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.
Estimate what Claude Fable 5 actually costs with caching, batch, and subscription credits included.
Default result is server-rendered: $20.00 for 1M input + 200K output.
Fable 5 uses a higher base price than Opus 4.8. This page does not claim a fixed task-level multiplier because output length, cache use, and routing decide the bill.
Advanced caching splits regular input, 5-minute writes, 1-hour writes, cache reads, and output. Batch and US-geo modifiers are then applied to the calculated base.
The credit estimate is a planning conversion. It is not an API invoice and it should not be used to reconcile your bill.
Fallback billing is intentionally not a calculator toggle. See Fable 5 specs and fallback notes.
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.
Yes. Simple mode maps cache hit percentage to cache-read tokens; Advanced mode exposes the actual billing buckets.
No. Fallback behavior belongs on the model page because the billing path depends on current Anthropic behavior and configuration.