Model Comparator
Compare up to four models using the same criteria.
2 models compared
- gpt-oss-20b (High)
OpenAI
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | gpt-oss-20b (High) OpenAI | Add a model · 2/4 | |
|---|---|---|---|
| Intelligence | 10 | 10 | |
| Coding | Best: 20.9 | 20.7 | |
| Math | 30.7 | Best: 89.3 | |
| Agentic | — | 1.2 | |
| Speed | 123 t/s | Best: 148 t/s | |
| TTFT | 1.09 s | Best: 0.72 s | |
| Input · $ / 1M tokens | $1.25×25 relative to the lowest price among the compared models | Best: $0.050 | |
| Output · $ / 1M tokens | $10×45.5 relative to the lowest price among the compared models | Best: $0.22 | |
| Context window | 1M | 1M | |
| Reasoning | No | Yes | |
| Open weights | No | Yes | |
| Cost / task | — | $0.0106 |
Intelligence
- Nova 2.0 Pro Preview (Non-reasoning)
- 10
- gpt-oss-20b (High)
- 10
Coding
- Nova 2.0 Pro Preview (Non-reasoning)
- Best: 20.9
- gpt-oss-20b (High)
- 20.7
Math
- Nova 2.0 Pro Preview (Non-reasoning)
- 30.7
- gpt-oss-20b (High)
- Best: 89.3
Agentic
- Nova 2.0 Pro Preview (Non-reasoning)
- —
- gpt-oss-20b (High)
- 1.2
Speed
- Nova 2.0 Pro Preview (Non-reasoning)
- 123 t/s
- gpt-oss-20b (High)
- Best: 148 t/s
TTFT
- Nova 2.0 Pro Preview (Non-reasoning)
- 1.09 s
- gpt-oss-20b (High)
- Best: 0.72 s
Input · $ / 1M tokens
- Nova 2.0 Pro Preview (Non-reasoning)
- $1.25×25 relative to the lowest price among the compared models
- gpt-oss-20b (High)
- Best: $0.050
Output · $ / 1M tokens
- Nova 2.0 Pro Preview (Non-reasoning)
- $10×45.5 relative to the lowest price among the compared models
- gpt-oss-20b (High)
- Best: $0.22
Context window
- Nova 2.0 Pro Preview (Non-reasoning)
- 1M
- gpt-oss-20b (High)
- 1M
Reasoning
- Nova 2.0 Pro Preview (Non-reasoning)
- No
- gpt-oss-20b (High)
- Yes
Open weights
- Nova 2.0 Pro Preview (Non-reasoning)
- No
- gpt-oss-20b (High)
- Yes
Cost / task
- Nova 2.0 Pro Preview (Non-reasoning)
- —
- gpt-oss-20b (High)
- $0.0106
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.
- Nova 2.0 Pro Preview (Non-reasoning)10 · $3.44
- gpt-oss-20b (High)10 · $0.092