Claude Fable 5.1 and Mythos 5.1: Anthropic's new coding/knowledge-work frontier
Same model, two safeguard tiers: Fable 5.1 is GA, Mythos 5.1 is gated behind trusted-access programs for cybersecurity and life-sciences work. Sets a new standard on coding and long-running problem-solving; ~25% cheaper than Fable 5 on typical workloads, up to ~45% for highly agentic work.
Why it matters
Introduces Enterprise Frontier Safeguards - zero-retention-grade privacy with data stored in customer-controlled cloud infrastructure - plus 60% fewer cybersecurity safeguard false positives, and Fable 5.1 can now be used to discover software vulnerabilities (not write exploits).
What you could build with it
A team running long coding or research agents can benchmark Fable 5.1 at medium effort against its current model; similar results at lower cost would cut the bill directly.
Does it hold up?
Strong evidence of real use: Cursor's own post calls Fable 5.1 its best-scoring model on CursorBench 3.2.0 (73.4%, shipped day one), and gateway vendors shipped migration guides for its breaking API changes - but several users report hallucination regressions and session-quota burn.
Built with Claude Fable 5.1 and Mythos 5.1
- Claude Fable 5.1: 55.8% Terminal-Bench, 25% Cheaperexplainx.ai · Unpacks the real numbers: Terminal-Bench-Science 0.1 doubled (52.6% vs 24.7%), CursorBench 3.2.0 73.4% independently confirmed, cache reads cut 75%.
- Fable 5.1 on Cencori: migration guidegithub · Gateway-side handling of 5.1's breaking API changes: forced tool use removed, always-on thinking via effort levels, append-only history.
- Frontier model releases map (Jun-Sep 2026)github · Charts Terminal-Bench 4.0/Science across Fable 5.1, Opus 5.5, GPT-6 Astra - Fable 5.1 at 55.8%/52.6%.
Learn more
First spotted on engineering blog: source.
More AI models
ElevenLabs Eleven v4 + v4 Turbo: new emotive TTS architecture with a 100ms real-time variantElevenLabs released Eleven v4 and Eleven v4 Turbo on September 28: a new text-to-speech architecture with inline…models · JEV 0.72VoiceStudio: open-source, fully-local ElevenLabs alternative with voice cloning, dubbing and dictation in 646 languagesVoiceStudio is an open-source, fully-local voice AI studio: voice cloning from a 3-second sample, voice design, video…models · JEV 0.71DeepSeek V4-Flash official API: public beta with upgraded agent capabilities, Responses API and Codex supportDeepSeek's official V4-Flash is now a public beta API with massively upgraded agent capabilities - benchmark scores…models · JEV 0.69Mistral Forge: enterprise platform for training and continuously improving proprietary modelsMistral AI launched Forge (Sep 28) — a full-lifecycle model training platform (pre-training, SFT, DPO/ODPO, RL…models · JEV 0.69kev: open, trainable Jev-like family of small decision models on Qwen3.5/3.8A family of small decision models (0.8B to 27B) built on Qwen3.5/3.8 that reproduces Jev's prefill-only typed-decision…models · JEV 0.68GPT-Synopsys: OpenAI and Synopsys multi-year deal to build an AI model that operates EDA toolsOpenAI and Synopsys signed a multi-year agreement on Sep 30 to jointly develop GPT-Synopsys, a specialized model that…models · JEV 0.66
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.