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.
GPT-5.6 SolGLM 5
| GPT-5.6 Sol OpenAI | GLM 5 Z.ai | |
|---|---|---|
| Pricing | ||
| Input $/1M | $2.00 | $0.60 |
| Output $/1M | $10.00 | $1.92 |
| Blended $/1M (3:1) | $4.00 | $0.93 |
| Cache read $/1M | $0.20 | $0.12 |
| Cache write $/1M | $2.50 | — |
| Quality indices | ||
| Intelligence Index | 47 | — |
| Coding Index | 77.4 | — |
| Math Index | — | — |
| Agentic Index | 50.2 | — |
| Speed & latency | ||
| Output speed (tok/s) | 65 | — |
| Time to first token | 64.46s | — |
| Time to first answer token | 64.46s | — |
| Benchmarks (Artificial Analysis) | ||
| GPQA Diamond | 94.1% | — |
| Humanity's Last Exam | 49.5% | — |
| SciCode | 57.1% | — |
| IFBench | 72.7% | — |
| AA-LCR (Long Context Reasoning) | 84.0% | — |
| Terminal-Bench Hard | 65.9% | — |
| Terminal-Bench 2.1 | 88.0% | — |
| τ²-Bench (Telecom) | 85.1% | — |
| τ³-Bench Banking | 44.3% | — |
| Design Arena (Elo) | ||
| Website | — | 1,260 |
| Full-stack | — | 1,132 |
| Mobile apps | — | 1,165 |
| Android native | — | 1,159 |
| UI components | — | 1,246 |
| Data viz | — | 1,243 |
| SVG | — | 1,192 |
| 3D | — | 1,257 |
| Game dev | — | 1,258 |
| Godot game dev | — | 1,144 |
| HTML slides | — | 1,150 |
| Code categories | — | 1,262 |
| ASCII art | — | 1,166 |
| Specs | ||
| Context window | 1.05M | 205K |
| Max output tokens | 128K | 128K |
| Input modalities | file, image, text | text |
| Tokenizer | GPT | Other |