4 hrs ago
Anthropic launches Claude Fable and Mythos 5.1 models
Anthropic has created two new versions of its Claude artificial intelligence system.
They are called Fable 5.1 and Mythos 5.1.
Both use the same basic AI technology.
Fable is available more generally, while Mythos has stronger limits and is only given to approved users.
Anthropic says Fable is better at coding, research and long tasks.
It also says the new model costs less to use than the previous version.
In one example, the model reportedly helped identify a very unusual software problem.
Anthropic says Mythos is intended for vetted cybersecurity and life-science researchers.
Anthropic launched Claude Fable 5.1 and Claude Mythos 5.1 for coding, knowledge work and scientific research.
The models use the same underlying AI system but have different safeguards and access rules.
Anthropic says Fable 5.1 costs about 25% less for typical workloads and up to 45% less for highly agentic tasks.
The company reports Fable 5.1 scored 52.6% on Terminal-Bench-Science, compared with 24.7% for Fable 5.
Mythos 5.1 is restricted to approved cybersecurity and life-science work, mainly through selected US organisations.
- Who
- Anthropic launched the Claude Fable 5.1 and Claude Mythos 5.1 models.
- What
- The company introduced AI models for coding, knowledge work, scientific research and related tasks.
- Where
- Mythos access is currently limited mainly to selected organisations in the United States; no launch location is specified.
- When
- The models were launched, but the article does not provide a specific date.
- Why
- Anthropic says the models are intended to improve performance on long-running, technical and research-oriented work while reducing costs for Fable users.
Key facts
- Models
- Claude Fable 5.1 and Claude Mythos 5.1
- Underlying system
- Both models use the same underlying AI system.
- Typical workload savings
- Anthropic says Fable 5.1 costs about 25% less than Fable 5.
- Highly agentic workload savings
- Anthropic says savings can reach about 45% because of lower cache-read pricing.
- Terminal-Bench-Science score
- Fable 5.1 scored 52.6%, compared with 24.7% for Fable 5, according to Anthropic.
- Fable access
- Fable 5.1 is generally available.
- Mythos access
- Mythos 5.1 is available through trusted access programmes for approved organisations, mainly in the United States.
Quotes
Dianne Penn
Anthropic’s product management leader for research and labs
“This was one area we wanted to make improvements on, responding to customer feedback”
deccanchronicle.com







