Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / ModelTC/lightllm issues and pull requests

#31 - generate text garbled

Issue - State: closed - Opened by ChristineSeven over 1 year ago - 14 comments
Labels: help wanted

#30 - Refactoring code for readability

Pull Request - State: closed - Opened by hiworldwzj over 1 year ago

#29 - Add error reports from sub process

Pull Request - State: closed - Opened by llehtahw over 1 year ago

#28 - OOM when prompt length exceeds 1020.

Issue - State: closed - Opened by tonylin52 over 1 year ago - 5 comments

#27 - add nccl_port to startup args

Pull Request - State: closed - Opened by hiworldwzj over 1 year ago

#26 - LLama 模型支持问题

Issue - State: closed - Opened by freeliuzc over 1 year ago - 6 comments

#25 - Can nccl init port address be exposed to command line interface?

Issue - State: closed - Opened by haoranchen06 over 1 year ago - 2 comments

#24 - Implementation of Positional Interpolation (PI) Feature

Pull Request - State: closed - Opened by andy-yang-1 over 1 year ago

#23 - Encounter error when serving with vicuna-13b-v1.3.

Issue - State: closed - Opened by shanshanpt over 1 year ago - 9 comments

#21 - llama2-70B 加载完成后，无法启动服务

Issue - State: closed - Opened by zackdist over 1 year ago - 7 comments

#20 - benchmark stuck

Issue - State: closed - Opened by leiwen83 over 1 year ago - 18 comments

#19 - triton kernel compile error

Issue - State: closed - Opened by leiwen83 over 1 year ago - 7 comments

#18 - Install lightllm in dockerfile

Pull Request - State: closed - Opened by hamelsmu over 1 year ago - 1 comment

#17 - Stream output

Issue - State: closed - Opened by tonylin52 over 1 year ago - 6 comments

#16 - 调用报错

Issue - State: open - Opened by xxm1668 over 1 year ago - 14 comments

#15 - I tried test_llama.py，but.... help...T^T

Issue - State: open - Opened by MissQueen over 1 year ago - 4 comments

#14 - Usage instructions for the server parameter

Pull Request - State: closed - Opened by hiworldwzj over 1 year ago

#13 - my A800 80G*8

Issue - State: open - Opened by weisihao over 1 year ago - 7 comments

#12 - Docker iamges

Issue - State: closed - Opened by Vincent131499 over 1 year ago - 1 comment
Labels: enhancement

#11 - Optimize test code to reduce duplicate code.

Pull Request - State: closed - Opened by hiworldwzj over 1 year ago

#10 - provide some examples

Issue - State: closed - Opened by lucasjinreal over 1 year ago - 1 comment

#9 - [feature request] add prompt styles support

Issue - State: open - Opened by jiacheo over 1 year ago - 4 comments
Labels: enhancement

#8 - Comparison with deepspeed inference?

Issue - State: closed - Opened by allanj over 1 year ago - 1 comment

#7 - the stream output is same to OpenAI?

Issue - State: closed - Opened by moseshu over 1 year ago - 1 comment

#6 - Fix filter batch

Pull Request - State: closed - Opened by llehtahw over 1 year ago

#5 - Reduce some gpu ops

Pull Request - State: closed - Opened by llehtahw over 1 year ago

#4 - why not support the n parameter?

Issue - State: closed - Opened by BaiMoHan over 1 year ago - 2 comments

#2 - Llama2 and llama-30b does not work

Issue - State: closed - Opened by sureshbhusare over 1 year ago - 26 comments

#1 - [feat]: support safetensors for llama && llama2

Pull Request - State: closed - Opened by XFPlus over 1 year ago - 1 comment