Together AI
1 graded endpoint serving gemma-4-31b · back to the matrix
| Endpoint | Harness | Billing | Faithfulness | Probed |
|---|---|---|---|---|
| Together AI OpenAI chat/completions | B− | ✓ | 26/29 | 2026-07-27 |
Findings affecting Together AI
-
red harness
Four providers drop replayed reasoning: the model never sees its own prior thinking
The gemma-4 chat template renders reasoning replayed on model turns, and the model card mandates preserving thinking across tool-call turns — the reference provider reproduces this live. These four providers delete it... -
yellow harness
Tool results replayed out of order are mis-paired on most providers
The chat dialect's contract is that a tool message is matched to its call by tool_call_id, in any order. After two parallel calls, replaying the results reversed swaps the data between them on these providers — the...