat: Make the table cells read only in the Leader Board

zakuru opened this issue 11 months ago

Add Whole History Rating to Leaderboard?

endolith opened this issue 11 months ago

vllm_worker will be hung after chating a while

hanswang1 opened this issue 11 months ago

Question on the Security Improvements in Vicuna-7B-v1.5

jc-ryan opened this issue 11 months ago

sglang worker raise exception in benchmark

zeyugao opened this issue 11 months ago

failed to run FastChat with stable lm-zephyr-3b

jasonzou opened this issue 11 months ago

How to let vllm_worker to support load-8bits?

hanswang1 opened this issue 11 months ago

Create a fine_tune_requirements.txt file

Oscarjia opened this pull request 11 months ago

code update

infwinston opened this pull request 11 months ago

fix: tokenization mismatch when training with different templates

congchan opened this pull request 11 months ago

Setup fine-tuning environment

Oscarjia opened this issue 11 months ago

fastchat 0.2.34 incompatible with vllm 0.3.0

wujingbo-web opened this issue 11 months ago

Incorrect Output of Umlauts in deepseek-llm-67b-chat Model

maninthemiddle01 opened this issue 11 months ago

Bugs when fine-tune tiny-llama with instructions using tiny-llama's conversation template

hychaochao opened this issue 11 months ago

convert sep_style to int raise ValueError in some case during base_model_worker init

vincentor opened this issue 11 months ago

BLEU Score

TheDeadcoder opened this issue 11 months ago

update yuan2.0 generation

wangpengfei1013 opened this pull request 11 months ago

Wrong reference answer in MT-Bench

jubgjf opened this issue 11 months ago

运行baichuan2报错

yuege613 opened this issue 11 months ago

[Dependency] webui dependency does not work along

Michaelvll opened this issue 11 months ago

Error: Llama-2-7B-Chat-AWQ safetensors_rust.SafetensorError: Error while deserializing header: HeaderTooLarge

zhangyilun opened this issue 11 months ago

Fix tokenizer for vllm worker

Michaelvll opened this pull request 11 months ago

FastChat API vs. transformers/generate speed difference

mpetruc opened this issue 11 months ago

ChatGLM3-6b 回复中内容问题

Bannerli opened this issue 11 months ago

关于openAI 服务太慢的问题 Regarding the problem that openAI service is too slow

shell-nlp opened this issue 11 months ago

How to register a new model API

yigediaosi opened this issue 11 months ago

deluxe chat

Bensolz opened this issue 11 months ago

vllm_worker - Mixtral - Tokenizer points to TokenizerGroup object

sudarshan-kamath opened this issue 11 months ago

The issue with model saving in train.py

Kong-Aobo opened this issue 11 months ago

fetch_remote needs to catch exception

zhouzaida opened this pull request 11 months ago

Update README.md

merrymercy opened this pull request 11 months ago

use cl100k_base as the default tiktoken encoding

bjwswang opened this pull request 11 months ago

Adding download folder information in README

dheeraj-326 opened this pull request 11 months ago

Anybody know what is the version of `flash_attn` used for finetune?

Oscarjia opened this issue 11 months ago

update yuan2.0 inference generation & delete some mistakes about yuan 2.0

wangpengfei1013 opened this pull request 11 months ago

Why device_map isn't passed on as kwargs when loading models?

sthemeow opened this issue 12 months ago

Gradio Web Server for Multimodal Models

BabyChouSr opened this pull request 12 months ago

Possibility of hiding aliases from the WEB UI but allowing it on the API endpoints

surak opened this issue 12 months ago

feat: train with template

congchan opened this pull request 12 months ago

chatglm3-6b-32k模型对话出现大量重复性内容

leojrAI opened this issue 12 months ago

OpenAI server does not work with stream: true

ArchiDevil opened this issue 12 months ago

书⽣·浦语2.0（InternLM2） is not supported?

oom2018 opened this issue 12 months ago

SGLang Worker

BabyChouSr opened this pull request 12 months ago

Inconsistency of vicuna v1.1, v1.3, v1.5 prompt

kazunaritakeichi opened this issue 12 months ago

add logprobs to vllm integration

aniemerg opened this pull request 12 months ago

add openai version >= 1.0 support to llm_judge

thavens opened this pull request 12 months ago

How to customize conv_template in fastchat?

empty2enrich opened this issue 12 months ago

Repetition Penalty

joeriakkerman opened this issue 12 months ago

qwen-72b-chat runs in Fastchat vllm , the input has weird message that is not mine.

dongxu opened this issue 12 months ago

--conv-template parameter has no effect。

empty2enrich opened this issue 12 months ago

GPT-4-Vision and Gemini Vision multimodal model support?

