lm-sys / lm-sys/FastChat

Add TemuClaude to Chatbot Arena — OpenAI-compatible API endpoint

Open
#3,899 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
39.5k
Forks
4.8k
PR merge metrics
No merged PRs in 30d

Description

## Model Submission: TemuClaude

I'd like to add **TemuClaude** to the Chatbot Arena for blind testing.

### API Details

- **Endpoint:** `https://saifulbusiness47--temuclaude-serve.modal.run/v1/chat/completions`
- **Format:** OpenAI-compatible Chat Completions API
- **Method:** POST
- **No API key required** (open access for evaluation)
- **Health check:** GET `https://saifulbusiness47--temuclaude-serve.modal.run/health`
- **Model list:** GET `https://saifulbusiness47--temuclaude-serve.modal.run/v1/models`
- **Model name:** `temuclaude`

### About TemuClaude

TemuClaude is a multi-model orchestration pipeline — not a single model. It routes queries across 8 models based on difficulty:

1. **Trivial queries** → cheapest model (Hy3 Preview)
2. **Medium queries** → specialist routing (DeepSeek for math, MiniMax for creative, MiMo for multimodal)
3. **Hard queries** → full Mixture-of-Agents pipeline:
- 3 models propose in parallel (GLM-5.2 + DeepSeek V4 Pro + Gemini 3.5 Flash)
- Aggregation by GLM-5.2 (consensus analysis, contradiction resolution)
- QA gate scoring by Nemotron (5-rubric, independent judge)
- Reflexion retry if QA < 8 (DeepSeek with feedback)
- Frontier fallback if QA < 6 (Claude Sonnet 5)
- Self-consistency for math (3 samples, majority vote)

### Test Command

```bash
curl -X POST https://saifulbusiness47--temuclaude-serve.modal.run/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{"model":"temuclaude","messages":[{"role":"user","content":"What is the capital of France?"}]}'
```

### Hosting

Hosted on Modal (serverless, auto-scaling). Max duration 120s per request. Timeout safeguard at 45s with single-model fallback.

The endpoint is live and ready for integration.

**Contact:** saifulbusiness47@gmail.com

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No repository files or tests are named. Start by locating FastChat's Chatbot Arena model-registration entry point, then verify the supplied health, model-list, and chat-completions endpoints with the curl example. Done means TemuClaude is available for blind testing under the model name `temuclaude` without breaking existing arena models.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai, api
Issue type
Feature
Difficulty
3/5
Estimated time
1-2 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.