# moonshotai/kimi-k2.6 across inference providers

Reference: context 262,144 tokens from https://huggingface.co/moonshotai/Kimi-K2.6/blob/7eb5002f6aadc958aed6a9177b7ed26bb94011bb/config.json (text_config.max_position_embeddings); quantization native INT4 (group size 32).

Provider rows: https://openrouter.ai/api/v1/models/moonshotai/kimi-k2.6/endpoints, read 2026-10-19. Declared values; no requests sent.

| Provider | Tag | Quantization | Context | Max output | Tools | Reasoning | Differs from the reference |
|---|---|---|---|---|---|---|---|
| AtlasCloud | atlas-cloud/int4 | int4 (same) | 262,144 | 235,929 | yes | yes | none declared |
| Baidu | baidu/fp4 | fp4 (different) | 262,144 | 235,929 | yes | yes | **declares fp4 weights; the official release is native INT4 (group size 32)** |
| Chutes | chutes/int4 | int4 (same) | 262,144 | 65,535 | yes | yes | none declared |
| Cloudflare | cloudflare | none (undisclosed) | 262,144 | 235,929 | yes | yes | none declared |
| CoreWeave | coreweave/fp4 | fp4 (different) | 262,144 | 235,929 | yes | yes | **declares fp4 weights; the official release is native INT4 (group size 32)** |
| Crusoe | crusoe/bf16 | bf16 (higher) | 262,144 | 235,929 | yes | yes | none declared |
| DeepInfra | deepinfra/fp4 | fp4 (different) | 262,144 | 16,384 | yes | yes | **declares fp4 weights; the official release is native INT4 (group size 32)** |
| DigitalOcean | digitalocean | none (undisclosed) | 262,144 | 235,929 | yes | yes | none declared |
| GMICloud | gmicloud/fp8 | fp8 (higher) | 262,144 | 235,929 | yes | yes | none declared |
| Inceptron | inceptron/int4 | int4 (same) | 262,144 | 235,929 | yes | yes | none declared |
| Moonshot AI | moonshotai/int4 | int4 (same) | 262,144 | 235,929 | yes | yes | none declared |
| Novita | novita | none (undisclosed) | 262,144 | 235,929 | yes | yes | none declared |
| Parasail | parasail/int4 | int4 (same) | 262,144 | 235,929 | yes | yes | none declared |
| Phala | phala | none (undisclosed) | 262,144 | 235,929 | yes | yes | none declared |
| SiliconFlow | siliconflow/fp8 | fp8 (higher) | 262,144 | 235,929 | yes | yes | none declared |
| StreamLake | streamlake/fp8 | fp8 (higher) | 256,000 | 230,400 | yes | yes | declares a 256,000-token context; the reference config says 262,144 (98%) |
| Venice | venice/int4 | int4 (same) | 256,000 | 65,536 | yes | yes | declares a 256,000-token context; the reference config says 262,144 (98%) |

## Changes since tracking began

None yet.