Claude Mythos Preview
C Tier · 6.5/10
Anthropic's most capable model -- a gated research preview via Project Glasswing, cybersecurity-specialized. 73% success on expert CTF tasks, 32-step autonomous network attacks. Not generally available.
Score Breakdown
Personality & Tone
The gated red-team specialist
Tone: When Anthropic does publish Mythos outputs (in sanitized research reports), the voice is careful, technically dense, and deliberately unperformed -- much more 'senior security researcher writing an internal memo' than Claude Opus's conversational style.
Quirks: Mythos is tuned to produce its cybersecurity reasoning with extensive show-your-work traces. Anthropic publishes some outputs with full CoT visible as evidence of capability claims. Outside of security tasks, the model reportedly sounds much like Opus 4.6 / 4.7 -- Anthropic hasn't published a distinct general-purpose voice for Mythos.
The Good and the Bad
What we like
- +The most capable Anthropic model available -- meaningfully stronger than Opus 4.7 on cybersecurity reasoning, long-horizon autonomy, and multi-step attack/defense planning per Anthropic's published evaluations
- +73% success rate on expert-level Capture-the-Flag tasks -- a benchmark other frontier models (GPT-5.x, Gemini 3.1 Pro, Opus 4.7) are well below
- +Autonomously executes 32-step network attacks in Anthropic's red-team evals -- demonstrates sustained agentic capability on security tooling without losing track
- +Paired with Project Glasswing: a coalition model where 8 founding enterprise partners get controlled access, $100M in credits, and shared threat intelligence
What could be better
- −Not available to the public. If you're reading this thinking you might use it: you probably can't. Invite-only rollout to ~50 orgs with active cybersecurity or research commitments
- −Even if you are in a Glasswing partner org, access is heavily gated -- deployment requires explicit use-case approval and extensive safety review
- −Specialized for security work. Anthropic explicitly notes Mythos is 'less broadly capable' than Opus 4.7 outside the cyber domain -- so it is NOT the answer for general coding, writing, or analysis work
- −Anthropic withholding the weights and API access is a policy call, not a technical one. This is the first time a frontier Claude model has been deliberately kept out of the API, signaling a new safety/release posture you should expect to see repeat
Pricing
Project Glasswing (Gated)
- ✓Not publicly available -- access limited to ~50 pilot organizations
- ✓Founding partners: Amazon, Apple, Google, Cisco, CrowdStrike, JPMorgan, Microsoft, Nvidia
- ✓$100M in total Anthropic credit commitments across partners
- ✓$4M in open-source security donations
- ✓Cybersecurity research and defense use cases only
Public access
- ✓Anthropic deliberately withholding broad release due to cybersecurity risk
- ✓For general-purpose work, use Claude Opus 4.7 (see /tools/claude)
- ✓Anthropic describes Mythos as 'less broadly capable' than Opus 4.7 outside cyber tasks
Known Issues
- Mythos's cybersecurity capability is the reason for its gated release. Anthropic's red-team evaluations showed the model could plan end-to-end network intrusion chains, which Anthropic deemed too risky for open API accessSource: Anthropic Project Glasswing announcement, Axios, CNBC, Schneier on Security · 2026-04
- Naming convention is confusing: 'Claude Mythos Preview' is the public product name, internal codename was Capybara, and it's sometimes referred to as 'Mythos 5' by third-party reporters (there is no Mythos 1-4)Source: Axios, Fortune · 2026-04
- Access applications are not open -- Anthropic is approaching partner orgs directly rather than accepting inbound requestsSource: Anthropic Glasswing page · 2026-04
- Axios reported 2026-04-19 that the NSA is among the ~40 orgs with Mythos access, despite the Pentagon's formal supply-chain risk designation of Anthropic. Dario Amodei reportedly met with W.H. Chief of Staff Susie Wiles and Treasury Secretary Scott Bessent on 2026-04-17. Material context if you are evaluating Mythos / Glasswing in a federal or defense-adjacent procurement -- the political posture inside the US government is not uniformSource: Axios, TechCrunch, Engadget · 2026-04
Best for
Partner organizations in Project Glasswing doing cybersecurity research, defensive red-teaming, threat intelligence, or large-scale vulnerability triage. If your use case is legitimate cybersecurity and you have enterprise Anthropic contact, ask about Glasswing admission.
