elevated Claude Fable 5 2 · Research

Claude Fable 5.2 Costs 20% Less at $8/$40, but Which Plans Actually Include It?

Data graphic: Claude Fable 5.2 costs 20% less. The Fable 5.1 price of $10/$50 per million tokens is struck through above the new $8/$40 input/output price; Max and Ultra get it up to 50% of the weekly limit, and Pro and standard Team pay usage credits after Nov 16.
AK

Threat intelligence editor · Published Nov 16, 2026, 11:31 AM EST

Anthropic cut Claude Fable 5.2 API prices to $8/$40 per million tokens. Max and Ultra include it up to 50% of weekly limits; Pro moves to usage credits.

Anthropic released Claude Fable 5.2 on November 9, 2026, at $8 per million input tokens and $40 per million output tokens, down from $10/$50 for Fable 5.1: a 20% cut on both. The model ID is claude-fable-5-2. On a subscription the picture is less simple. Max and Ultra include Fable up to 50% of the weekly usage limit, while Pro and standard Team seats pay usage credits for it once the launch week ends on November 16, 2026 at 11:59 pm PT.

What to do now

  • Pro and standard Team users: the launch week that includes Fable in your plan ends November 16, 2026 at 11:59 pm PT. After that, Fable usage runs on usage credits, if you have them enabled.
  • Max, Ultra and premium seat holders: check how many launch resets you have under Settings > Usage. They expire March 8, 2027 at 11:59 pm PT.
  • Enterprise admins: decide whether to turn Fable 5.2 on, or opt out before it is added automatically to roles that already have a Fable model enabled.
  • API teams: the price cut is real, but Anthropic itself recommends Opus 5.5 as the daily driver and Fable for the most ambitious work. Test both on your own evals.
  • Security teams: expect Fable 5.2 to refuse most offensive work and hand it to an older model.

Which plans include Fable 5.2

Fable 5.2 is available on Pro, Max, Ultra, Team and Enterprise, but it is not included the same way on each.

Data graphic: who gets Fable 5.2. Max, Ultra and premium seats run it on plan limits up to 50% of the weekly limit, then credits or another model; Pro and standard Team have it included through Nov 16, then pay usage credits; standard Enterprise seats use credits only if enabled, and it is off until an admin opts in. No one-time credit.

How each Claude plan pays for Fable 5.2 once the Pro launch week ends on November 16, 2026. Source: Anthropic help centre.

  • Max, Ultra, premium Team seats and premium seat-based Enterprise seats: Fable is part of the plan, up to 50% of weekly usage limits at no extra cost. Past that cap you continue on usage credits or switch to another model.
  • Pro and standard Team seats: outside the launch offer, Fable is not covered by plan limits and runs on usage credits from the start.
  • Standard seat-based Enterprise: usage credits only, and only if the organization has enabled them. Fable 5.2 is off by default until an admin turns it on. Seven days after launch it is added to roles that already have a Fable model enabled, unless an admin opts out.

There is no one-time credit this time. Eligible Pro and standard Team seats got one when Fable 5 moved to usage credits; Anthropic says none applies to Fable 5.2 or 5.1. For how the Max plan changed when Fable 5 became a permanent part of it, see our earlier look at Claude Max.

The launch week and the resets

Anthropic's plan page puts Pro on usage credits; its launch-offer page includes Pro for one week. Both are true: the launch offer is a temporary exception. Pro and standard Team seats get all Fable models included, up to 50% of weekly limits, through November 16, 2026 at 11:59 pm PT, plus one reset. After that, Fable usage goes to usage credits if they are enabled.

Max, Ultra, premium Team and premium legacy seat-based Enterprise seats get launch resets based on how long the account has paid, counted from the first paid bill:

  • Under six months: one reset
  • Six to 12 months: two resets
  • Over 12 months: three resets

These resets expire March 8, 2027 at 11:59 pm PT. A reset restores Fable usage on the weekly limit and refills the five-hour window. It does not raise limits and has no cash value. Use it under Settings > Usage, "Reset for free".

To qualify, paid billing must have been in place before November 9, 2026 at 10:00 am PT. Free, standard legacy seat Enterprise, self-serve Enterprise, usage-based Enterprise and HIPAA-ready Enterprise plans are excluded. Our Claude Code Max limits explainer covers how the weekly and five-hour windows work.

API details

  • Standard: $8/MTok input, $40/MTok output.
  • Fast mode: $16/MTok input, $80/MTok output, for up to 2.5x faster generation. It runs in Claude Code, the Claude Platform, Claude Models on Google Cloud and Microsoft Foundry (Anthropic hosted).
  • Context: 1M tokens, with 128K max output and default effort high.
  • Discounts: prompt cache reads cost 2.5% of the base input price, and the Batch API is 50% off.
  • Lifetime: Anthropic says the model will not be retired sooner than November 9, 2027.

Safeguards and Mythos 5.2

Fable 5.2 keeps its cyber and biology safeguards. Anthropic says it blocks most cybersecurity tasks, including penetration testing, exploit generation and binary-based vulnerability work. Blocked queries are answered by Opus 4.8 (cyber) or Opus 5 (biology). Anthropic says blocks are down about 55% compared with Fable 5.1. The same Opus 4.8 fallback applies to Opus 5.5's cyber safeguards.

Mythos 5.2 is the same underlying model without certain cyber and biology safeguards, at the same $8/$40, and only for vetted organizations.

Benchmarks (vendor-reported)

All figures below come from Anthropic's launch post and have not been independently reproduced.

  • FrontierCode 2.0 (Xhigh effort): Fable 5.2 49.1%, Opus 5.5 45.5%
  • FrontierSWE v2: Fable 5.2 69.9%, Opus 5.5 62.3%
  • OSWorld 2.1: Fable 5.2 89.3%, Opus 5.5 87.2%
  • CVE Rediscovery: Mythos 5.2 89.6%

Sources

Keep reading

All latest →
  1. watchResearchClaude Max 20x Is Now Ultra: No Five-Hour Limit, if Your Account Is Verified4 min
  2. watchResearchEdge0 Claims Its 145B MoE Runs on One 24 GB GPU at 79.6 Tokens a Second5 min
  3. watchResearchCopilot Fusion Goes GA: Per-Model Billing, 30% Off, and How It Compares With Claude Code Max4 min
  4. stableResearchRipple-65B Preview: A Dense 65B Open Model That Fits One DGX Spark and Claims 80.5 on Terminal-Bench 2.17 min
  5. elevatedResearchGemini 2.5 Pro and 2.5 Flash API Calls Reroute on 1 December: What Changes in Your Bill and Your Code4 min
  6. watchResearchGitHub Enterprise Now Includes Copilot at $0, but Chat, Code Review and the Cloud Agent Bill From the First Request3 min