kvcache-ai / kvcache-ai/ktransformers

部署Qwen3.6-35B-A3B-FP8出现的问题

Open
#2,040 4 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
19.5k
Forks
1.6k
Avg merge
19h 32m
Merged PRs (30d)
27

Description

### Reminder

- [x] I have read the above rules and searched the existing issues.

### System Info

本地硬件是4090+Intel(R) Xeon(R) Gold 5222 CPU,采用官方的镜像一直部署不上去,出现EOFError,错误为TP MOE layer 0, pool: 0x4ed44e30, expert num: 256, num_experts_per_tok: 8
Creating AMX_MOE_TP 0 at numa 0
Creating AMX_MOE_TP 1 at numa 1
terminate called after throwing an instance of 'std::runtime_error'
what(): no weight source
Fatal Python error: Aborted
有朋友部署成功过么?

### Reproduction

```text
Put your message here.
```

### Others

_No response_

Contributor guide

Open the contributing guide

Research direction

The reproduction section is empty, so start by recording the exact official image and deployment command used with the 4090 and Xeon Gold 5222 system. Trace the first failure around the reported “no weight source” error and AMX_MOE initialization; done should be a reproducible diagnosis or a clearly documented compatible setup.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.