Compare models
Pick up to 4 models for a side-by-side spec sheet. Hover any dotted metric name for what the test measures. Best value in each row is highlighted.
DeepSeek V4 FlashGPT-5.6 Sol
| DeepSeek V4 Flash DeepSeek | GPT-5.6 Sol OpenAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $0.094 | $5.00 |
| Output $/1M | $0.188 | $30.00 |
| Blended $/1M (3:1) | $0.117 | $11.25 |
| Cache read $/1M | $0.019 | $0.50 |
| Cache write $/1M | — | $6.25 |
| Quality indices | ||
| Intelligence Index | 40.3 | 58.9 |
| Coding Index | 56.2 | 77.4 |
| Math Index | — | — |
| Agentic Index | 31.1 | 54 |
| Speed & latency | ||
| Output speed (tok/s) | 115 | 70 |
| Time to first token | 0.86s | 112.90s |
| Time to first answer token | 49.64s | 112.90s |
| Benchmarks (Artificial Analysis) | ||
| GPQA Diamond | 89.4% | 94.1% |
| Humanity's Last Exam | 32.1% | 47.2% |
| SciCode | 44.9% | 56.1% |
| IFBench | 79.2% | 72.7% |
| AA-LCR (Long Context Reasoning) | 63.0% | 73.7% |
| Terminal-Bench Hard | 35.6% | 65.9% |
| Terminal-Bench 2.1 | 61.8% | 88.0% |
| τ²-Bench (Telecom) | 95.0% | 85.1% |
| τ³-Bench Banking | 22.9% | 33.0% |
| Design Arena (Elo) | ||
| Website | 1,235 | — |
| UI components | 1,205 | — |
| Data viz | 1,159 | — |
| SVG | 1,202 | — |
| 3D | 1,248 | — |
| Game dev | 1,254 | — |
| Code categories | 1,239 | — |
| ASCII art | 1,157 | — |
| Specs | ||
| Context window | 1.05M | 1.05M |
| Max output tokens | — | 128K |
| Input modalities | text | file, image, text |
| Tokenizer | DeepSeek | GPT |