## Feature Request

### What's missing?

`elevenlabs/scribe_v2` is not present in `model_prices_and_context_window.json`, so cost tracking for this model does not work in the LiteLLM proxy.

Scribe v2 was released by ElevenLabs in January 2026 and is their current recommended batch speech-to-text model. The LiteLLM docs and config examples only reference `elevenlabs/scribe_v1` — `scribe_v2` has never been added.

### Proposed entry

ElevenLabs prices Scribe v2 at **$0.22/hour** of audio, which converts to `$0.0000611111` per second:

```json "elevenlabs/scribe_v2": { "litellm_provider": "elevenlabs", "mode": "audio_transcription", "input_cost_per_second": 0.0000611111, "output_cost_per_second": 0 } ```

### References - ElevenLabs Scribe v2 announcement: https://elevenlabs.io/blog/introducing-scribe-v2 - ElevenLabs pricing: https://elevenlabs.io/pricing - Related issue: #8852 (Add ElevenLabs as Provider) - Related issue: #18058 (Cost calculation not working for ElevenLabs) - LiteLLM docs on adding model pricing: https://docs.litellm.ai/docs/provider_registration/add_model_pricing

### Workaround

Users can currently override pricing in their `config.yaml` via `model_info`: …