Claude Fable 5.1

Anthropic released Claude Fable 5.1 on September 1, 2026, as a large language model built for coding, research, and knowledge work, available to most business users, with sibling Claude Mythos 5.1 limited to select programs and carrying additional safeguards for cybersecurity and biology. According to Anthropic's publi Anthropic publishes benchmark scores for Fable 5.1 across Terminal-Bench-Science 0.1 (which evaluates agentic scientific research tasks via a simulated terminal), Terminal-Bench 4.0 (agentic coding and software engineering including debugging and strategic code synthesis), Humanity's Last Exam (broad multidisciplinary

An independent capabilities review of Claude Fable 5.1 and Mythos 5.1 published on September 5, 2026 describes the launch as introducing what Anthropic calls the most powerful model by a substantial margin, with Fable 5.1 arriving alongside reduced cache prices, an option for zero data retention, and substantially more Qualitative capability observations highlighted in the review include noticeably improved writing, fewer but still-present Claudisms, better admission of mistakes, more enjoyable conversations, code simplification, and more permissive safety classifiers. The reviewer observes that Fable 5.1 is proactive and tends to us

Anthropic released Claude Fable 5.1 on September 1, 2026, as a generally available successor to Claude Fable 5, alongside a restricted twin, Claude Mythos 5.1, that shares the same weights with loosened safeguards for vetted cybersecurity defenders and life-science researchers. According to the supplied Handy AI launch Pricing holds at $10 per million input tokens and $50 per million output tokens, with cache reads cut 75% to $0.25 per million (from $1 on Fable 5), batch at $5/$25, and unchanged 512-token minimum and cache-write rates ($12.50 five-minute, $20 one-hour). Headline benchmark figures cited in the coverage include Termina

On September 1, 2026, Anthropic announced Claude Fable 5.1 and Claude Mythos 5.1, with the most striking aspect of the launch being a distribution choice rather than a technical leap: both models run the same underlying model, and the only difference is how tightly the applied safeguard layer is tuned. Fable 5.1 is the The article reports a 5.1-point performance gap measured on the same model parameters and the same benchmark under two different safeguard layers, identifying the gap as the measurable cost of safeguard interventions, something the industry has long known intuitively but that is here quantified by numbers the vendor pu

According to the Emergent.sh release-date article, Anthropic launched Claude Fable 5.1 on September 1, 2026, making it generally available the same day across the Claude API, Amazon Web Services, Google Cloud, and Microsoft Azure under the model ID claude-fable-5-1. The article reports that list pricing held steady at Positioned by Anthropic above the Opus, Sonnet, and Haiku tiers, Fable 5.1 is framed as its most advanced model for coding, knowledge work, and long-running problem-solving. The article characterizes the release as confirming earlier leaks that predicted an August-to-September window with unchanged pricing, and identif

Anthropic has released Claude Fable 5.1 for general use alongside Claude Mythos 5.1, a more permissive configuration restricted to approved cybersecurity and life sciences users. According to Anthropic, the two models share the same underlying capabilities but differ in safeguards, with Fable 5.1 available across Claud Anthropic estimates the cache-read cut will reduce typical workload costs by about 25% compared with Fable 5, rising to around 45% for complex coding and highly agentic tasks where large amounts of previous context are repeatedly reused. Cache reads allow the model to reuse context it has already processed rather than

Smart AI Workspace's Tariq Osmani piece documents Claude Fable 5.1 as Anthropic's flagship coding and knowledge-work model released September 1, 2026, shipping as two safeguard configurations of the same underlying model: Fable 5.1 (generally available as claude-fable-5-1 on the Anthropic API, Amazon Bedrock, Google Cl The article highlights AutomationBench results as the headline automation change for Claude Fable 5.1, scoring 31.4% versus 17.1% for Claude Fable 5 — roughly a near-doubling that the author frames as the decisive indicator for unattended multi-step agentic work. It also catalogs partner-reported long-horizon runs: Mon

A third-party deep-dive analyzes Anthropic's September 1, 2026 launch of Claude Fable 5.1 and Claude Mythos 5.1, noting the release includes new benchmark records, cheaper cache pricing, and expanded agentic coding power. The post explicitly cites Anthropic's announcement page and platform documentation, and highlights The analysis positions Fable 5.1 as Anthropic's strongest option for long-running agentic work, coding, and knowledge work that spans hours rather than minutes. Fable 5.1 is generally available to anyone with API access, while Mythos 5.1 is described in Anthropic's platform documentation as identical to Fable 5.1 but d

Claude Fable 5.1 is Anthropic's top-end Claude model, released 1 September 2026 as a point release on Claude Fable 5, with API ID "claude-fable-5-1". The context window is 1,000,000 tokens (default and maximum), max output is 128,000 tokens, multimodal input is text and images with text output, adaptive thinking is alw Pricing keeps Fable 5's $10 per million input and $50 per million output rates, but the cache-read price drops 4× from $1.00 to $0.25 per million tokens — the most consequential change of the release. Fable 5.1 uses the tokenizer introduced with Claude Opus 4.7, which produces roughly 30% more tokens for the same text

DataCamp describes Claude Fable 5.1 as Anthropic's generally available frontier model for long-horizon coding and research work, sitting above Opus 5 ($5/$25 per million tokens) and Sonnet 5 ($2/$10) on price, with latency rated as slower than both. The page confirms a 1 million-token context window with 128K maximum o Pricing details confirm Fable 5.1 keeps Fable 5's list price of $10/$50 per million input/output tokens while cutting cache reads 75% to $0.25 per million, which Anthropic estimates makes typical workloads about 25% cheaper and highly agentic workloads up to roughly 45% cheaper than Fable 5. The headline benchmark gain

Anthropic released Claude Fable 5.1 and Claude Mythos 5.1 on September 2, 2026, as reported by SD Times. Internal testing shows Fable 5.1 outperforms Fable 5 on coding, knowledge work, and problem-solving tasks, with similar results at lower cost when running at low or medium effort levels. Craig Falls of Jane Street C Fable 5.1 and Mythos 5.1 are the same underlying model with different safeguard regimes: Fable 5.1 is generally available, while Mythos 5.1's safeguards target cybersecurity and life sciences and is available only through trusted-access programs. Enterprise Frontier Safeguards (EFS) provide zero-data-retention-level pr