Inference provider
CoreWeave: every model it hosts
10 endpoints tracked. Hosts may use different quantizations, context limits, and serving defaults, so compare price together with reliability and configuration. On this host family, the same model costs up to 19× more depending on which endpoint serves it.
Verified 2026-10-21 · source: OpenRouter public API
| Model | Input /1M | Output /1M | Cache read | Context | Uptime 1d | Throughput |
|---|---|---|---|---|---|---|
| NVIDIA: Nemotron 3.5 Lightningbf16 | $0.070 | $0.200 | $0.040 | 262K | 100.0% | — |
| DeepSeek: DeepSeek V4 Flash 0423fp8 | $0.130 | $0.280 | $0.070 | 262K | 100.0% | — |
| DeepSeek: DeepSeek V4.1 Flashfp8 | $0.150 | $0.650 | $0.0020 | 1049K | 100.0% | — |
| Z.ai: GLM 5.3 Flashnvfp4 | $0.150 | $0.500 | $0.020 | 1049K | 96.0% | — |
| NVIDIA: Nemotron 3.5 Super VLnvfp4 | $0.150 | $0.500 | $0.020 | 1049K | 92.2% | — |
| MiniMax: MiniMax M3fp4 | $0.230 | $0.960 | $0.050 | 262K | 99.9% | — |
| Qwen: Qwen3.6 35B A3Bfp8 | $0.250 | $1.25 | $0.250 | 262K | 99.9% | — |
| Z.ai: GLM 5.2fp4 | $0.620 | $2.42 | $0.140 | 1049K | 99.7% | — |
| Z.ai: GLM 5.3fp8 | $1.20 | $4.00 | $0.190 | 1049K | 96.6% | — |
| DeepSeek: DeepSeek V4 Pro 0423fp8 | $1.31 | $3.96 | $0.044 | 1049K | 100.0% | — |
Prices are list prices at build time; providers run promotions. Uptime/throughput are OpenRouter's rolling measurements. Compare hosts on the providers index.