Claude Fable 5.1
Anthropic's September 2026 Claude update: cheaper cache reads and sharper agentic coding at the same per-token price.
Last verified · 2026-10-23 · by Moe Ameen
Published 2026-09-01 · Updated
Short answer
Claude Fable 5.1 is Anthropic's September 2026 update to Fable 5. It is the same model as Mythos 5.1 with broader safeguards. It keeps the $10 input and $50 output price per million tokens, cuts cache reads 75% to $0.25, and scores 55.8% on Terminal-Bench 4.0. It runs on Claude.ai, the API, AWS, Google Cloud, and Azure. It writes and codes but makes no images or video.
Key takeaways
- Fable 5.1 and Mythos 5.1 are the same model; Mythos is limited to vetted cyber and life-sciences users.
- Pricing stays $10 in / $50 out per million tokens; cache reads drop 75% to $0.25.
- Anthropic puts typical savings near 25%, and up to 45% on highly agentic tasks.
- Terminal-Bench 4.0 rose to 55.8% from 42.0% on Fable 5.
- It is text only; Kompozy renders its scripts as video, carousels, and posts.
What is Claude Fable 5.1?
Claude Fable 5.1 is Anthropic's updated frontier language model, released in September 2026 as a refresh of Fable 5. It is the publicly available, safeguarded face of Anthropic's restricted Mythos line — the same underlying capability with guardrails that route a small set of high-risk topics to more constrained handling. Its sibling, Claude Mythos 5.1, relaxes some of those restrictions for cybersecurity and life-sciences work and is limited to vetted organizations through Anthropic's trusted-access programs.
The update keeps Fable 5's per-token price ($10 per million input, $50 per million output) but cuts cache-read costs by roughly 75% (to about $0.25 per million tokens), with typical workloads landing around 25% cheaper overall and up to 45% cheaper for highly agentic tasks. Anthropic reports gains over Fable 5 on agentic coding benchmarks, including Terminal-Bench 4.0 (55.8% versus 42.0%).
The safeguards also got less jumpy. Anthropic reports 85% fewer false positives on benign medical queries and a 60% drop in false-positive cybersecurity interventions, and Fable 5.1 now permits vulnerability detection for defensive security work. Fable 5.1 is available on Claude.ai and via the API (model id claude-fable-5-1) on AWS, Google Cloud, and Microsoft Azure.
One thing to keep straight for creative work: Fable 5.1 is a text and reasoning model. It writes, structures, analyzes, codes, and reasons — it does not render images, video, or audio. Treat any exact benchmark or price as a launch snapshot and check Anthropic's docs for current figures.
What can you make with Claude Fable 5.1?
- Video scripts and hooks with strong narrative structure for shorts, avatar video, and podcasts
- Long-form blog articles, newsletters, and research summaries from raw notes or transcripts
- A carousel narrative arc or listicle outline written to a specific slide count and beat
- Platform-native captions and thread drafts written to a brief, with effort tuned per pass
- Rewriting and editing passes: tighten, restructure, or shift the voice of existing copy
- Code and technical docs for builders shipping content alongside a product
How does Kompozy turn Claude Fable 5.1 output into content?
Fable 5.1's real edge for creators is the writing itself — long-form structure, a coherent narrative arc, and effort you can dial up for a hard creative brief. That makes it a strong script-and-story brain, but a script is not a video and an outline is not a carousel. Kompozy is the production layer that turns those words into finished media. Take a Fable-class script and Kompozy renders it as a HeyGen persona or avatar short with burned-in captions; take a carousel arc and it builds a brand-exact multi-slide post through HyperFrames; take a blog outline and it drafts the full article and a matching newsletter. The idea travels from words to a whole content set without you opening a separate editor for each format.
Kompozy's per-post review gate is also where you add your own AI-disclosure line before anything ships. In practice you write the concept in Fable 5.1, generate the formats in Kompozy governed by your Persona Brief so the voice stays consistent, and publish the set across the eight social platforms plus blog and email on autopilot. Fable 5.1 owns the creative draft; Kompozy owns the render, the format fan-out, the schedule, and the publish.
