# Anthropic Launches Claude Fable 5 With Safety Caps > Source: [https://botensten.com/articles/anthropic-claude-fable-5-mythos-launch](https://botensten.com/articles/anthropic-claude-fable-5-mythos-launch) (canonical) > Author: iCharles News — Botensten, https://botensten.com > Published: 2026-06-10 ## TL;DR Anthropic has launched Claude Fable 5, a Mythos-class model it describes as exceeding the capabilities of any model it has previously made generally available. The release includes safeguards that redirect certain queries to Claude Opus 4.8; those safeguards trigger in fewer than 5% of sessions on average. A separate, less restricted version called Claude Mythos 5 is being deployed for a small group of cyberdefenders and infrastructure providers through Project Glasswing, in collaboration with the US government. ## What did Anthropic release? Anthropic launched Claude Fable 5, a Mythos-class model now available for general use. According to [Anthropic's announcement](https://www.anthropic.com/news/claude-fable-5-mythos-5), Fable 5 outperforms every model the company has previously made publicly available. It shows strong results across software engineering, knowledge work, vision, and scientific research. Anthropic also notes a clear pattern: the longer and more complex a task, the wider Fable 5's lead over its other models. --- ## What safeguards are in place? Fable 5 is capable enough in areas like cybersecurity that misuse is a real concern. Because of this, Anthropic launched it with safeguards that redirect certain queries to [Claude Opus](/articles/claude-opus-4-8-anthropic-s-latest-upgrade-leaves-many) 4.8 — its next-most-capable model — instead of Fable 5. Anthropic tuned these safeguards conservatively. That means they will sometimes block harmless requests. On average, the safeguards trigger in fewer than 5% of sessions. The company says it will work to cut false positives as more capable models arrive in the coming months. If you want a closer look at Claude Opus 4.8 itself, see the [Claude Opus 4.8 review covering whether it's an incremental upgrade or hype](/articles/claude-opus-4-8-anthropic-s-latest-upgrade-leaves-many). --- ## Who gets access to Mythos 5? Alongside the general release, Anthropic is launching Claude Mythos 5 for a small group of cyberdefenders and infrastructure providers. As [Anthropic details](https://www.anthropic.com/news/claude-fable-5-mythos-5), Mythos 5 uses the same underlying model as Fable 5. The key difference: safeguards are lifted in some areas. Mythos 5 will first be deployed through Project Glasswing. That project runs in collaboration with the US government and serves as an upgrade to Claude Mythos Preview. A separate report from [sources.news](https://sources.news/p/inside-apples-siri-revamp-anthropic) noted that Anthropic was preparing to release a public version of Mythos ahead of the official announcement. --- ## Frequently asked questions **Q: What is the difference between Claude Fable 5 and Claude Mythos 5?** Both use the same underlying model. Fable 5 is the general-use version with safety safeguards active. Mythos 5 has those safeguards lifted in some areas. It is available only to a small group of cyberdefenders and infrastructure providers. **Q: How often do Fable 5's safeguards redirect a request to Claude Opus 4.8?** On average, the safeguards trigger in fewer than 5% of sessions. Anthropic acknowledges they are tuned conservatively and will sometimes catch harmless requests. **Q: Where is Claude Mythos 5 being deployed first?** Mythos 5 is being deployed through Project Glasswing, in collaboration with the US government. It replaces Claude Mythos Preview. --- Anthropic's confirmed next step: improve the safeguards and reduce false positives as more capable models arrive in the coming months. ## Frequently asked questions **What is the difference between Claude Fable 5 and Claude Mythos 5?** Both use the same underlying model, but Fable 5 is the general-use release with safety safeguards in place. Mythos 5 has those safeguards lifted in some areas and is available only to a small group of cyberdefenders and infrastructure providers. **How often do Fable 5's safeguards redirect a request to Claude Opus 4.8?** On average, the safeguards trigger in fewer than 5% of sessions, though Anthropic acknowledges they are tuned conservatively and will sometimes catch harmless requests. **Where is Claude Mythos 5 being deployed first?** Mythos 5 is being deployed through Project Glasswing, in collaboration with the US government, as an upgrade to Claude Mythos Preview.