Claude Fable 5.1 Review: Features, Benchmarks & Pricing

Claude Fable 5.1 is Anthropic's most capable generally available model, released on September 1, 2026 as the successor to Claude Fable 5. It keeps Fable 5's $10/$50 price, cuts cache-read costs by 75%, and at launch took first place on independent leaderboards. If you mainly use Claude for everyday work, Claude Sonnet 5 is the closest option already available in Chat Smith.

This Claude Fable 5.1 review covers what the model is, what its benchmark numbers show, what it costs, how its safeguards work, and how it now compares with Claude Opus 5.5. It also covers access: Claude Fable 5.1 is not yet integrated into Chat Smith, but several other Claude models already are.

Claude Fable 5.1 sits at the top of Anthropic's public lineup, built for the hardest coding, research, and long autonomous tasks. It is the generally available version of the same underlying model as Claude Mythos 5.1, which Anthropic offers only to a small number of trusted organizations; Fable 5.1 adds production safeguards for areas such as cybersecurity and biology.

It supports a 1-million-token context window with up to 128,000 tokens of output, and it is available through the Claude API as claude-fable-5-1. Anthropic says it solves more of its internal coding problems than Fable 5 or Opus 5 and stays readable over long sessions, where earlier models became harder to follow.

Fable 5.1 follows Claude Fable 5, which launched on June 9, 2026. Access to Fable 5 was briefly suspended in June to comply with U.S. export controls and restored on July 1, and Fable 5.1 arrived two months later.

Four areas defined Fable 5.1 at launch.

On Terminal-Bench 4.0, Fable 5.1 scores 55.8% against 52.3% for Opus 5, and on Terminal-Bench-Science 0.1, which covers data analysis and simulations, it reaches 52.6%, far ahead of Opus 5's 29.0%. The team behind the coding agent Devin said it moved its Opus 5 traffic to Fable 5.1 on launch day after it matched or edged out Fable 5 at a lower cost per task.

On OSWorld 2.0 it scores 80.7%, and on GDPval-AA v2.1, which measures economically valuable professional work, it has an Elo of 1,735. Independent tester Vals AI ranked it first on its Vals Index at launch, ahead of Opus 5 and Fable 5.

Vals AI also measured a perfect 100% on ProofBench v1.1, a test of formal mathematical proofs, up from 95% for Fable 5. On Humanity's Last Exam, it scores 65.6% in Anthropic's later comparison table. For students and researchers who work through hard problems with an AI math solver, that depth is what sets the Fable tier apart.

The headline pricing change is caching: cache reads now cost $0.25 per million tokens, 75% less than on Fable 5. Anthropic estimates this cuts typical workload costs by about 25% and highly agentic workloads by up to roughly 45%, which matters for long research sessions or when you need to summarize long articles and reuse the same context many times.

The table below uses Anthropic's published figures from its current benchmark suite. Prices are API rates per million tokens.

The pattern is clear. Fable 5.1 beat Opus 5 across the board at launch, but Opus 5.5, released three weeks later, now leads Fable 5.1 on every row in this table at well under half the price. Fable 5.1 remains Anthropic's top generally available tier by price, but it is no longer the obvious default.

Claude Fable 5.1 costs $10 per million input tokens and $50 per million output tokens, the same as Fable 5 and the highest rate in Anthropic's lineup. Cache reads cost $0.25 per million tokens, just 2.5% of the base input price instead of the usual 10%.

• Developers: available as claude-fable-5-1 through the Claude API.
• Claude Mythos 5.1: the same underlying model without Fable's extra safeguards, available only through Anthropic's trusted-access programs such as Project Glasswing.
• Chat Smith: Claude Fable 5.1 is not integrated yet. You can use Claude Sonnet 5 and other Claude models today, with higher usage limits on Chat Smith Pro at $9.99 per month or $39.99 per year.

Opus 5.5 now beats it on most benchmarks for less than half the price. Unless you specifically need the Fable tier for the hardest tasks, Opus 5.5 is the better value on the API.

Safeguards can affect results. Anthropic evaluated Fable 5.1 with production safeguards on, and on tasks where they intervened it scored zero on OSWorld 2.0. In OpenAI's AutomationBench comparison, Fable 5.1 fell back to Opus 5 on roughly 40% of tasks, so some legitimate security or research work may be redirected.

It is also slow and costly at maximum effort. Artificial Analysis found that at max effort it costs about 20% more per task than Fable 5, because it generates roughly 1.7 times as many output tokens. Test it against alternatives such as GPT-6 Sol on your real tasks before committing.

Claude Fable 5.1 is not integrated into Chat Smith yet, but several Claude models already are. Pick one from the model list and start chatting on Android, iOS, or in the browser, with no Anthropic account needed.

Claude Sonnet 5 is the closest option for most work, with strong writing, research, and coding at a fraction of Fable's price and a 1M-token context window; our Claude Sonnet 5 review covers it in detail. For quick tasks, Claude Haiku 4.5 is faster, and Claude Sonnet 4.6 remains available if you have prompts tuned for it. Fable 5.1's biggest advantages show up mainly in the hardest long-running coding, science, and research tasks.

Because Chat Smith also includes models from OpenAI, Google, xAI, and DeepSeek, you can put the same prompt to a Claude model and its rivals and keep the answer that holds up. Chat Smith Pro adds higher usage limits and more premium models for $9.99 per month or $39.99 per year.

Claude Fable 5.1 pushed Anthropic's top public tier further: stronger long-horizon coding and science, first place on independent leaderboards at launch, a perfect ProofBench score, and much cheaper cached sessions at the same $10/$50 price. Its weak spots are cost, safeguard fallbacks, and the fact that Opus 5.5 now beats it on most benchmarks for far less.

If you want Claude's strengths without API costs, the Claude models already in Chat Smith are an easy way in. You can try them now in Chat Smith, an AI chat app that puts Claude next to the latest models from OpenAI, Google, xAI, and DeepSeek.