Put models side by side on context, price, speed and availability.
| Model | |
|---|---|
| Overview | |
| Developer | NVIDIA |
| Context tokens | 256K |
| Max output tokens | 178K |
| Input | |
| Output | |
| Providers | 4 |
| Released | Jun 4, 2026 |
| Pricing | |
| Input per 1M | $0.60 |
| Output per 1M | $2.40 |
| Cache read per 1M | $0.12 |
| Performance | |
| Latency p50, median provider | 1.10s |
| Throughput p50, median provider | 61 tps |
| Uptime last 24h | 100.00% |