Anthropic has made its most advanced artificial intelligence model accessible to the general public for the first time, implementing rigorous safety protocols.
On Tuesday, the AI company officially launched Claude Fable 5, marking the inaugural public release of its powerful Mythos model. Anthropic highlights Fable 5's exceptional capabilities in areas such as software engineering, complex knowledge work, and computer vision. However, it operates with strict safety limitations; in high-risk domains like cybersecurity, biology, chemistry, and distillation, the model will automatically block responses, defaulting instead to Claude Opus 4.8.
The Mythos model was initially introduced in April as a preview, with access restricted to a select group of partners due to significant cybersecurity considerations. Just last week, Anthropic broadened this access to include hundreds of organizations spanning 15 countries, maintaining a strategic focus on entities responsible for managing critical infrastructure.
This sophisticated technology is now widely available to individuals via Anthropic’s Claude API and its consumption-based Enterprise plans. Access for existing subscribers will be phased in: until June 22, Fable 5 is included without additional cost in Pro, Max, Team, and seat-based Enterprise plans. Starting June 23, Fable 5 will be temporarily removed from these plans, requiring usage credits for access, with Anthropic aiming to reintegrate it as a standard subscription feature as swiftly as possible.
Concurrently, Anthropic is rolling out Mythos 5, an updated version of the Mythos model, to organizations that have already secured approval for its advanced capabilities.
The introduction of Fable 5 occurs as Anthropic prepares for its public market debut, joining contemporaries like OpenAI and Elon Musk’s SpaceX. This launch also follows Anthropic's public appeal for major global AI laboratories to collaboratively establish a "brake pedal" on frontier AI development. The company has expressed concerns that AI systems are advancing at such a rapid pace that they could soon achieve recursive self-improvement (RSI), enabling them to autonomously enhance themselves without human intervention.
Mindful of the potential dangers a Mythos-class model could pose if misused, Anthropic conducted extensive stress-testing of its classifiers through simulated "jailbreak" attempts prior to the release of Fable 5.
“Internally, we ran an external bug bounty that produced no universal jailbreaks in over 1,000 hours of testing. We then worked with external red-teaming orgs which also failed to find universal jailbreaks,” the company stated.
Despite these robust tests, the possibility of novel attack vectors remains. Consequently, with the launch of Fable 5 and Mythos 5, Anthropic will mandate a 30-day data retention policy on all traffic, even for enterprises that previously held zero-retention agreements. The company clarifies that this data will not be used for training purposes but solely to “defend against complex and novel attacks, including new jailbreaks,” and to “identify and reduce false positives.” This policy could establish a new industry benchmark, where access to increasingly powerful AI models is contingent upon mandatory data-retention policies framed as a critical safety measure.
For ongoing users of the model, it is important to note that not every query will receive a Fable 5 response. Anthropic indicates that instances where Fable defers to Opus 4.8 are infrequent, with initial data suggesting that Fable 5 independently handles at least 95% of its sessions.
In independent evaluations, analytics firm Hex reported that Fable 5 was the first model to achieve a 90% score on its core analytics benchmark, which assesses complex, long-running analytical tasks.
“On the hardest questions, it shows strong judgement and attention to nuance,” Hex commented.
Similarly, vibe-coding platform Base44 highlighted Fable's superior ability in "one-shotting full apps" and its exceptional tool-calling capabilities. AI-powered workspace and agent platform Genspark also noted that Fable surpassed all other models in its evaluations, demonstrating significantly enhanced performance in tasks such as UI design and game coding.
The pricing for both Fable 5 and Mythos 5 is set at $10 per million input tokens and $50 per million output tokens, which is double the cost of Opus 4.8. This pricing structure alone may act as a significant deterrent to widespread adoption.
Many enterprises are increasingly scrutinizing AI expenses, having faced unexpected bills or depleted their annual AI budgets prematurely. Advanced models like Opus 4.8 can exacerbate these financial challenges, as their sophisticated reasoning capabilities might split a single request into multiple, cost-accruing tasks.
Anthropic anticipates a very high and unpredictable demand for Fable 5. However, some organizations, such as the shopping rewards platform Rakuten, believe the benefits outweigh the higher price point.
“At the highest effort, Fable reflects on and validates its own work,” Rakuten stated. “For us, that’s what makes highly autonomous operations possible — the extra thinking pays for itself.”
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.