Model Comparator
Compare up to four models using the same criteria.
2 models compared
- Gemma 4 31B (Reasoning)
Google
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | Gemma 4 31B (Reasoning) | Add a model · 2/4 | |
|---|---|---|---|
| Intelligence | Best: 14.8 | 14.7 | |
| Coding | Best: 43.5 | 43.4 | |
| Math | 53.7 | — | |
| Agentic | Best: 6.8 | 4.2 | |
| Speed | Best: 55.1 t/s | 35.4 t/s | |
| TTFT | Best: 0.75 s | 1.07 s | |
| Input · $ / 1M tokens | $0.27 | — | |
| Output · $ / 1M tokens | $1 | — | |
| Context window | 1M | 1M | |
| Reasoning | No | Yes | |
| Open weights | Yes | Yes | |
| Cost / task | — | $0 |
Intelligence
- DeepSeek V3.1 Terminus (Non-reasoning)
- Best: 14.8
- Gemma 4 31B (Reasoning)
- 14.7
Coding
- DeepSeek V3.1 Terminus (Non-reasoning)
- Best: 43.5
- Gemma 4 31B (Reasoning)
- 43.4
Math
- DeepSeek V3.1 Terminus (Non-reasoning)
- 53.7
- Gemma 4 31B (Reasoning)
- —
Agentic
- DeepSeek V3.1 Terminus (Non-reasoning)
- Best: 6.8
- Gemma 4 31B (Reasoning)
- 4.2
Speed
- DeepSeek V3.1 Terminus (Non-reasoning)
- Best: 55.1 t/s
- Gemma 4 31B (Reasoning)
- 35.4 t/s
TTFT
- DeepSeek V3.1 Terminus (Non-reasoning)
- Best: 0.75 s
- Gemma 4 31B (Reasoning)
- 1.07 s
Input · $ / 1M tokens
- DeepSeek V3.1 Terminus (Non-reasoning)
- $0.27
- Gemma 4 31B (Reasoning)
- —
Output · $ / 1M tokens
- DeepSeek V3.1 Terminus (Non-reasoning)
- $1
- Gemma 4 31B (Reasoning)
- —
Context window
- DeepSeek V3.1 Terminus (Non-reasoning)
- 1M
- Gemma 4 31B (Reasoning)
- 1M
Reasoning
- DeepSeek V3.1 Terminus (Non-reasoning)
- No
- Gemma 4 31B (Reasoning)
- Yes
Open weights
- DeepSeek V3.1 Terminus (Non-reasoning)
- Yes
- Gemma 4 31B (Reasoning)
- Yes
Cost / task
- DeepSeek V3.1 Terminus (Non-reasoning)
- —
- Gemma 4 31B (Reasoning)
- $0
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.
- DeepSeek V3.1 Terminus (Non-reasoning)14.8 · 55.1 t/s
- Gemma 4 31B (Reasoning)14.7 · 35.4 t/s