Merge pull request #1352 from jlw463195395/master

修复deepseekcoder爆显存,加入int8,int4通用加载量化。
This commit is contained in:
binary-husky
2023-12-06 21:37:05 +08:00
committed by GitHub
4 changed files with 53 additions and 5 deletions

View File

@@ -167,6 +167,14 @@ git clone --depth=1 https://github.com/OpenLMLab/MOSS.git request_llms/moss #
# 【可选步骤IV】确保config.py配置文件的AVAIL_LLM_MODELS包含了期望的模型目前支持的全部模型如下(jittorllms系列目前仅支持docker方案)
AVAIL_LLM_MODELS = ["gpt-3.5-turbo", "api2d-gpt-3.5-turbo", "gpt-4", "api2d-gpt-4", "chatglm", "moss"] # + ["jittorllms_rwkv", "jittorllms_pangualpha", "jittorllms_llama"]
# 【可选步骤V】支持本地模型INT8,INT4量化这里所指的模型本身不是量化版本目前deepseek-coder支持后面测试后会加入更多模型量化选择
pip install bitsandbyte
# windows用户安装bitsandbytes需要使用下面bitsandbytes-windows-webui
python -m pip install bitsandbytes --prefer-binary --extra-index-url=https://jllllll.github.io/bitsandbytes-windows-webui
pip install -U git+https://github.com/huggingface/transformers.git
pip install -U git+https://github.com/huggingface/accelerate.git
pip install peft
```
</p>