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.
Nemotron 3.5 Lightning (free)GPT-5.6 Sol
| Nemotron 3.5 Lightning (free) NVIDIA | GPT-5.6 Sol OpenAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | Free | $2.00 |
| Output $/1M | Free | $10.00 |
| Blended $/1M (3:1) | Free | $4.00 |
| Cache read $/1M | — | $0.20 |
| Cache write $/1M | — | $2.50 |
| Quality indices | ||
| Intelligence Index | 12.9 | 47 |
| Coding Index | 26.8 | 77.4 |
| Math Index | — | — |
| Agentic Index | 3.5 | 50.2 |
| Speed & latency | ||
| Output speed (tok/s) | 290 | 65 |
| Time to first token | 0.46s | 64.46s |
| Time to first answer token | 7.35s | 64.46s |
| Benchmarks (Artificial Analysis) | ||
| GPQA Diamond | 74.3% | 94.1% |
| Humanity's Last Exam | 10.6% | 49.5% |
| SciCode | 32.1% | 57.1% |
| IFBench | — | 72.7% |
| AA-LCR (Long Context Reasoning) | 60.3% | 84.0% |
| Terminal-Bench Hard | — | 65.9% |
| Terminal-Bench 2.1 | 24.3% | 88.0% |
| τ²-Bench (Telecom) | — | 85.1% |
| τ³-Bench Banking | 8.9% | 44.3% |
| Specs | ||
| Context window | 1M | 1.05M |
| Max output tokens | 66K | 128K |
| Input modalities | text | file, image, text |
| Tokenizer | Other | GPT |