Inference provider
Wafer: every model it hosts
12 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 21× 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 |
|---|---|---|---|---|---|---|
| Z.ai: GLM 5.2fp4 | $0.034 | $7.00 | $0.033 | 1049K | 99.4% | — |
| Z.ai: GLM 5.3 Flashnvfp4 | $0.074 | $0.200 | $0.060 | 1049K | 97.4% | — |
| Z.ai: GLM 5.3fp4 | $0.079 | $1.40 | $0.078 | 1049K | 99.5% | — |
| DeepSeek: DeepSeek V4 Flash 0423nvfp4 | $0.140 | $0.230 | $0.135 | 1049K | 99.4% | — |
| DeepSeek: DeepSeek V4.1 Flashmxfp4 | $0.160 | $0.500 | $0.0030 | 1049K | 99.6% | — |
| Z.ai: GLM 5.3fp4 | $0.179 | $1.40 | $0.178 | 1049K | 99.2% | — |
| DeepSeek: DeepSeek V4.1 Flashmxfp4 | $0.190 | $0.650 | $0.0060 | 1049K | 99.4% | — |
| Z.ai: GLM 5.2fp4 | $0.190 | $8.00 | $0.180 | 1049K | 99.2% | — |
| MoonshotAI: Kimi K3mxfp4 | $0.238 | $24.00 | $0.228 | 1049K | 99.3% | — |
| DeepSeek: DeepSeek V4 Pro 0423mxfp4 | $0.250 | $5.00 | $0.249 | 1049K | 99.4% | — |
| MoonshotAI: Kimi K3mxfp4 | $0.360 | $11.90 | $0.359 | 1049K | 99.3% | — |
| DeepSeek: DeepSeek V4 Flash 0423mxfp4 | $0.700 | $0.170 | $0.013 | 1049K | 99.4% | — |
Prices are list prices at build time; providers run promotions. Uptime/throughput are OpenRouter's rolling measurements. Compare hosts on the providers index.