Not for
Everyone else. For general coding, writing, analysis, agent work, or consumer use: use Claude Opus 4.7 (see /tools/claude). It is Anthropic's most capable generally-available model, and for >95% of real-world tasks it's functionally equivalent.
Our Verdict
Claude Mythos Preview is the first frontier Claude model Anthropic deliberately kept out of the public API. Announced alongside Project Glasswing on April 7, 2026, it's a cybersecurity-specialized model that posts uncommonly high scores on expert CTF tasks and long-horizon agentic security work -- high enough that Anthropic judged broad release too risky. For the ~50 pilot organizations with access (including Apple, Google, Microsoft, Nvidia, JPMorgan), Mythos is a real capability leap on security-domain tasks. For everyone else, it's a signal about where frontier release policy is heading: expect more 'gated preview' drops that never reach broad GA. If you're not in Glasswing, use Opus 4.7 and don't lose sleep over it -- the general-purpose quality gap is small outside the cyber niche.
Sources
- Anthropic: Project Glasswing (accessed 2026-04-17)
- Anthropic Red: Mythos Preview (accessed 2026-04-17)
- Fortune: Anthropic's Mythos model + Project Glasswing (accessed 2026-04-17)
- Axios: Anthropic releases Opus 4.7, concedes it trails unreleased Mythos (accessed 2026-04-17)
- Schneier on Security: On Mythos Preview and Project Glasswing (accessed 2026-04-17)
- CNBC: Anthropic Opus 4.7 less risky than Mythos (accessed 2026-04-17)
- Axios: NSA uses Mythos despite Pentagon feud (2026-04-19) (accessed 2026-04-20)
- TechCrunch: NSA spies reportedly using Anthropic Mythos (accessed 2026-04-20)
Explore more Claude Mythos Preview rankings
Deeper leaderboards, benchmarks, task-specific tier lists, and status/pricing pages for Claude Mythos Preview.
The Tier List Tuesday
Weekly newsletter: tier movers, new entrants, and the VS of the week. Built from our daily AI-tool sweeps. No spam, unsubscribe anytime.
Alternatives to Claude Mythos Preview
Claude (Anthropic)
Anthropic's flagship LLM -- Opus 4.7 (launched April 16, 2026) with 1M-token context, high-res vision, new xhigh reasoning level, and the most natural conversational style. Note: 2026-04-04 policy excluded third-party agent harnesses (OpenClaw etc.) from Pro/Max flat-rate, and 2026-04-16 Enterprise pricing dropped bundled tokens
Gemini (Google)
Google's LLM with deep Google Workspace integration, 2M token context window, and native code execution
Grok
xAI's irreverent chatbot with a direct line to X/Twitter -- real-time data meets unfiltered personality. Grok 4.3 production launched 2026-05-02 with Custom Voices cloning + Imagine Agent Mode + ~40% API price cut to $1.25/$2.50 per 1M tokens
Muse Spark (Meta)
Meta's first model from its Superintelligence Lab -- natively multimodal with Contemplating mode for multi-agent reasoning
GPT-Rosalind (OpenAI)
OpenAI's first domain-specific model -- life sciences, drug discovery, translational medicine. Launched 2026-04-16 as a Trusted Access research preview. Launch partners: Amgen, Moderna, Allen Institute, Thermo Fisher. Paired with a Life Sciences Codex plugin (50+ scientific tool integrations)
GPT-5.4-Cyber (OpenAI)
OpenAI's defensive-cybersecurity variant of GPT-5.4, launched 2026-04-16. Lowered refusal boundary for security-research tasks and native binary reverse-engineering. Access gated via Trusted Access for Cyber (TAC) program -- thousands of verified defenders, hundreds of teams, no public pricing
Hunyuan 3 (Tencent Hy3)
Tencent's Hy3 Preview launched 2026-04-23 -- 295B total / 21B active MoE, 256K context, open-sourced on HuggingFace under tencent/Hy3-preview. Cheapest frontier-class API at ~1.2 RMB per million input tokens. Integrated into Yuanbao, WeChat, QQ
MiMo (Xiaomi)
Xiaomi's MiMo-V2.5 family launched 2026-04-22 -- Pro (1T total / 42B active MoE, 1M context, native vision+audio reasoning), Multimodal base, TTS (3 sub-models: base, VoiceDesign, VoiceClone), and ASR (open-source, English + Chinese + major dialects). Full voice pipeline for the agent era. Extra-charge 1M-context tier removed at launch