AI models
Qwen3.8 Max vs DeepSeek V4 Pro
Same criteria, side by side. Where the gap is small, it may be measurement noise.
Also searched as
Qwen Max vs DeepSeek V4
Direct answer: GPQA Diamond: Qwen3.8 Max 92.7%, DeepSeek V4 Pro 91.7% (Epoch AI). OTIS Mock AIME: Qwen3.8 Max 100%, DeepSeek V4 Pro 98.6% (Epoch AI). FrontierMath (tiers 1 to 3): Qwen3.8 Max 74.7%, DeepSeek V4 Pro 64.6% (Epoch AI). SimpleQA Verified: Qwen3.8 Max 47.3%, DeepSeek V4 Pro 52.9% (Epoch AI).
| Metric | Qwen3.8 Max | DeepSeek V4 Pro | Source |
|---|---|---|---|
| Weights ?what it is | closed | open | no data |
| Input US$/1M ? | US$ 2.00 | US$ 0.96 | no data |
| Output US$/1M ? | US$ 6.00 | US$ 1.91 | no data |
| Context ? | 1,000,000 | 1,048,576 | no data |
| GPQA Diamond ? | 92.7% | 91.7% | Epoch AI |
| OTIS Mock AIME ? | 100% | 98.6% | Epoch AI |
| FrontierMath (tiers 1 to 3) ? | 74.7% | 64.6% | Epoch AI |
| SimpleQA Verified ? | 47.3% | 52.9% | Epoch AI |
| SWE-bench Verified ? | no data | 77.6% | Epoch AI |
| GPQA Diamond (OpenRouter) ? | no data | 91.9% | OpenRouter Benchmarks |
| τ-bench Verified (airline) ? | no data | 78.6% | OpenRouter Benchmarks |
Compare with
Sources
- Epoch AI, CC BY 4.0, latest version we checked.
- OpenRouter Benchmarks, CC BY 4.0, latest version we checked.
- OpenRouter, dados públicos da API; exports de benchmarks e rankings em CC BY 4.0, latest version we checked.
Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.