# openai/gpt-oss-120b across inference providers

Reference: context 131,072 tokens from https://huggingface.co/openai/gpt-oss-120b/blob/b5c939de8f754692c1647ca79fbf85e8c1e70f8a/config.json (max_position_embeddings); quantization MXFP4 on the MoE experts, bf16 elsewhere.

Provider rows: https://openrouter.ai/api/v1/models/openai/gpt-oss-120b/endpoints, read 2026-10-19. Declared values; no requests sent.

| Provider | Tag | Quantization | Context | Max output | Tools | Reasoning | Differs from the reference |
|---|---|---|---|---|---|---|---|
| AkashML | akashml/bf16 | bf16 (higher) | 131,072 | 117,964 | yes | yes | none declared |
| Amazon Bedrock | amazon-bedrock | none (undisclosed) | 131,072 | 117,964 | no | yes | **declares no tools parameter; the official model supports tool calling** |
| Amazon Bedrock | amazon-bedrock/eu-west-1 | none (undisclosed) | 131,072 | 117,964 | no | yes | **declares no tools parameter; the official model supports tool calling** |
| BaseTen | baseten/fp4 | fp4 (same) | 128,072 | 115,264 | yes | yes | declares a 128,072-token context; the reference config says 131,072 (98%) |
| BaseTen | baseten/fp4/us | fp4 (same) | 128,072 | 115,264 | yes | yes | declares a 128,072-token context; the reference config says 131,072 (98%) |
| Cerebras | cerebras/fp16 | fp16 (higher) | 131,072 | 40,960 | yes | yes | none declared |
| CoreWeave | coreweave/fp4 | fp4 (same) | 131,072 | 117,964 | yes | yes | none declared |
| Crusoe | crusoe/bf16 | bf16 (higher) | 131,072 | 117,964 | yes | yes | none declared |
| DeepInfra | deepinfra/bf16 | bf16 (higher) | 131,072 | 117,964 | yes | yes | none declared |
| DekaLLM | dekallm/bf16 | bf16 (higher) | 131,072 | 117,964 | yes | yes | none declared |
| DigitalOcean | digitalocean | none (undisclosed) | 128,000 | 4,096 | no | yes | declares a 128,000-token context; the reference config says 131,072 (98%); **declares no tools parameter; the official model supports tool calling**; **declares at most 4,096 output tokens, which the reasoning tokens also count against** |
| Google | google-vertex/global | none (undisclosed) | 131,072 | 117,964 | no | yes | **declares no tools parameter; the official model supports tool calling** |
| Groq | groq | none (undisclosed) | 131,072 | 65,536 | yes | yes | none declared |
| Mancer 2 | mancer/fp8 | fp8 (higher) | 131,072 | 117,964 | yes | yes | none declared |
| Mara | mara | none (undisclosed) | 131,072 | 117,964 | yes | yes | none declared |
| Near AI | near-ai/fp4 | fp4 (same) | 131,072 | 117,964 | yes | yes | none declared |
| Nebius | nebius/fp4 | fp4 (same) | 131,072 | 117,964 | yes | yes | none declared |
| Novita | novita/fp4 | fp4 (same) | 131,072 | 32,768 | yes | yes | none declared |
| Parasail | parasail/fp4 | fp4 (same) | 131,072 | 117,964 | yes | yes | none declared |
| Phala | phala | none (undisclosed) | 131,072 | 32,768 | yes | yes | none declared |
| SambaNova | sambanova | none (undisclosed) | 131,072 | 117,964 | yes | yes | none declared |
| SiliconFlow | siliconflow/fp8 | fp8 (higher) | 131,072 | 8,192 | no | yes | **declares no tools parameter; the official model supports tool calling** |
| Together | together | none (undisclosed) | 131,072 | 117,964 | yes | yes | none declared |
| Venice | venice/int4 | int4 (different) | 128,000 | 32,768 | yes | yes | declares a 128,000-token context; the reference config says 131,072 (98%); **declares int4 weights; the official release is MXFP4 on the MoE experts, bf16 elsewhere** |

## Changes since tracking began

None yet.