youself64github opened this issue about 1 year ago

Inconsistencies between OpenAI API and FastChat's OpenAI serving model

brandonbiggs opened this issue about 1 year ago

API For JS

duplicate2991 opened this issue about 1 year ago

Update token spacing for mistral conversation.py

thavens opened this pull request about 1 year ago

Bugs when fine-tune llama-2-7b with instructions using llama-2's conversation template

YJiangcm opened this issue about 1 year ago

check if hm in models before deleting to avoid errors

joshua-ne opened this pull request about 1 year ago

Update web server commands

merrymercy opened this pull request about 1 year ago

vllm worker becomes unresponsive after running for a long time.

Future-FC opened this issue about 1 year ago

leaderboard code update

infwinston opened this pull request about 1 year ago

support bge-reranker-large model

kztao opened this issue about 1 year ago

Fix the problem of not using the decoding method corresponding to the base model in peft mode

Jingsong-Yan opened this pull request about 1 year ago

update a new sota model on MT-Bench which touch an 8.8 scores.

xiechengmude opened this pull request about 1 year ago

Update a new dpo-v2 trained model by xDAN-AI achieving new sota

xiechengmude opened this pull request about 1 year ago

Update a new dpo-v2 trained model by xDAN-AI achieving new sota scores (8.8) on MT-Bench

xiechengmude opened this pull request about 1 year ago

mistral 7b train

bigdante opened this issue about 1 year ago

Support "Openchat 3.5 - 1210"

Graveyard9988 opened this issue about 1 year ago

Modify doc description

zhangsibo1129 opened this pull request about 1 year ago

Update docs

Ying1123 opened this pull request about 1 year ago

Clean error messages

merrymercy opened this pull request about 1 year ago

Add new models (Perplexity, gemini) & Separate GPT versions

merrymercy opened this pull request about 1 year ago

Any plan for supporting NexusRaven-V2-13B?

engchina opened this issue about 1 year ago

Format code

merrymercy opened this pull request about 1 year ago

Add StripedHyena-Nous-7B

Zymrael opened this issue about 1 year ago

Update README.md

eltociear opened this pull request about 1 year ago

New

olachinkei opened this pull request about 1 year ago

Mixtral-8x7B-Instruct-v0.1: Chat arena vs local inference

jalajthanaki opened this issue about 1 year ago

Feature request: Along with accuracy benchmark, is it possible to get the throughput information for all the models on leaderboard?

jalajthanaki opened this issue about 1 year ago

TypeError: unsupported operand type(s) for |: 'dict' and 'dict'

PhaneendraGunda opened this issue about 1 year ago

fix: 'compeletion' typo

congchan opened this pull request about 1 year ago

Add Tunnelmole as an open source alternative to ngrok and include usage instructions

robbie-cahill opened this pull request about 1 year ago

vllm + chatglm3, it can't generate content as expected

yaofeng opened this issue about 1 year ago

any plan for 33b based on llama2?

warvyvr opened this issue about 1 year ago

NPU needs to be initialized when starting a new process

jq460494839 opened this pull request about 1 year ago

deploy with npu raise The context is empty

jq460494839 opened this issue about 1 year ago

安装fastchat报错

Dongmei97 opened this issue about 1 year ago

Add support for CatPPT

rishiraj opened this pull request about 1 year ago

warn when input text cut

GianniGi opened this issue about 1 year ago

Mixtral-8x7B-Instruct-v0.1 will be officially integrated inside the fastchat codebase

jalajthanaki opened this issue about 1 year ago

fun:/v1/chat/completions , use "stream":"False". error ? why

sjyttkl opened this issue about 1 year ago

Support "Phi-2"

Graveyard9988 opened this issue about 1 year ago

prompt template for SOLAR-10.7B-Instruct-v1.0

HelloCard opened this issue about 1 year ago

mistral-8x-7b stops generating in mid

tacacs1101-debug opened this issue about 1 year ago

使用vllm_worker进行模型加载，卡着不动

wfs420100 opened this issue about 1 year ago

Add functionality to ping AI2 InferD endpoints for tulu 2

natolambert opened this pull request about 1 year ago

fastchat.serve.vllm_worker方法部署Yi-34B-Chat-4bits，实体抽取类任务请求服务失败，内部错误

houhailun opened this issue about 1 year ago

add download models from www.modelscope.cn

liuyhwangyh opened this pull request about 1 year ago

Inference using NF4 weight

DevNullx64 opened this pull request about 1 year ago

Add Huggingface's apply chat template functionality

BabyChouSr opened this pull request about 1 year ago