Claude on WholeTech network
home/products/mythos
Research preview · gated

Claude Mythos preview

A specialized Claude model for defensive cybersecurity research. Announced April 7, 2026 as part of Project Glasswing. Invitation-only — not generally available.

Status: Research preview, gated. Available only to Project Glasswing partners (12 founding orgs) and 40+ critical-infrastructure organizations. Anthropic states no general public release is currently planned. Pricing: $25 input / $125 output per million tokens — research-grade, not consumer-grade.
01 — What it is

A model trained for vulnerability research.

Mythos is the first publicly named Claude model specialized for a single research domain: defensive cybersecurity. It identifies zero-day vulnerabilities in operating systems, web browsers, and critical software, and can autonomously discover and exploit security flaws with minimal human guidance. Anthropic positions it as a tool for the defensive side of the AI-era security arms race.

The most-quoted finding from the launch announcement: Mythos uncovered a 27-year vulnerability in OpenBSD and a 16-year flaw in FFmpeg that automated testing had encountered five million times without detecting. The number is intended to dramatize the capability gap; the underlying point is that pre-Mythos automated testing had a ceiling, and this model breaks it.

02 — Benchmarks

How it compares.

Anthropic published four benchmarks at launch, each comparing Mythos to Claude Opus 4.6:

CyberGym

83.1% Mythos vs. 66.6% Opus 4.6. Vulnerability reproduction. The headline number.

SWE-bench Verified

93.9% Mythos vs. 80.8% Opus 4.6. General software-engineering tasks.

SWE-bench Pro

77.8% Mythos vs. 53.4% Opus 4.6. Harder professional engineering benchmark.

Terminal-Bench 2.0

82.0% Mythos vs. 65.4% Opus 4.6. Terminal-driven agentic tasks.

Real-world finds

Thousands of high-severity vulnerabilities across every major OS and browser; specific finds include OpenBSD (27 yrs old) and FFmpeg (16 yrs).

Pricing

$25 / $125 per million input/output tokens. Roughly 5× Opus 4.7’s rate — reflects the research-tier positioning.

03 — Project Glasswing

The launch vehicle.

Mythos didn’t ship as a product. It shipped as the engine of an industry initiative. Project Glasswing is a 12-org coalition launched alongside it — named for the glasswing butterfly, the “hidden vulnerabilities and transparency in defense” metaphor. The founders:

Amazon Web Services · Anthropic · Apple · Broadcom · Cisco · CrowdStrike · Google · JPMorgan Chase · Linux Foundation · Microsoft · NVIDIA · Palo Alto Networks.

Plus 40+ additional critical-infrastructure organizations with access. The stated goal: get Mythos-class capability into defenders’ hands before attackers reach parity.

Anthropic also committed: $100M in model-usage credits for participating researchers, $2.5M to Alpha-Omega/OpenSSF (Linux Foundation), and $1.5M to the Apache Software Foundation. There’s a 90-day public-reporting cycle on findings and fixed vulnerabilities.

04 — Updates timeline

Thirty days in.

A running log of post-launch movement, refreshed daily by the local Mythos watcher (cron 06:15 UTC) that diffs Anthropic’s public surfaces and flags changes. Newest first.

2026-05-09
Mythos enters the canonical model menu. The anthropic.com top navigation now lists Models > Mythos preview alongside Opus, Sonnet, and Haiku. A subtle promotion — on April 7 the announcement page lived under /glasswing and didn’t appear in the global product nav. As of today, every page on Anthropic’s marketing site offers Mythos as a navigable surface. Read: it has graduated from a launch event to a permanent shelf position. (Detected by the watcher this morning; research, news, and release-notes all changed.)
2026-05-09
Release notes refreshed. The docs.anthropic.com/release-notes overview page changed; the watcher saw a content delta. No GA-style entry yet — the change is most likely a Mythos-context insertion, not a new public-tier announcement. Worth re-reading directly.
2026-05 (week 1)
Industry response settled into a posture. Major security publications framed Mythos as a defender-side capability whose primary risk is uneven distribution — the founding 12 + 40 critical-infrastructure orgs hold capability, smaller defenders do not. Expect this to drive the licensing / access debate over the next quarter.
2026-04-21
Quiet tools-only period. No public Glasswing findings yet (the 90-day cycle starts ticking from launch). Recipient orgs are presumably triaging internally before disclosure. The first batch of mandated public reports is due around ~July 6, 2026.
2026-04-07
Launch. Mythos preview announced as the engine of Project Glasswing. 12 founding orgs (AWS, Anthropic, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorgan Chase, Linux Foundation, Microsoft, NVIDIA, Palo Alto Networks). $25/$125 per million tokens. 27-year OpenBSD vulnerability and 16-year FFmpeg flaw cited as headline finds. Benchmarks: 83.1% CyberGym vs. 66.6% on Opus 4.6.
05 — Adoption signals

What we’re reading between the lines.

Mythos’s public footprint is intentionally narrow. The signal is in adjacent moves — pricing, product-nav placement, partner statements, and what shows up in sibling Anthropic surfaces.

Nav promotion is the strongest signal so far

April 7 launched at /glasswing — a campaign URL. May 9, Mythos joins the global model menu. Anthropic doesn’t list research-preview models in nav unless they expect them to stay. This is the institutional commitment showing.

Pricing tier is doing strategic work

$25 / $125 per million tokens is roughly 5× the consumer Opus rate. That’s not “bigger model is more expensive” pricing — it’s gating. Anthropic is using the price tag to keep usage to org-scale defenders, not curious individuals.

The $4M in grants is buying narrative cover

$2.5M to Alpha-Omega/OpenSSF, $1.5M to Apache — the non-profit OSS security ecosystem. When Mythos finds something nasty in a widely-deployed library, the disclosure path is already funded and friendly. Smart defensive PR move; also actually useful.

The benchmark delta is the real product

SWE-bench Verified jumped from 80.8% (Opus 4.6) to 93.9% (Mythos). That’s a 13-point improvement on a general-purpose engineering benchmark, not just security. Whatever training recipe makes Mythos better at vulnerability-hunting also makes it better at code in general. That recipe will surface in next-generation Opus / Sonnet long before Mythos itself does.

The watcher catches what the headlines don’t

Most of the last 30 days has been silent on the marketing side and visible only as small content diffs — release-note paragraphs, nav reshuffles, doc cross-links. That’s why the watcher exists: launch energy fades but the institutional footprint keeps moving. Dashboard at wholetech.com/admin/mythos-watcher/.

Defender / attacker delta still the unspoken subtext

Every Mythos discussion eventually returns to the question Anthropic carefully doesn’t answer: when does the same capability show up on the offensive side? CrowdStrike’s “minutes” framing at launch was the most direct industry comment. Watch for analogous statements from the defensive vendor coalition over the summer.

06 — What to watch for next

Where this is going.

07 — Practical takeaways

What this means for builders.

08 — Sources

Read it from Anthropic.

Last refreshed 2026-05-09 after the Mythos watcher (cron daily 06:15 UTC, dashboard at wholetech.com/admin/mythos-watcher/) detected three Anthropic-surface changes today.