Model Comparator
Compare up to four models using the same criteria.
2 models compared
- DeepSeek V3 0324
DeepSeek
- Qwen3 Coder 30B A3B Instruct
Alibaba
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | DeepSeek V3 0324 DeepSeek | Qwen3 Coder 30B A3B Instruct Alibaba | Add a model · 2/4 |
|---|---|---|---|
| Intelligence | Best: 9.7 | 9.6 | |
| Coding | 21.2 | — | |
| Math | Best: 41 | 29 | |
| Agentic | 0.8 | — | |
| Speed | 109 t/s | Best: 156 t/s | |
| TTFT | Best: 0.71 s | 1.01 s | |
| Input · $ / 1M tokens | $0.24×1.6 relative to the lowest price among the compared models | Best: $0.15 | |
| Output · $ / 1M tokens | $0.90×1.2 relative to the lowest price among the compared models | Best: $0.748 | |
| Context window | 128K | Best: 262K | |
| Reasoning | No | No | |
| Open weights | Yes | Yes | |
| Cost / task | $0.0211 | — |
Intelligence
- DeepSeek V3 0324
- Best: 9.7
- Qwen3 Coder 30B A3B Instruct
- 9.6
Coding
- DeepSeek V3 0324
- 21.2
- Qwen3 Coder 30B A3B Instruct
- —
Math
- DeepSeek V3 0324
- Best: 41
- Qwen3 Coder 30B A3B Instruct
- 29
Agentic
- DeepSeek V3 0324
- 0.8
- Qwen3 Coder 30B A3B Instruct
- —
Speed
- DeepSeek V3 0324
- 109 t/s
- Qwen3 Coder 30B A3B Instruct
- Best: 156 t/s
TTFT
- DeepSeek V3 0324
- Best: 0.71 s
- Qwen3 Coder 30B A3B Instruct
- 1.01 s
Input · $ / 1M tokens
- DeepSeek V3 0324
- $0.24×1.6 relative to the lowest price among the compared models
- Qwen3 Coder 30B A3B Instruct
- Best: $0.15
Output · $ / 1M tokens
- DeepSeek V3 0324
- $0.90×1.2 relative to the lowest price among the compared models
- Qwen3 Coder 30B A3B Instruct
- Best: $0.748
Context window
- DeepSeek V3 0324
- 128K
- Qwen3 Coder 30B A3B Instruct
- Best: 262K
Reasoning
- DeepSeek V3 0324
- No
- Qwen3 Coder 30B A3B Instruct
- No
Open weights
- DeepSeek V3 0324
- Yes
- Qwen3 Coder 30B A3B Instruct
- Yes
Cost / task
- DeepSeek V3 0324
- $0.0211
- Qwen3 Coder 30B A3B Instruct
- —
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 03249.7 · $0.405
- Qwen3 Coder 30B A3B Instruct9.6 · $0.30