GPT-OSS 120B
OpenAIOpen weightsNon-reasoningVerifiedReleased Aug 5, 2025Input: text
Overall score
37.3
Rank 153 of 222 · 23.5–51.0
Context window
128K
128,000 tokens
Price per 1M tokens
$0.15 in · $0.60 out
Source: OpenRouter
Blended price
$0.26
3:1 input:output tokens
Output speed
—
Not measured yet
Category scores
Scores per capability area with the model's rank among ranked models in that category.
- 1Instruction following#5084.3
- 2Reasoning60.7
- 3Coding45.4
- 4Knowledge34.1
Benchmark results (9)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Knowledge
| GPQA | 80.1 | #59 of 117 |
| HLE | 14.9 | #64 of 69 |
| MMLU | 90.0 | #6 of 28 |
Coding
| SWE-bench Verified | 62.4 | #72 of 98 |
| React Native Evals | 71.6 | #12 of 15 |
Math
| AIME 2025 | 92.5 | #10 of 38 |
| AIME 2024 | 95.8 | #1 of 19 |
Agentic & tool use
| Gert Labs | 29.61 | #49 of 52 |
Multimodal
| Design Arena Website | 969 | #87 of 91 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| List price (OpenRouter) | $0.15 | $0.60 | — |
| OpenRouter openai/gpt-oss-120b | $0.15 | $0.60 | $0.075 |
Open-weight model. Self-hosted or third-party hosted costs vary by provider and infrastructure.
Arena ratings
Human-preference ratings from LMArena. Best-rated variant shown.
| Text | 1366 | #192 · 30.2K votes |
Open weights
From the Hugging Face repository.
- Parameters
- 117B
- Licence
- apache-2.0
- Downloads (30 days)
- 3.1M
- Repository
- openai/gpt-oss-120b ↗
Speed
Measured by us through OpenRouter (median of trials).
Not measured yet.
Related models
Same family
- GPT-OSS 20B33.3
Similar score
- GPT-5 nano37.3 · $0.14
- GPT-4.137.6 · $3.50
- LFM2.5-24B-A2B37.7 · $0.33
- Claude 4 Sonnet38.1 · $6.00
- DeepSeek V3 032436.5 · $0.50