Inference provider
DeepInfra: every model it hosts
27 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 48× 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 |
|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V4 Flash 0423fp8 | $0.060 | $0.180 | $0.015 | 1049K | 100.0% | — |
| NVIDIA: Nemotron 3.5 Lightningbf16 | $0.060 | $0.160 | $0.030 | 262K | 99.9% | — |
| DeepSeek: DeepSeek V4.1 Flashfp8 | $0.078 | $0.235 | $0.0024 | 1049K | 94.6% | — |
| NVIDIA: Nemotron 3 Superbf16 | $0.085 | $0.400 | — | 262K | 99.9% | — |
| DeepSeek: DeepSeek V4 Flash 0423fp8 | $0.090 | $0.180 | $0.018 | 1049K | 100.0% | — |
| DeepSeek: DeepSeek V4.1 Flashfp8 | $0.098 | $0.294 | $0.0029 | 1049K | 100.0% | — |
| Qwen: Qwen3.6 35B A3Bfp8 | $0.100 | $0.950 | $0.100 | 262K | 98.3% | — |
| DeepSeek: DeepSeek V4.1 Flashfp8 | $0.112 | $0.336 | $0.0034 | 1049K | 94.1% | — |
| Xiaomi: MiMo-V2.6-Flashfp8 | $0.133 | $0.266 | $0.0027 | 1049K | 99.9% | — |
| DeepSeek: DeepSeek V4.1 Flashfp8 | $0.140 | $0.420 | $0.0042 | 1049K | 100.0% | — |
| DeepSeek: DeepSeek V4.1 Flashfp8 | $0.147 | $0.441 | $0.0044 | 1049K | 100.0% | — |
| Z.ai: GLM 5.3 Flashfp4 | $0.150 | $0.500 | $0.030 | 1049K | 96.9% | — |
| NVIDIA: Nemotron 3.5 Super VLfp4 | $0.150 | $0.500 | $0.020 | 262K | 99.9% | — |
| DeepSeek: DeepSeek V4.1 Flashfp8 | $0.210 | $0.630 | $0.0063 | 1049K | 100.0% | — |
| MiniMax: MiniMax M3fp8 | $0.224 | $0.880 | $0.045 | 524K | 71.4% | — |
| MiniMax: MiniMax M3fp8 | $0.280 | $1.10 | $0.056 | 524K | 99.6% | — |
| Xiaomi: MiMo-V2.6-Profp8 | $0.387 | $0.783 | $0.0032 | 1049K | 100.0% | — |
| NVIDIA: Nemotron 3 Ultrafp4 | $0.400 | $1.76 | $0.080 | 262K | 100.0% | — |
| Z.ai: GLM 5.2fp4 | $0.450 | $1.44 | $0.084 | 1049K | 98.7% | — |
| NVIDIA: Nemotron 3 Ultrafp4 | $0.500 | $2.20 | $0.100 | 262K | 99.9% | — |
| Z.ai: GLM 5.3fp4 | $0.563 | $2.50 | $0.125 | 1049K | 97.5% | — |
| Z.ai: GLM 5.2fp4 | $0.563 | $1.80 | $0.105 | 1049K | 99.9% | — |
| Tencent: Hy4 previewfp4 | $0.834 | $2.50 | $0.042 | 1049K | 98.5% | — |
| StepFun: Step 5 Previewfp8 | $1.00 | $2.70 | $0.100 | 1049K | 99.2% | — |
| DeepSeek: DeepSeek V4 Pro 0423fp8 | $1.30 | $2.60 | $0.100 | 1049K | 100.0% | — |
| DeepSeek: DeepSeek V4 Pro 0423fp8 | $1.30 | $2.60 | $0.100 | 1049K | 100.0% | — |
| MoonshotAI: Kimi K3mxfp4 | $2.85 | $14.25 | $0.285 | 1049K | 99.3% | — |
Prices are list prices at build time; providers run promotions. Uptime/throughput are OpenRouter's rolling measurements. Compare hosts on the providers index.