- Use Fable 5.1 (via Claude.ai or the API) to write the script, carousel arc, or blog outline, dialing effort up for the hard creative beats.
- Bring the draft or transcript into Kompozy and pick the formats — persona/avatar short, carousel, listicle video, blog, newsletter, text posts.
- Let Kompozy generate each format in your voice via your Persona Brief, rendering video, branded captions, and slides.
- Reframe and caption the video for each platform automatically, adding your AI-disclosure line at the per-post review gate.
- Schedule and publish the whole set across the eight social platforms plus blog and email from one queue.
What does this look like in Kompozy?
This carousel came out of the Kompozy workspace as is. Its subject has nothing to do with Claude Fable 5.1, so look at it for the format, not the topic.
Who is this for, and who should skip it?
Best for:
- Teams running agentic coding or long, cached workloads
- Writers who want a strong drafting model for scripts and long-form
- Defensive security and medical teams hit by false refusals before
Not ideal for:
- Anyone who needs image, video, or audio output
- High-risk cyber or life-sciences work, which Anthropic routes to Mythos 5.1 for verified users
- Budget bulk tasks where a cheaper model is enough
Frequently asked questions
What is Claude Fable 5.1?
Claude Fable 5.1 is Anthropic's September 2026 update to Fable 5 — a frontier text and reasoning model and the publicly available, safeguarded face of the restricted Mythos line. It keeps Fable 5's per-token price but cuts cache reads roughly 75% and improves on coding and agentic tasks. It does not generate images, video, or audio.
How is Fable 5.1 different from Fable 5?
It holds Fable 5's $10/$50 per-million-token pricing while cutting cache reads about 75%, reports higher agentic coding scores (Terminal-Bench 4.0 at 55.8% versus 42.0%), and cuts false-positive safety interventions on benign medical and cybersecurity requests.
How is Fable 5.1 different from Mythos 5.1?
They are the same underlying model. Fable 5.1 is generally available with standard safeguards; Mythos 5.1 relaxes some restrictions for cybersecurity and life-sciences work and is limited to vetted organizations through Anthropic's trusted-access programs.
Can Claude Fable 5.1 make videos or images?
No. Fable 5.1 is a language model — it writes, reasons, analyzes, and codes but produces no video, images, or audio. To turn its scripts and outlines into published persona video, carousels, blogs, and posts, pair it with a content engine like Kompozy that renders the media and publishes across platforms.
Can Claude Fable 5.1 help with security or medical topics?
Yes, within limits. Anthropic says Fable 5.1 gives 85% fewer false positives on benign medical queries and 60% fewer false-positive cybersecurity interventions, and now permits vulnerability detection for defensive security work. Higher-risk cyber and life-sciences work goes to Mythos 5.1 through Anthropic's verification programs.
How do we know this?
We checked every figure and product claim on this page against the sources listed below on 2026-10-23, and removed or corrected anything they did not support. Features Anthropic's announcement does not describe were removed. Kompozy plan and platform details come from kompozy.io/pricing and kompozy.io.
Sources
- Anthropic: Claude Fable 5.1 and Claude Mythos 5.1 (read 2026-10-23)
- Kompozy: Home (read 2026-10-23)
Related tools
- Fable 5 — Anthropic's most powerful publicly available Claude model — a Mythos-class model made safe for general use.
- Mythos — Anthropic's frontier Claude language model — restricted at launch, reaching the public through Claude Fable 5.
- Claude Content Watermarking — Anthropic's provenance system for Claude — an invisible, machine-readable watermark in generated text plus signed C2PA metadata on files, so content Claude had a hand in can be identified and verified.
- Fable 5.1 World Modeling — PhiloLabs' open-source project that uses Claude Fable 5.1 agent swarms to generate explorable, browser-native 3D reconstructions of real places as plain Three.js apps.
- Claude Opus 5.5 — Anthropic's September 2026 frontier Claude model — about 20% cheaper per token than Opus 5, more than 30% faster, with a #1 debut on the Artificial Analysis Intelligence Index and a new class of safeguards. A text-output model, not an image, audio, or video generator.