# Together AI - gemma-4-31b grades

1 graded endpoint serving gemma-4-31b.
HTML version: [https://inferencecanary.com/togetherai](https://inferencecanary.com/togetherai) · matrix: [https://inferencecanary.com/index.md](https://inferencecanary.com/index.md)


## Endpoints

| Endpoint | Dialect | Inference | Billing | Probed |
|---|---|---|---|---|
| [Together AI](https://inferencecanary.com/togetherai/gemma-4-31b.md) | OpenAI chat/completions | B | OK | 2026-08-25 |


## Findings affecting Together AI

- **[high · inference] Several providers drop replayed reasoning: the model never sees its own prior thinking** - details: [/replayed-reasoning-dropped.md](https://inferencecanary.com/replayed-reasoning-dropped.md)