mirror of
https://github.com/kvcache-ai/ktransformers.git
synced 2026-11-05 01:07:28 +00:00
[GH-ISSUE #883] ktransformers --type transformers这个模式里config.json #638
Labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
kvcache-ai/ktransformers#638
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @113514089 on GitHub (Mar 13, 2025).
Original GitHub issue: https://github.com/kvcache-ai/ktransformers/issues/883
Originally assigned to: @qiyuxinlin, @KMSorSMS on GitHub.
用这个模式始终不得要领,但是又查不到相关资料,我想问下这个模式有什么好处没?还有这个模式对系统的要求是什么,还是和前面一样没变吗,4090 48g跑671b 要什么条件,设置什么参数 比如我现在就遇到一个问题:TypeError: GPTQConfig.init() missing 1 required positional argument: 'bits'
其实config.json里面参数很多,但是不知道该怎么设置,那里有相关得说明文件可以看
@qiyuxinlin commented on GitHub (Mar 20, 2025):
transformers 模式没有好处,并且很久没有维护,启动命令中只会有下列参数有用:
--model_path: safetensor config 目录;
--gguf_path: gguf 模型权重目录;
--max_new_tokens: 最大生成 token 个数
--cache_lens:最大上下文长度
--cpu_infer:启动进程数
--optimize_config_path: 注入规则 yaml 文件
--force_think: 强制输出 think
--port: server 启动端口号
我们尽量在下个版本中删除 config.json 没有用到的配置信息,并更新到文档中