AI models
DeepSeek V4 Pro vs GPT-6.1 Sol
Same criteria, side by side. Where the gap is small, it may be measurement noise.
Also searched as
DeepSeek V4 vs GPT-6
Direct answer: GPQA Diamond: DeepSeek V4 Pro 91.7%, GPT-6.1 Sol 95.4% (Epoch AI). OTIS Mock AIME: DeepSeek V4 Pro 98.6%, GPT-6.1 Sol 100% (Epoch AI). FrontierMath (tiers 1 to 3): DeepSeek V4 Pro 64.6%, GPT-6.1 Sol 93.7% (Epoch AI). SimpleQA Verified: DeepSeek V4 Pro 52.9%, GPT-6.1 Sol 73.9% (Epoch AI). GPQA Diamond (OpenRouter): DeepSeek V4 Pro 91.9%, GPT-6.1 Sol 94.8% (OpenRouter Benchmarks). τ-bench Verified (airline): DeepSeek V4 Pro 78.6%, GPT-6.1 Sol 70.7% (OpenRouter Benchmarks).
| Metric | DeepSeek V4 Pro | GPT-6.1 Sol | Source |
|---|---|---|---|
| Weights ?what it is | open | closed | no data |
| Input US$/1M ? | US$ 0.96 | US$ 2.00 | no data |
| Output US$/1M ? | US$ 1.91 | US$ 10.00 | no data |
| Context ? | 1,048,576 | 1,050,000 | no data |
| GPQA Diamond ? | 91.7% | 95.4% | Epoch AI |
| OTIS Mock AIME ? | 98.6% | 100% | Epoch AI |
| FrontierMath (tiers 1 to 3) ? | 64.6% | 93.7% | Epoch AI |
| SimpleQA Verified ? | 52.9% | 73.9% | Epoch AI |
| SWE-bench Verified ? | 77.6% | no data | Epoch AI |
| GPQA Diamond (OpenRouter) ? | 91.9% | 94.8% | OpenRouter Benchmarks |
| τ-bench Verified (airline) ? | 78.6% | 70.7% | 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.