FableGuide › Pricing
Claude Fable 5 costs $10 per million input tokens and $50 per million output tokens on the Claude API, twice the per-token price of Claude Opus 4.8 ($5 / $25). The 1M-token context window carries no long-context premium. The June suspension and July 1 restoration changed availability, not the API price — $10/$50 held throughout.
Full price table
| Model | Model ID | Input / MTok | Output / MTok | Context |
|---|---|---|---|---|
| Claude Fable 5 | claude-fable-5 | $10.00 | $50.00 | 1M |
| Claude Opus 4.8 | claude-opus-4-8 | $5.00 | $25.00 | 1M |
| Claude Sonnet 5 (new, Jun 30) | claude-sonnet-5 | $2.00* | $10.00* | — |
| Claude Sonnet 4.6 | claude-sonnet-4-6 | $3.00 | $15.00 | 1M |
| Claude Haiku 4.5 | claude-haiku-4-5 | $1.00 | $5.00 | 200K |
API list prices per million tokens. *Sonnet 5 launched June 30, 2026 at intro pricing of $2 / $10 through August 31, 2026, then $3 / $15. Prompt caching reads cost roughly a tenth of the input rate; cache writes cost about 1.25x the input rate. Batch processing runs at 50% of standard prices.
The return window: July 1–7 at half limits, then usage credits
Updated July 1, 2026. The original launch offer — free on paid plans June 9 through June 22 — was cut short when the June 12 export-control suspension pulled Fable 5 worldwide; subscribers got about three of the promised fourteen days. With the controls lifted June 30 and the model restored July 1, the replacement offer is this: on Pro, Max, Team, and select Enterprise plans, Fable 5 is included at 50% of weekly usage limits through July 7, 2026. From July 8, subscription use moves behind usage credits (standard Enterprise is credits-only throughout; premium Enterprise is included through July 7, then credits). None of this touches the API, which bills per token the entire time — there is no free public API tier. Access walkthroughs are in Start.
When Fable is worth it, and when it is not
Fable 5 is the most capable widely released Claude model: reach for it on the hardest reasoning and long-horizon agentic work, where its capability ceiling earns the premium. For everyday work, Opus 4.8 at half the price is usually the better value, and the new Claude Sonnet 5 — launched June 30 at $2 / $10 intro pricing, positioned as a cheaper way to run agents — is the budget alternative for routine agentic tasks. Fable also tends to be more token-efficient than prior models on hard problems, so judge by completed-task cost, not the sticker rate, but the right move is to route by task difficulty rather than default to the most expensive model. The capability comparison lives in Benchmarks.
Moral: $10 / $50 buys the ceiling, not the floor. Pay it for the hard problems; send the rest to a cheaper model. Through July 7 the return window makes trying it cheap — see Start.