Token counting
When the server proxy is configured, the counter calls Anthropic count_tokens with model: "claude-fable-5". If the API is unavailable, the UI keeps working in estimate mode and labels the result.
A compact reference for the Fable 5 cost and access details that affect token counting and launch planning.
| Claude API model ID | claude-fable-5 |
|---|---|
| Amazon Bedrock ID | anthropic.claude-fable-5 |
| Vertex AI ID | claude-fable-5 |
| Knowledge cutoff | January 2026 |
| Prompt caching | 5m write, 1h write, and cache-read pricing supported |
Subscription access is expected to shift toward usage credits. Treat that as a product-access question, not an API invoice calculation.
Anthropic says safeguard-flagged requests can be routed away from Fable 5 and are not charged at Fable prices. The calculator keeps fallback out of the cost form so this page can be updated if Anthropic wording changes.
model: "claude-fable-5" for model-specific token counting.When the server proxy is configured, the counter calls Anthropic count_tokens with model: "claude-fable-5". If the API is unavailable, the UI keeps working in estimate mode and labels the result.
Regular input, cache writes, cache reads, and output are priced separately. Batch and US-geo modifiers are applied after the base cost, matching the current MVP contract.
The MVP proxy does not store pasted text. It only needs request status, rough character bucket, and timing for abuse and reliability checks.
Pricing changes should be fixed in PRICING_CONFIG, rebuilt, and noted with a new last-verified date.
Pricing constants are isolated in one config file and should be rechecked before launch.
Estimates are for planning. Check your Anthropic invoice and current pricing before committing spend.