Inference provider
Claude Platform on AWS: every model it hosts
4 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 50× 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 |
|---|---|---|---|---|---|---|
| Anthropic: Claude Haiku 5.5unknown | $0.100 | $0.500 | $0.010 | 1000K | 100.0% | — |
| Anthropic: Claude Sonnet 5.5unknown | $2.00 | $10.00 | $0.100 | 1000K | 100.0% | — |
| Anthropic: Claude Opus 5.5unknown | $4.00 | $20.00 | $0.200 | 1000K | 100.0% | — |
| Anthropic: Claude Opus 5unknown | $5.00 | $25.00 | $0.500 | 1000K | 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.