feat(together-ai): update model YAMLs [bot]

[models-bot]

Auto-generated by poc-agent for provider together-ai.

---

Note

Low Risk
Low risk config-only change to a single provider model YAML; main risk is mis-declaring capabilities (e.g., supported modes/features) which could affect routing or UI expectations.

Overview
Updates providers/together-ai/mistralai/Voxtral-Mini-3B-2507.yaml to replace the previous unknown mode with audio transcription configuration and to declare additional model metadata.

Adds declared features (function_calling), limits (context_window: 32000), explicit modalities (audio/text in, text out), provisioning: serverless, status: active, a model sources URL, and expands supportedModes to include chat.

^{Reviewed by Cursor Bugbot for commit 70e3d8b. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

/test-models

[harshiv-26]

Gateway test results

• Total: 2
• Passed: 0
• Failed: 0
• Validation failed: 0
• Errored: 2
• Skipped: 0
• Success rate: 0.0%

Provider	Model	Scenarios
together-ai	mistralai/Voxtral-Mini-3B-2507	error: tool-call, params

Failures (2)

together-ai/mistralai/Voxtral-Mini-3B-2507 — tool-call (error)

Error:

Snippet request error: Code snippet request failed for model test-v2-together-ai/mistralai-Voxtral-Mini-3B-2507 [500]: {"statusCode":500,"message":"Unknown error happened"}

together-ai/mistralai/Voxtral-Mini-3B-2507 — params (error)

Error:

Snippet request error: Code snippet request failed for model test-v2-together-ai/mistralai-Voxtral-Mini-3B-2507 [500]: {"statusCode":500,"message":"Unknown error happened"}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 70e3d8b. Configure here.}

[cursor]

Token pricing inconsistent with audio_transcription mode models

Medium Severity

This model uses input_cost_per_token/output_cost_per_token with mode: audio_transcription, but every other audio_transcription model in the together-ai provider uses input_cost_per_second pricing. Downstream cost-calculation logic that expects input_cost_per_second for audio_transcription models would fail to find a cost entry or compute an incorrect value for this model.

Additional Locations (1)

• providers/together-ai/mistralai/Voxtral-Mini-3B-2507.yaml#L14-L15

 

^{Reviewed by Cursor Bugbot for commit 70e3d8b. Configure here.}