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.
Ling 3.0 FlashGPT-5.6 Sol
| Ling 3.0 Flash Inclusionai | GPT-5.6 Sol OpenAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $0.021 | $2.00 |
| Output $/1M | $0.063 | $10.00 |
| Blended $/1M (3:1) | $0.032 | $4.00 |
| Cache read $/1M | $0.0042 | $0.20 |
| Cache write $/1M | — | $2.50 |
| Quality indices | ||
| Intelligence Index | 24.9 | 47 |
| Coding Index | 50.6 | 77.4 |
| Math Index | — | — |
| Agentic Index | — | 50.2 |
| Speed & latency | ||
| Output speed (tok/s) | 354 | 65 |
| Time to first token | 1.64s | 64.46s |
| Time to first answer token | 7.30s | 64.46s |
| Benchmarks (Artificial Analysis) | ||
| GPQA Diamond | 85.5% | 94.1% |
| Humanity's Last Exam | 23.7% | 49.5% |
| SciCode | 42.0% | 57.1% |
| IFBench | — | 72.7% |
| AA-LCR (Long Context Reasoning) | 73.0% | 84.0% |
| Terminal-Bench Hard | — | 65.9% |
| Terminal-Bench 2.1 | 55.4% | 88.0% |
| τ²-Bench (Telecom) | — | 85.1% |
| τ³-Bench Banking | 27.2% | 44.3% |
| Specs | ||
| Context window | 262K | 1.05M |
| Max output tokens | 33K | 128K |
| Input modalities | text | file, image, text |
| Tokenizer | Other | GPT |