OpenAI · Released Aug 5, 2025GPT OSS 120B
Open GPT reasoning model for self-hosted agents and controllable deployments.
- Capability
- 140.0ECI · #99 of 148
- Input
- $0.15per 1M tokens
- Output
- $0.60per 1M tokens
- Context
- 131K33K max output
What it costs
- Input
- $0.15per million tokens
- Output
- $0.60per million tokens
- Blended (3:1)
- $0.263cheaper than 73% of priced models
Typical monthly bills
| Monthly usage | Estimated cost |
|---|---|
| Side project2M input + 0.5M output tokens | $0.60 |
| Team assistant25M input + 5M output tokens | $6.75 |
| Production app250M input + 50M output tokens | $67.50 |
Median of 36 paid API providers on models.dev.
Independent benchmarks
#99 of 148 scored models
The shaded band is Epoch AI’s confidence range (135.3–142.3); the tick marks the median scored model.
Scores from Epoch AI, run independently of OpenAI.
The details
- Lab
- OpenAI
- Released
- Aug 5, 2025
- Context window
- 131,072 tokens
- Max output
- 32,768 tokens
- Inputs
- Text
- Output
- Text
- Reasoning
- Always on
- Tool calling
- Yes
- Structured output
- Yes
- Weights
- Open weights
- Availability
- 39 API providerslisted on models.dev
More from OpenAI
About GPT OSS 120B
How much does GPT OSS 120B cost?
GPT OSS 120B costs $0.15 input / $0.60 output per million tokens (median across 36 API providers). At a 3:1 input-to-output mix that is $0.263 per million tokens, cheaper than 73% of the 360 priced models we track.
What is the context window of GPT OSS 120B?
GPT OSS 120B accepts up to 131,072 tokens per request and can write up to 32,768 tokens in one response.
How good is GPT OSS 120B?
Epoch AI gives GPT OSS 120B a Capabilities Index score of 140.0 (likely range 135.3–142.3), ranking it #99 of 148 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 75.8%, OTIS Mock AIME 2024–2025 88.9%.
Is GPT OSS 120B open source?
Yes. OpenAI publishes the weights, so it can be downloaded and self-hosted.
What inputs does GPT OSS 120B support?
GPT OSS 120B accepts text and replies in text. It is a reasoning model, supports tool calling and can return structured JSON output.
When was GPT OSS 120B released?
OpenAI released GPT OSS 120B on Aug 5, 2025.
What are the best alternatives to GPT OSS 120B?
The closest current models from other labs on capability, price and release date are Qwen3 235B-A22B Instruct 2507 (Alibaba (Qwen), ECI 138.9, $0.15 / $0.75), DeepSeek-V3.1 (DeepSeek, ECI 139.9, $0.385 / $1.25), Gemini 2.5 Flash (Google, ECI 140.8, $0.30 / $2.50) and Mistral Small 3.2 (Mistral AI, ECI 131.7, $0.10 / $0.30).