Model Comparator
Compare up to four models using the same criteria.
2 models compared
- Sarvam 105B (High)
Sarvam
- Qwen3 4B 2507 (Reasoning)
Alibaba
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | Sarvam 105B (High) Sarvam | Qwen3 4B 2507 (Reasoning) Alibaba | Add a model · 2/4 |
|---|---|---|---|
| Intelligence | 8.8 | 8.8 | |
| Math | — | 82.7 | |
| Speed | 224 t/s | — | |
| TTFT | 1.05 s | — | |
| Input · $ / 1M tokens | $0.042 | — | |
| Output · $ / 1M tokens | $0.17 | — | |
| Context window | 128K | Best: 262K | |
| Reasoning | Yes | Yes | |
| Open weights | Yes | Yes |
Intelligence
- Sarvam 105B (High)
- 8.8
- Qwen3 4B 2507 (Reasoning)
- 8.8
Math
- Sarvam 105B (High)
- —
- Qwen3 4B 2507 (Reasoning)
- 82.7
Speed
- Sarvam 105B (High)
- 224 t/s
- Qwen3 4B 2507 (Reasoning)
- —
TTFT
- Sarvam 105B (High)
- 1.05 s
- Qwen3 4B 2507 (Reasoning)
- —
Input · $ / 1M tokens
- Sarvam 105B (High)
- $0.042
- Qwen3 4B 2507 (Reasoning)
- —
Output · $ / 1M tokens
- Sarvam 105B (High)
- $0.17
- Qwen3 4B 2507 (Reasoning)
- —
Context window
- Sarvam 105B (High)
- 128K
- Qwen3 4B 2507 (Reasoning)
- Best: 262K
Reasoning
- Sarvam 105B (High)
- Yes
- Qwen3 4B 2507 (Reasoning)
- Yes
Open weights
- Sarvam 105B (High)
- Yes
- Qwen3 4B 2507 (Reasoning)
- Yes
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.
- Sarvam 105B (High)8.8 · $0.074
Not plotted for lack of data on these axes: Qwen3 4B 2507 (Reasoning).