gpt-oss-120b
OpenAIOpen weights131k context
gpt-oss-120b is an open-weights model from OpenAI with a 131k-token context window. 15 providers list it, from $0.037 per million input tokens and $0.17 per million output tokens (DeepInfra).
- Maker
- OpenAI
- Licence
- Open weights
- Context window
- 131k
- Providers listed
- 15
- Lowest input, per 1M
- $0.037
- Lowest output, per 1M
- $0.17
- Effort settings
- Low, Medium, High
Providers serving gpt-oss-120b
Cheapest first, by the blended rate of three input tokens to one output token. Prices in US dollars per million tokens, as of 26 Oct 2026.
| Provider | Input | Output | Context | Tools | JSON mode | OpenAI-compatible | Model ID |
|---|---|---|---|---|---|---|---|
DeepInfra | $0.037 | $0.17 | 131k | Yes | Yes | No | openai/gpt-oss-120b |
Novita | $0.05 | $0.25 | 131k | Yes | Yes | Yes | openai/gpt-oss-120b |
Google Vertex | $0.09 | $0.36 | 131k | Yes | Yes | No | openai/gpt-oss-120b-maas |
Baseten | $0.10 | $0.50 | 128k | Yes | Yes | — | gpt-oss-120b |
Amazon Bedrock | $0.15 | $0.60 | 131k | Yes | Yes | No | openai.gpt-oss-120b-1:0 |
Databricks Billed in DBUs; converted at Databricks' $0.07 per DBU (AWS). | $0.15 | $0.60 | 131k | Yes | Yes | Yes | databricks-gpt-oss-120b |
Fireworks | $0.15 | $0.60 | 131k | Yes | Yes | Yes | accounts/fireworks/models/gpt-oss-120b |
Groq | $0.15 | $0.60 | 131k | Yes | Yes | No | openai/gpt-oss-120b |
Microsoft Azure | $0.15 | $0.60 | 131k | Yes | Yes | — | gpt-oss-120b |
Nebius | $0.15 | $0.60 | 131k | Yes | Yes | Yes | openai/gpt-oss-120b |
Together AI | $0.15 | $0.60 | 131k | Yes | Yes | No | openai/gpt-oss-120b |
SambaNova | $0.22 | $0.59 | 131k | Yes | Yes | Yes | gpt-oss-120b |
Cerebras | $0.35 | $0.75 | 131k | Yes | Yes | No | gpt-oss-120b |
Cloudflare | $0.35 | $0.75 | 128k | Yes | Yes | Yes | @cf/openai/gpt-oss-120b |
Scaleway Priced in EUR only: €0.15 in / €0.60 out per 1M tokens (not converted) | — | — | 128k | Yes | — | Yes | gpt-oss-120b |
Benchmark scores
Published scores only, each linked to its source. Compare every model in Bench rankings