Claude models
Which Claude models are current and how they differ
In short
Claude Fable 5.2 is on all paid plans, the Claude Platform and major clouds at $8/$40 per million tokens, down from $10/$50. Opus 5.5 costs $4/$20, with cache reads at $0.20 per million tokens. Sonnet 5.5 costs $2/$10, and Haiku 5.5 costs $0.10/$0.50 on prompts under 100k. Mythos 5.2 is the same underlying model as Fable, but only organizations verified through Anthropic’s Life Sciences and Cyber Verification Programs can get it (source).
The latest change is Fable 5.2. According to Anthropic, its output runs up to 30% faster than Fable 5.1. In Anthropic’s early FrontierCode 2.0 run, Fable 5.2 at High effort scored about level with Opus 5.5 at Max for roughly half the cost per task. Anthropic still recommends starting most coding and knowledge work on Opus 5.5. Claude Code 2.1.321 makes Fable 5.2 the default Fable model with 1M context. Its cache reads cost $0.20/Mtok, against $0.25 for Fable 5.1. That release also fixes an MCP OAuth redirect that sent secrets to another host.
Earlier, Anthropic halved Sonnet 5.5 cache reads from $0.20 to $0.10 per million tokens. The safeguards affect which model actually answers. On Opus 5.5, most cybersecurity tasks get rerouted to Opus 4.8. On Sonnet 5.5, higher-risk security tasks fall back to Sonnet 5. Fable 5.5 and Mythos 5.5 have appeared only in leaks and a hidden page string, and Anthropic has not confirmed either.
At a glance
| Model | ID | Status | As of | Source |
|---|---|---|---|---|
| Claude Fable 5.2 | Available on all paid plans, the Claude Platform and major clouds; $8/$40 per million tokens, down from $10/$50; output up to 30% faster than Fable 5.1 | 10 November 2026 | Read | |
| Claude Opus 5.5 | Live on all platforms, including AWS, Google Cloud and Azure; $4/$20 per million tokens, cache reads $0.20 per million tokens; most cybersecurity tasks rerouted to Opus 4.8 | 22 September 2026 | Read | |
| Claude Sonnet 5.5 | Released; $2/$10 per million tokens; first Sonnet with cyber safeguards, higher-risk security tasks fall back to Sonnet 5 | 28 September 2026 | Read | |
| Claude Haiku 5.5 | Released; $0.10/$0.50 per million tokens on prompts under 100k, smaller cut on prompts over 100k; first Haiku with an effort setting | 7 October 2026 | Read | |
| Claude Mythos 5.2 | Same underlying model as Fable; only for organizations verified through Anthropic's Life Sciences and Cyber Verification Programs | 10 November 2026 | Read | |
| Claude Fable 5.1 | claude-fable-5-1 | 1M context at $10/$50 per Mtok, $0.25/Mtok cache reads | 2 September 2026 | Read |
| Claude Mythos 5.1 | Same model as Fable 5.1 with more permissive safeguards; open only to vetted cyberdefenders and life scientists through two trusted access programs, currently a set of US organizations | 1 September 2026 | Read | |
| Claude Opus 5 | claude-opus-5 | Available across all platforms at $5/$25 per million tokens; Fast mode about 2.5x faster at double the price | 24 July 2026 | Read |
| Claude Sonnet 5 | Generally available via the Claude interface and API | 1 July 2026 | Read | |
| Claude Opus 4.8 | Released; Fast mode 2.5x faster than standard at 3x less | 28 May 2026 | Read |
Milestones
- Anthropic releases Claude Opus 4.8, with a Fast mode that runs 2.5x faster than standard and costs 3x less.
- Claude Fable 5 and Mythos 5 launch at $10/$50 per million tokens, with Mythos 5 limited to Project Glasswing participants.
- Anthropic temporarily disables Fable 5 and Mythos 5 for all users after a US Commerce Department letter demanding a government license.
