Claude Fable 5.1: API price, limits and plan access

Claude Fable 5.1 at a glance

Claude Fable 5.1 is the newer of Anthropic's two Fable models, and Anthropic recommends it for demanding reasoning and long-horizon agentic work, or when your evals on Opus 5.5 at higher effort still fall short. Here are the specs for claude-fable-5-1.

Fable 5.1 vs Fable 5

The table below compares claude-fable-5-1 and claude-fable-5 across release dates, context windows and pricing.

Both models are active right now. Fable 5.1 keeps the same base price as Fable 5, but it cuts the cost of cache reads by 75%, down from $1 to $0.25 per million tokens. If you use Claude Code, you'll need version 2.1.255 or later to run Fable 5.1 (2.1.170 or later for Fable 5).

What Fable costs on the API

A single request to claude-fable-5-1 with 100,000 input tokens and 10,000 output tokens costs $1.00 for input (100,000 x $10 / 1M) plus $0.50 for output (10,000 x $50 / 1M), coming to $1.50 in total, or $0.75 if you send it through the Batch API at $5 and $25 per million tokens. For workloads that pass the same large prompt on every turn, repeated context read from the cache costs $0.25 per million tokens instead of the $10 base input rate (5-minute cache writes are $12.50 per million tokens, and 1-hour writes are $20).

Anthropic says Claude Opus 5.5 performs at the level of Fable 5.1 on most work while running at $4 per million input tokens and $20 per million output. That same 100,000-input, 10,000-output request on Opus 5.5 costs $0.40 for input plus $0.20 for output, or $0.60 in total; see the full Claude API pricing breakdown for the rest of the lineup.

Fable on Claude Pro and Max

How you pay for Fable in the Claude apps depends on whether you're on Claude Pro or Max, and the table below breaks down how each subscription tier handles Fable access and Claude usage limits.

Safeguards and where Fable runs

• Anthropic's safeguards automatically reroute many flagged cyber and biology queries to less capable models, and when a request gets rerouted you aren't billed at Fable prices.
• Using Fable requires 30-day data retention by default.
• Fable 5.1 runs on the Claude API, Amazon Web Services, Google Cloud and Microsoft Foundry, and inside the Claude apps on Pro, Max, Team and Enterprise plans (it's not available on the Free plan).

Call Fable from your Claude plan in code

Plan2API connects your existing Claude Pro or Max subscription and exposes claude/claude-fable-5-1 and claude/claude-fable-5 (both with a 1M-token context window and up to 128K output) through an OpenAI-compatible Chat Completions API at https://api.plan2api.com/v1. You can also use the anthropic/ prefix or the dotted model name claude-fable-5.1. The curl example that follows calls POST /v1/chat/completions with claude/claude-fable-5-1.

Requests run directly on the connected Claude account's own plan, which means Anthropic's Fable rules for that plan still apply (up to 50% of the weekly limit on Max, or usage credits from the start on Pro). Connecting your account is free to start: the Plan2API Free tier ($0, no card needed) covers 1 connected subscription and 50 chat requests a day, and Pro is $12 a month (or $99 a year) with 3 subscriptions and no daily request cap.

curl https://api.plan2api.com/v1/chat/completions \
  -H "Authorization: Bearer $PLAN2API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude/claude-fable-5-1",
    "reasoning_effort": "high",
    "messages": [
      {"role": "user", "content": "Audit this migration plan and list the risks in order."}
    ]
  }'