Models
Hugging Face repo IDs on the chat endpoint, with context length, quantization, and price per million tokens.
| Model | Context | Quant | Input | Output | Cache | tok/s |
|---|---|---|---|---|---|---|
| 262128 | FP8 | $0.10 | $1.60 | $0.050 | 46 | |
| 1048576 | NVFP4 | $0.15$0.12 | $0.50$0.30 | $0.030 | 104 | |
| 1048576 | NVFP4 | $0.20 | $0.50 | $0.050 | 240 | |
99.73%
Uptime, last 7 days