- Sonnet 5 becomes generally available alongside its System Card.
- Claude Opus 5 launches at $5/$25 per million tokens and becomes the default on Claude Max.
- Anthropic retunes Fable 5’s biology safeguards, and in its own testing refusals on biological queries fall roughly 85%.
- A statistical text watermark goes live on Claude models released after August 2, 2026.
- Claude Fable 5.1 ships with cache reads cut 75% to $0.25 per million tokens, and Mythos 5.1 goes to vetted cyberdefenders and life scientists.
- Claude Opus 5.5 launches at $4/$20 per million tokens, with cache reads at $0.20.
- Claude Code 2.1.280 makes Opus 5.5 the default model on Pro and Team Standard plans.
- Claude Sonnet 5.5 launches at Sonnet 5’s price of $2/$10 per million tokens and is the first Sonnet with cyber safeguards.
- Haiku 5.5 launches at $0.10/$0.50 per million tokens on prompts under 100k, and Sonnet 5.5 cache reads are halved to $0.10.
- Claude Code 2.1.321 makes Fable 5.2 the default Fable model with 1M context.
- Claude Fable 5.2 launches at $8/$40 per million tokens, down from $10/$50, alongside Mythos 5.2 for verified organizations.
FAQ
What is the current Claude Fable model and how much does it cost?
The current Fable model is Claude Fable 5.2. It is on all paid plans, the Claude Platform and major clouds at $8/$40 per million tokens, down from $10/$50. In Claude Code it has 1M context, and its cache reads cost $0.20/Mtok, against $0.25 for Fable 5.1. To switch back to Fable 5.1 in Claude Code, run /model claude-fable-5-1.
Should I use Claude Opus 5.5 or Fable 5.2 for coding?
Anthropic still recommends starting most coding and knowledge work on Opus 5.5. In Anthropic’s own early FrontierCode 2.0 run, Fable 5.2 at High effort scored about level with Opus 5.5 at Max for roughly half the cost per task.
On Terminal-Bench 4.0, Sonnet 5.5 scores 70.6% and Opus 5.5 scores 66.4%. Anthropic says Opus 5.5 remains clearly stronger at complex, open-ended work.
What is the difference between Claude Fable and Claude Mythos?
Mythos 5.2 is the same underlying model as Fable. The difference is the safeguards: Fable has strong safeguards against misuse. Mythos goes only to organizations verified through Anthropic’s Life Sciences and Cyber Verification Programs.
Why do Claude Opus 5.5 and Sonnet 5.5 hand security tasks to older models?
Both models have safeguards that send this work elsewhere. On Opus 5.5, most cybersecurity tasks get rerouted to Opus 4.8. Sonnet 5.5 is the first Sonnet with cyber safeguards, and its higher-risk security tasks visibly fall back to Sonnet 5.
Has Claude Fable 5.5 or Mythos 5.5 been released?
Anthropic has confirmed neither. A LiteLLM entry for Fable 5.5 was removed after about a day. A hidden mention of Mythos 5.5 on Anthropic’s Mythos page came from one page string spotted by one account. The poster labeled the idea of a release as speculation.
Latest
- Claude Fable 5.2 drops to $8/$40 and runs up to 30% faster
- Claude Code makes Fable 5.2 default at $8/$40 per Mtok
- Hidden Claude Mythos 5.5 line found on Anthropic’s site
- Over 80% of Anthropic’s Q3 revenue came from new models
- Kimi K3.1 Pro lands just behind Opus 5.5 on Terminal-Bench
- Leakers say partners already have Claude Fable 5.5
- Claude Fable 5.5 surfaces in LiteLLM’s Azure model list
- Opus 5.5 traffic may be hitting a new Claude checkpoint
- Haiku 5.5 ranks #2 of 182 and burns 440M tokens doing it
- Haiku 5.5 cuts token prices 90% on prompts under 100k
