openai-gpt-56-luna via Venice.ai fails due to thinking level

I have Venice.ai configured as the model provider. I am trying to use openai-gpt-56-luna, which worked fine in the past. Now I get the following error.

Ran into this error: Request failed: Bad request (400): {"error":"Function tools with reasoning_effort are not supported for gpt-5.6-luna in /v1/chat/completions. To use function tools, use /v1/responses or set reasoning_effort to 'none'.","request_id":"r3mgV1AKXDuERfp_Gxj7V"}.

I am unable to use the /v1/responses option (instead of /v1/chat/completions), and I am unable to set the thinking/reasoning effort to none.

---

**To Reproduce**
Steps to reproduce the behavior:
1. Set up Venice.ai as the model provider with a Venice API key
2. Select openai-gpt-56-luna as the model
3. Try any prompt
4. See error above

---

**Expected behavior**
This model provider and model should work, and has worked fine in the past.

---

**Please provide the following information**
- **OS & Arch:** macos 25.6.0 aarch64
- **Interface:** UI
- **Version:** 1.48.0
- **Extensions enabled:** todo, tom, developer, skills, analyze, summon, Extension Manager, Playwright, scheduler
- **Provider & Model:** venice - openai-gpt-56-luna

[diagnostics_20260902_22.json](https://github.com/user-attachments/files/31748860/diagnostics_20260902_22.json)

Look at the Venice.ai provider integration code, likely in a directory like `src/providers/venice` or similar. Find where the API endpoint and request parameters (like `reasoning_effort`) are set for the `openai-gpt-56-luna` model. The fix involves either switching the endpoint to `/v1/responses` or ensuring `reasoning_effort` is set to 'none' for this model. Check existing tests for the provider to understand the expected behavior and run them after making changes.

Written by the indexing model from the issue text.