Claude Mythos: Access, Pricing, and How It Relates to Claude Fable 5.1

📊 See Claude Fable 5.1 rankings

📱 Get the AI Rank app

Claude Mythos is Anthropic's Mythos-class tier of models, a level Anthropic places above its Opus class in capability. The current version, Claude Mythos 5.1 (model ID claude-mythos-5-1, released September 1, 2026), is the same underlying model as the generally available Claude Fable 5.1, but with different safeguards: Mythos 5.1 is offered only to organizations verified through Anthropic's trusted access programs for cybersecurity and life sciences. Its list price is the same as Fable 5.1: $10 per million input tokens and $50 per million output tokens.

Because Mythos itself is not on the public Arena or OpenRouter leaderboards, the rankings below are for Claude Fable 5.1, its generally available twin. If you arrived here from the anthropic cyber mission announcement, the section below explains how that October 8, 2026 initiative relates to Mythos access. All facts were checked against Anthropic's official pages and dated AI Rank data on October 12, 2026.

What Claude Mythos Is: Preview, Mythos 5, and Mythos 5.1

Anthropic has released three Mythos-class models so far:

Anthropic describes Mythos as its most capable model for cybersecurity and biology research. Its name pairing is deliberate: Anthropic says the safeguards are what distinguish Fable and Mythos, which is why the two releases carry different names. Fable 5.1 is the generally available version with cybersecurity and biology safeguards; Mythos 5.1 is the version with those safeguards adjusted for vetted defenders and researchers.

The Mythos 5 rollout was not smooth. Anthropic suspended access to Fable 5 and Mythos 5 on June 12, 2026, and on July 1 announced that export controls had been lifted and that access to Mythos 5 had been restored for a set of US organizations, following the US government's approval.

Claude Mythos 5.1 Access: Who Can Use It

Mythos 5.1 is available to vetted cyber defenders and life scientists through two programs:

• Cyber Verification Program (CVP): security teams can apply for advanced cyber capabilities on Claude Opus 5.5, Claude Sonnet 5.5, Claude Mythos 5.1, and future models. Since October 6, 2026, the program has three tiers: Defense Access (defensive work such as incident response and vulnerability validation), Red Team Access (adds authorized penetration testing, organizations only), and Specialized Access (the fewest cyber blocks, reviewed in depth with the US government). Existing Project Glasswing members moved to the Specialized Access tier.
• Life Sciences Verification Program: life sciences researchers can apply for advanced biology capabilities.

On the Claude Platform docs, Mythos 5.1 is listed as "Active (verification required)" on the Claude API, Amazon Bedrock (anthropic.claude-mythos-5-1), Google Cloud, and Microsoft Foundry, with retirement not sooner than September 1, 2027. Anthropic also says using Mythos 5.1 requires accepting a 30-day data retention policy for safety monitoring by default, and that Claude Security now runs on Mythos 5.1.

Claude Mythos Pricing and Specs

Anthropic's docs list the same specifications and prices for Mythos 5.1 as for Fable 5.1:

Mythos 5 has the same $10 / $50 base price, but its cache reads cost $1 per million tokens and its knowledge cutoff is January 2026. Anthropic says Fable 5 and Mythos 5 cost less than half the price of Claude Mythos Preview. Prices are in US dollars and were checked on the official pricing page on October 12, 2026.

Claude Mythos vs Claude Fable 5.1

Anthropic says the two are the same underlying model. The practical difference is what the safeguards allow:

• Fable 5.1 can be used to identify software vulnerabilities in source code, but its safeguards route dual-use biology and chemistry research questions to Opus models and prevent penetration testing, exploit generation, and binary-based vulnerability scanning.
• Mythos 5.1 gives verified organizations access to those cybersecurity and biology capabilities.

The difference shows up in Anthropic's own numbers. On Terminal-Bench 4.0, Anthropic reports 55.8% for Fable 5.1 and 60.9% for Mythos 5.1, and says the gap reflects tasks on which its earlier, less precise cyber safeguards intervened. Anthropic expects the difference to be much smaller with the safeguard improvements it shipped with 5.1. For most coding and knowledge work without cyber or biology content, Fable 5.1 is the model you can actually buy.

Claude Mythos on AI Rank: What the Fable 5.1 Data Shows

Claude Mythos 5.1 has no public entry on Arena or OpenRouter, so AI Rank cannot rank it directly. The closest public signal is Claude Fable 5.1. Its Arena entry is claude-fable-5.1-max (Arena text leaderboard with style control, data published October 8, 2026):

→ Track it on AI Rank: Claude Fable 5.1 — Arena rating, rank and 90-day history

📱 Follow it in the AI Rank app

The overall top of the board is tight. Fable 5.1 max's interval overlaps with the models ranked #2 to #5, including Claude Fable 5 high at #4 (1503.7, 40,026 votes), so the order among them is not clear-cut. Gemini 4 Argon high leads overall at 1525.4. In coding, Fable 5.1 max moved up to #15 from #19 in the previous snapshot. The Chinese and math #3 results rest on fewer than 1,000 votes each, so their intervals are wide. Arena measures human preference in chat, not cybersecurity work, so these numbers say nothing about the extra capabilities Mythos unlocks.

On OpenRouter, there is no Mythos model in the public catalog. Fable 5.1 (anthropic/claude-fable-5.1-20260831) ranked #53 by tokens in the week of October 4–10, 2026, with about 225B tokens, down 31% from the prior week. Over the 30 days from September 11 to October 10 it ranked #41 with about 1.53T tokens. You can compare it with other models in the OpenRouter usage ranking on AI Rank.

Anthropic Cyber Mission: How It Relates to Mythos

The Anthropic Cyber Mission, launched October 8, 2026, is Anthropic's long-term commitment to securing the systems everyone depends on. It starts with two areas:

• Critical infrastructure: the new Critical Infrastructure Defense Program brings frontier Claude models, on-site engineers, and threat research to trusted providers that secure operational technology. Founding partners are Accenture, Booz Allen, CrowdStrike, Deloitte, Dragos, Hitachi, Insane Cyber, Nozomi Networks, Palo Alto Networks, PwC, and Rockwell Automation.
• Open-source software: OSS Scanner offers enrolled open-source projects free, regular security scans from Anthropic's most capable models. See our guide to Anthropic OSS Scanner enrollment for how maintainers sign up.

The link to Mythos is Project Glasswing. Anthropic says the Cyber Mission builds on lessons from Glasswing, the program that first gave Mythos Preview to cyber defenders, and that earlier the same week it merged Glasswing into the expanded Cyber Verification Program. In practice, the Cyber Mission is the umbrella effort; the CVP is the route through which security teams apply for Mythos 5.1.

Limitations and What to Check

• Mythos access depends on program approval. The public docs list specs and prices, but they do not mean any account can call claude-mythos-5-1.
• The Arena and OpenRouter figures above belong to Claude Fable 5.1, not Mythos 5.1. They are dated snapshots and will move as votes and usage change.
• Benchmark numbers are Anthropic's own reports under its evaluation settings.
• Check Anthropic's pricing page before budgeting; cache-read pricing differs between Mythos 5 and Mythos 5.1.

Compare before you choose: pricing changes and rankings move daily — check the current model rankings on AI Rank before you commit.

Or 📱 get the AI Rank app to follow them on your phone.

Prepared with automated editorial assistance and checked against official sources and dated AI Rank data on October 12, 2026. We did not run our own benchmarks for this article.