# Claude Mythos Preview Claude Mythos Preview is an AI model released by Anthropic on Apr 7 2026. It has a 1M context window. Tracked results include 94.5% on GPQA Diamond, 93.9% on SWE-Bench Verified and 64.7% on Humanity's Last Exam (with tools). ## Facts | Field | Value | | --- | --- | | Model | Claude Mythos Preview | | Developer | Anthropic | | Release date | Tuesday, Apr 7 2026 | | Licensing | Closed | | Context window | 1M | ## Tracked benchmark scores | Benchmark | Score | Source | What it measures | | --- | --- | --- | --- | | SWE-Bench Pro | 77.8% | Lab | Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better. | | SWE-Bench Verified | 93.9% | Lab | Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | | SWE-Bench Multilingual | 87.3% | Lab | Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | | SWE-Bench Multimodal | 59% | Lab | Real bug reports that arrive with pictures attached — a screenshot, a mockup, a page rendering wrongly — so the AI has to read the image as well as the code to work out what to fix. Higher is better. | | Terminal-Bench 2.0 | 82% | Lab | Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? (Version 2.0 of the test.) Higher is better. | | CyberGym | 83.1% | Lab | Tests the AI on cybersecurity challenges — finding and exploiting software weaknesses inside a safe sandbox. Higher is better. | | Humanity's Last Exam (no tools) | 56.8% | Lab | Humanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better. | | Humanity's Last Exam (with tools) | 64.7% | Lab | Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | | GPQA Diamond | 94.5% | Lab | Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | | OSWorld-Verified | 79.6% | Lab | Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better. | | CharXiv Reasoning | 86.1% | Lab | Can the AI read and reason about complex charts and figures, not just text? Higher is better. | | CharXiv Reasoning (with tools) | 93.2% | Lab | The same chart-and-figure reasoning test, run with the AI allowed to use tools — writing code to inspect the image, for instance — rather than reading the chart unaided. Scores run higher than the unaided version, so read the two as separate tests. Higher is better. | | GraphWalks BFS (256K–1M) | 80% | Lab | breadth-first search F1 on the 200 GraphWalks problems with 256K to 1M tokens of context | ## About Claude Mythos Preview Anthropic announced Claude Mythos Preview on April 7, 2026 through Project Glasswing. Selected software maintainers and security teams used it to find and help patch vulnerabilities in widely deployed software. One investigation uncovered a flaw that had been present in OpenBSD for 27 years. In Anthropic’s CyberGym evaluation, the preview reproduced the target vulnerability on its first attempt in 83.1% of tasks. Opus 4.6 scored 66.6% in the same evaluation. Anthropic committed usage credits to the project and announced rates of $25 per million input tokens and $125 per million output tokens for participants after the preview. In June it offered those partners an upgrade to Claude Mythos 5. ## Questions and answers ### When was Claude Mythos Preview released? Claude Mythos Preview was released by Anthropic on Tuesday, Apr 7 2026. ### Who made Claude Mythos Preview? Claude Mythos Preview was built by Anthropic. AI safety company building the Claude family of models. Founded in 2021 by former OpenAI researchers. ### What benchmark scores did Claude Mythos Preview get? Claude Mythos Preview reports 13 tracked benchmark scores — SWE-Bench Pro: 77.8%; SWE-Bench Verified: 93.9%; SWE-Bench Multilingual: 87.3%; SWE-Bench Multimodal: 59%; Terminal-Bench 2.0: 82%; CyberGym: 83.1%; Humanity's Last Exam (no tools): 56.8%; Humanity's Last Exam (with tools): 64.7%; GPQA Diamond: 94.5%; OSWorld-Verified: 79.6%; CharXiv Reasoning: 86.1%; CharXiv Reasoning (with tools): 93.2%; GraphWalks BFS (256K–1M): 80%. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data. ### What is the context window of Claude Mythos Preview? Claude Mythos Preview has a context window of 1M. Providers may specify separate input and output token limits. ### Is Claude Mythos Preview open source? No. Claude Mythos Preview is a closed model. The weights are not published — it is available only through the provider's own API, apps, or partner platforms. ### What came before and after Claude Mythos Preview? Anthropic's previous tracked release was Claude Sonnet 4.6 on Feb 17 2026, 49 days earlier. It was followed by Claude Opus 4.7 on Apr 16 2026. --- Canonical page: https://aireleasetracker.com/model/anthropic/claude-mythos-preview Site index: https://aireleasetracker.com/llms.txt Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.