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.
Gemini 3.5 FlashGPT-5.6 Sol
| Gemini 3.5 Flash | GPT-5.6 Sol OpenAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $1.50 | $5.00 |
| Output $/1M | $9.00 | $30.00 |
| Blended $/1M (3:1) | $3.38 | $11.25 |
| Cache read $/1M | $0.15 | $0.50 |
| Cache write $/1M | $0.083 | $6.25 |
| Quality indices | ||
| Intelligence Index | 50.2 | 58.9 |
| Coding Index | 70.1 | 77.4 |
| Math Index | — | — |
| Agentic Index | 37.4 | 54 |
| Speed & latency | ||
| Output speed (tok/s) | 283 | 67 |
| Time to first token | 11.30s | 120.38s |
| Time to first answer token | 11.30s | 120.38s |
| Benchmarks (Artificial Analysis) | ||
| GPQA Diamond | 92.2% | 94.1% |
| Humanity's Last Exam | 41.0% | 47.2% |
| SciCode | 53.1% | 56.1% |
| IFBench | 76.3% | 72.7% |
| AA-LCR (Long Context Reasoning) | 69.3% | 73.7% |
| Terminal-Bench Hard | 40.9% | 65.9% |
| Terminal-Bench 2.1 | 78.7% | 88.0% |
| τ²-Bench (Telecom) | 95.3% | 85.1% |
| τ³-Bench Banking | 25.4% | 33.0% |
| Design Arena (Elo) | ||
| Website | 1,283 | — |
| Web apps | 1,254 | — |
| Full-stack | 1,253 | — |
| Mobile apps | 1,250 | — |
| Android native | 1,224 | — |
| UI components | 1,304 | — |
| Data viz | 1,261 | — |
| SVG | 1,295 | — |
| 3D | 1,291 | — |
| Game dev | 1,319 | — |
| Agentic game dev | 1,190 | — |
| Godot game dev | 1,171 | — |
| HTML slides | 1,170 | — |
| PPTX slides | 1,244 | — |
| Python→PPTX slides | 1,247 | — |
| Agentic slides | 1,244 | — |
| Agentic HTML slides | 1,162 | — |
| Agentic slides (HTML) | 1,162 | — |
| Agentic slides (Python PPTX) | 1,242 | — |
| Code categories | 1,289 | — |
| ASCII art | 1,305 | — |
| Specs | ||
| Context window | 1.05M | 1.05M |
| Max output tokens | 66K | 128K |
| Input modalities | text, image, video, file, audio | file, image, text |
| Tokenizer | Gemini | GPT |