Model Comparator
Compare up to four models using the same criteria.
2 models compared
- gpt-oss-120b (High)
OpenAI
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | gpt-oss-120b (High) OpenAI | Add a model · 2/4 | |
|---|---|---|---|
| Intelligence | Best: 11.7 | 11.6 | |
| Coding | — | 30.4 | |
| Math | — | 93.4 | |
| Agentic | — | 3.7 | |
| Speed | 162 t/s | Best: 221 t/s | |
| TTFT | Best: 0.6 s | 0.83 s | |
| Input · $ / 1M tokens | $0.20×1.3 relative to the lowest price among the compared models | Best: $0.15 | |
| Output · $ / 1M tokens | $1.25×2.1 relative to the lowest price among the compared models | Best: $0.59 | |
| Context window | Best: 400K | 131K | |
| Reasoning | No | Yes | |
| Open weights | No | Yes | |
| Cost / task | — | $0.107 |
Intelligence
- GPT-5.4 nano (Non-reasoning)
- Best: 11.7
- gpt-oss-120b (High)
- 11.6
Coding
- GPT-5.4 nano (Non-reasoning)
- —
- gpt-oss-120b (High)
- 30.4
Math
- GPT-5.4 nano (Non-reasoning)
- —
- gpt-oss-120b (High)
- 93.4
Agentic
- GPT-5.4 nano (Non-reasoning)
- —
- gpt-oss-120b (High)
- 3.7
Speed
- GPT-5.4 nano (Non-reasoning)
- 162 t/s
- gpt-oss-120b (High)
- Best: 221 t/s
TTFT
- GPT-5.4 nano (Non-reasoning)
- Best: 0.6 s
- gpt-oss-120b (High)
- 0.83 s
Input · $ / 1M tokens
- GPT-5.4 nano (Non-reasoning)
- $0.20×1.3 relative to the lowest price among the compared models
- gpt-oss-120b (High)
- Best: $0.15
Output · $ / 1M tokens
- GPT-5.4 nano (Non-reasoning)
- $1.25×2.1 relative to the lowest price among the compared models
- gpt-oss-120b (High)
- Best: $0.59
Context window
- GPT-5.4 nano (Non-reasoning)
- Best: 400K
- gpt-oss-120b (High)
- 131K
Reasoning
- GPT-5.4 nano (Non-reasoning)
- No
- gpt-oss-120b (High)
- Yes
Open weights
- GPT-5.4 nano (Non-reasoning)
- No
- gpt-oss-120b (High)
- Yes
Cost / task
- GPT-5.4 nano (Non-reasoning)
- —
- gpt-oss-120b (High)
- $0.107
Each reasoning level remains a separate configuration.
Score and cost
Each line joins the reasoning levels of a compared model; the large point marks the configuration in the table. The top-right corner holds the best score for the lowest cost. Select another point to compare that level.
- GPT-5.4 nano (Non-reasoning)11.7 · $0.463
- gpt-oss-120b (High)11.6 · $0.26