Model Comparator
Compare up to four models using the same criteria.
2 models compared
- Qwen3.8 27B (Non-reasoning)
Alibaba
- gpt-oss-20b (High)
OpenAI
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | Qwen3.8 27B (Non-reasoning) Alibaba | gpt-oss-20b (High) OpenAI | Add a model · 2/4 |
|---|---|---|---|
| Intelligence | 10 | 10 | |
| Coding | Best: 44.6 | 20.7 | |
| Math | — | 89.3 | |
| Agentic | Best: 22.4 | 1.2 | |
| Speed | 43.6 t/s | Best: 148 t/s | |
| TTFT | 3.89 s | Best: 0.72 s | |
| Input · $ / 1M tokens | $0.50×10 relative to the lowest price among the compared models | Best: $0.050 | |
| Output · $ / 1M tokens | $3×13.6 relative to the lowest price among the compared models | Best: $0.22 | |
| Context window | 256K | Best: 1M | |
| Reasoning | No | Yes | |
| Open weights | Yes | Yes | |
| Cost / task | $3.77×356 relative to the lowest price among the compared models | Best: $0.0106 |
Intelligence
- Qwen3.8 27B (Non-reasoning)
- 10
- gpt-oss-20b (High)
- 10
Coding
- Qwen3.8 27B (Non-reasoning)
- Best: 44.6
- gpt-oss-20b (High)
- 20.7
Math
- Qwen3.8 27B (Non-reasoning)
- —
- gpt-oss-20b (High)
- 89.3
Agentic
- Qwen3.8 27B (Non-reasoning)
- Best: 22.4
- gpt-oss-20b (High)
- 1.2
Speed
- Qwen3.8 27B (Non-reasoning)
- 43.6 t/s
- gpt-oss-20b (High)
- Best: 148 t/s
TTFT
- Qwen3.8 27B (Non-reasoning)
- 3.89 s
- gpt-oss-20b (High)
- Best: 0.72 s
Input · $ / 1M tokens
- Qwen3.8 27B (Non-reasoning)
- $0.50×10 relative to the lowest price among the compared models
- gpt-oss-20b (High)
- Best: $0.050
Output · $ / 1M tokens
- Qwen3.8 27B (Non-reasoning)
- $3×13.6 relative to the lowest price among the compared models
- gpt-oss-20b (High)
- Best: $0.22
Context window
- Qwen3.8 27B (Non-reasoning)
- 256K
- gpt-oss-20b (High)
- Best: 1M
Reasoning
- Qwen3.8 27B (Non-reasoning)
- No
- gpt-oss-20b (High)
- Yes
Open weights
- Qwen3.8 27B (Non-reasoning)
- Yes
- gpt-oss-20b (High)
- Yes
Cost / task
- Qwen3.8 27B (Non-reasoning)
- $3.77×356 relative to the lowest price among the compared models
- gpt-oss-20b (High)
- Best: $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.
- Qwen3.8 27B (Non-reasoning)10 · $3.77
- gpt-oss-20b (High)10 · $0.0106