Skip to content

配置不上vllm部署的qwen #270

@rongchenrui

Description

@rongchenrui

这是我curl访问的:
(base) rcr@zsyz-pc:/data/ccb$ curl http://localhost:8889/v1/chat/completions \

-H "Content-Type: application/json"
-d '{
"model": "qwen2.5-vl-7b",
"messages": [
{
"role": "user",
"content": "nihao"
}
],
"temperature": 0,
"max_tokens": 512
}'
{"id":"chatcmpl-c744e8304b264aaba703678ddc8735c7","object":"chat.completion","created":1776217541,"model":"qwen2.5-vl-7b","choices":[{"index":0,"message":{"role":"assistant","reasoning_content":null,"content":"你好!很高兴见到你。请问有什么我可以帮助你的吗?","tool_calls":[]},"logprobs":null,"finish_reason":"stop","stop_reason":null}],"usage":{"prompt_tokens":21,"total_tokens":35,"completion_tokens":14,"prompt_tokens_details":null},"prompt_logprobs":null}

这是配置信息:
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Login

OpenAI Compatible API Setup

Configure an OpenAI Chat Completions compatible endpoint (e.g. Ollama, DeepSeek, vLLM).

Base URL   http://192.168.192.195:8889/v1
                                                          
API Key    sk-123
                                                             
Haiku      qwen2.5-vl-7b
                                              
Sonnet     qwen2.5-vl-7b
                                                                    
Opus       qwen2.5-vl-7b

↑↓/Tab to switch · Enter on last field to save · Esc to go back

Esc to cancel

这是访问不了:
❯ /login
⎿ Login successful

❯ nihao
⎿  API Error: 400 status code (no body)

❯ /model

────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Select model
Switch between Claude models. Applies to this session and future Claude Code sessions. For other/previous model names, specify with --model.

1. Default (recommended)  Use the default model (currently qwen2.5-vl-7b)                                                                                                             

❯ 2. qwen2.5-vl-7b ✔ Custom Sonnet model
3. qwen2.5-vl-7b Custom Opus model
4. qwen2.5-vl-7b Custom Haiku model

○ Effort not supported for qwen2.5-vl-7b

Enter to confirm · Esc to exit

❯ /login
⎿ Login successful

❯ nihao
⎿  API Error: 400 status code (no body)

❯ /model
⎿ Set model to qwen2.5-vl-7b

❯ nihao
⎿  API Error: 400 status code (no body)

────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
❯ 
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
? for shortcuts

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions