| Model | Endpoint | Context | Input /M | Output /M | Cached input /M | Cache write /M | Uptime 1d |
|---|---|---|---|---|---|---|---|
 Beam 1.0 501B A23B | nebius/fp4 | 1M | $0.25 | $1.00 | $0.05 | — | 100.00% |
 DeepSeek V4 Pro 0813 | nebius/fp8 | 1M | $1.32 | $3.96 | $0.04 | — | 99.79% |
| DeepSeek V4.1 Flash 1017 | nebius/fp8 | 1M | $0.30 | $1.20 | $0.006 | — | 100.00% |
 GLM 5.2 | nebius/fp4 | 1M | $1.40 | $4.40 | $0.14 | — | 76.88% |
| GLM 5.3 | nebius/fp4 | 1M | $1.40 | $4.40 | $0.26 | — | 98.67% |
| GLM 5.3 Flash | nebius/fp4 | 1M | $0.15 | $0.50 | $0.03 | — | 99.56% |
 Gemma 3 27B | nebius/fp8 | 110K | $0.10 | $0.30 | $0.04 | — | 94.06% |
 Hermes 4 405B | nebius/fp8 | 131K | $1.00 | $3.00 | $0.20 | — | 100.00% |
 Kimi K3.1 Pro | nebius/fp4 | 1M | $2.00 | $10.00 | $0.20 | — | 95.39% |
 MiniMax M3.1 Flash | nebius/fp4 | 1M | $0.15 | $0.60 | $0.015 | — | 100.00% |
 Muse Spark Open | nebius/fp4 | 1M | $1.25 | $4.25 | $0.15 | — | 99.61% |
 Nemotron 3 Nano 30B A3B | nebius/fp8 | 262K | $0.06 | $0.24 | $0.012 | — | 99.93% |
| Nemotron 3.5 Super VL | nebius/bf16 | 1M | $0.30 | $0.90 | $0.06 | — | 93.54% |
 Qwen3 235B A22B Instruct 2507 | nebius/fp8 | 262K | $0.20 | $0.60 | $0.05 | — | 87.71% |
| Qwen3 30B A3B Instruct 2507 | nebius/fp8 | 262K | $0.10 | $0.30 | $0.10 | — | 97.08% |
| Qwen3 Embedding 8B | nebius/bf16 | 41K | $0.01 | $0.00 | — | — | 100.00% |
| Qwen3.8 27B | nebius/fp4 | 1M | $0.45 | $3.00 | $0.10 | — | 99.37% |
 gpt-oss-120b | nebius/fp4 | 131K | $0.15 | $0.60 | $0.10 | — | 99.93% |