Provider price table
Cheapest first, our own row marked. Show the rows we cannot route and label them clearly rather than omitting them.
| Provider | Input $/1M | Output $/1M | Cache read $/1M | Notes |
|---|---|---|---|---|
| Anthropic | $10.00 | $50.00 | $0.25 | List price; claude-fable-5-1 model ID |
| AWS Bedrock | $10.00 | $50.00 | $0.25 | Identical list rates; IAM auth |
| Microsoft Foundry | $10.00 | $50.00 | $0.25 | Azure regions; reserved capacity available |
| Google Vertex | $11.00 | $55.00 | $0.275 | +10% US data-residency multiplier |
| OneInfer | $10.00 | $50.00 | $0.25 | OpenAI-compatible endpoint; free credits available |
What does not change between providers
Cache read at $0.25, cache write at $12.50 (five-minute) and $20.00 (one-hour), and batch rates at $5.00/$25.00 are identical to Anthropic's published schedule. The 1.1x multiplier on Google Vertex applies to input and output, not to cache. Cloud regions and quota differ; the routing decision is about latency, regional availability, and commitments rather than headline price.
Ready to test the workflow?
Create account & add creditsRouting notes
Use OneInfer to access Claude Fable 5.1 without an Anthropic, AWS or Google account, on free credits to start. Use Anthropic direct for the lowest latency in the US east region. Use AWS Bedrock when you have committed-use discounts on EC2 or Bedrock. Use Google Vertex when you need the 1.1x data-residency multiplier for US-only inference.
Frequently asked questions
Which providers serve Claude Fable 5.1?
Claude Fable 5.1 is served through the Claude API, AWS Bedrock, Google Cloud Vertex, Microsoft Foundry, and routed platforms including OneInfer. List price is identical across Anthropic, AWS Bedrock, Azure and OpenRouter at $10 and $50 per 1M tokens; Google's rows carry an 11% premium, at $11.00 and $55.00, reflecting the data-residency multiplier.
Put Claude Fable 5.1 to work
Fund a controlled evaluation, start with a prepared prompt, and measure quality and cost on your own workload.