AI Model Release Tracker

Claude Fable 5

Released

Claude Fable 5 is an AI model released by Anthropic on Tuesday, Jun 9 2026, 12 days after Claude Opus 4.8. It has a 1M token context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider.. Benchmark results (shown below) cover HealthBench Professional, GDPval-AA, BullshitBench v2, Gray Swan IPI, SWE-Bench Pro, SWE-Bench Verified, and 18 more.

API pricing

InputWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it.
$10.00
Cached inputA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again.
$1.00
5 min cache writeA one-off charge for storing text so later requests can reuse it at the cheaper cached rate. This option keeps it for five minutes.
$12.50
1 hr cache writeA one-off charge for storing text so later requests can reuse it at the cheaper cached rate. This option keeps it for an hour, so it costs more than the five-minute one.
$20.00
OutputWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one.
$50.00
  • Adaptive thinking is always on.
USD per 1M tokensEvery price here is for one million tokens. A token is roughly three-quarters of a word, so a million tokens is about 750,000 words of text.Verified October 19, 2026Read from platform.claude.com. Click the check to open the source.

Available from

Amazon Bedrock—$10.00$50.001M
Anthropic—$10.00$50.001M
Azure—$10.00$50.001M
Claude Platform on AWS—$10.00$50.001M
Googleglobal$10.00$50.001M
Googleeu$11.00$55.001M
USD per 1M tokensEvery price here is for one million tokens. A token is roughly three-quarters of a word, so a million tokens is about 750,000 words of text.Checked October 22, 2026Third-party rates from openrouter.ai, not the lab’s own list price.

Benchmarks

Coding

SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better.
80.3%
#2 of 24
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better.
95.5%
#2 of 70
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better.
86.6%
#3 of 19
SWE-Bench MultimodalMultimodal coding — Real bug reports that arrive with pictures attached — a screenshot, a mockup, a page rendering wrongly — so the AI has to read the image as well as the code to work out what to fix. Higher is better.
54.1%
#3 of 3
DeepSWE 1.1Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better.
70%
#12 of 36
DeepSWE 1.0Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. Higher is better.
66.1%
#2 of 6
Next.js EvalsNext.js coding — Vercel's open eval of how well AI coding agents build and migrate real Next.js apps — measured as the share of tasks the agent completes successfully. Higher is better.
77%
#13 of 33

Terminal & CLI

Frontier-Bench v0.1Agentic computer work — A hard, ever-evolving set of real computer tasks — coding, system administration, data work, and more — that an AI agent has to complete on its own. Run by the Harbor / Laude Institute team as the successor to Terminal-Bench (v0.1 is the first release of the task set). The score is the share of tasks solved. Higher is better.
33.8%
#3 of 9
Terminal-Bench 4.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Version 4.0 recalibrated how much time, CPU and memory each task gets, removed eight tasks and fixed nineteen, so fewer runs fail for reasons that have nothing to do with the model. Scores are not comparable with earlier versions. Higher is better.via Terminal-Bench
44.55%
#10 of 29
Terminal-Bench 2.1Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better.
88%
#8 of 36
Terminal-Bench-Science 0.1Agentic scientific computing — The same command-line setup as Terminal-Bench, pointed at scientific work: the AI has to drive research tooling and computational workflows through to a result, rather than administer a machine. Version 0.1 is the first release of the task set, and scores run lower than on the general board. Higher is better.
24.7%
#7 of 7

Agentic & tool use

BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better.via BenchLM
86.9%
#9 of 31
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better.
85%
#2 of 29

Reasoning & science

Humanity's Last ExamMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better.via BenchLM
64.5%with tools
#3 of 53
ARC-AGI-2Abstract reasoning — Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better.via BenchLM
89.2%
#9 of 41

Knowledge work

GDPval-AAKnowledge work — Measures how well the AI does economically valuable knowledge work, judged against human experts. Shown as a rating (like a chess Elo) — higher is better.
1932
#1
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo)
1760
#5 of 21

Healthcare

HealthBench ProfessionalHealth — Realistic health conversations graded against detailed rubrics written by physicians — can the AI respond the way a careful medical professional would? Higher is better.
63.3%
#1
MedScribeMedical admin work — Can the AI support doctors with their administrative work, like notes and paperwork? Created by Vals AI. Higher is better.
88.52%
#2 of 5

Robustness

BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better.
56%
#32 of 98
Gray Swan IPIPrompt injection robustness — Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. Gray Swan's indirect prompt injection benchmark measures how often such an attack succeeds when the attacker gets a single try. Lower is better.
0.4%k = 1
2.3%k = 10
2.8%k = 15
k = 1
#2 of 13
k = 10
#2 of 13
k = 15
#3 of 14

Community preference

threejsevalCommunity preference (Three.js) — Every model gets the same prompt — "the Eiffel Tower", "a glass fishbowl", "a robot arm picking toys into a box" — and builds a 3D scene in Three.js. Real people then see two scenes side by side, names hidden, and vote for the one they prefer. The votes become a chess-style Elo rating on threejseval.com, averaged across all the prompts. It measures whether the scene looks and moves right to a human eye, not whether the code passes a test. Higher is better.
1511
#14 of 29

Source: BenchLM, retrieved 28 September 2026 · Source: Terminal-Bench, retrieved 30 August 2026. Other sources are identified on the linked benchmark pages.

Compare Claude Fable 5 with

Claude Fable 5

Suggested comparisons

Claude Fable 5vsClaude Haiku 5.5Claude Fable 5vsGPT-6.1 SolClaude Fable 5vsGemini 4 ArgonClaude Fable 5vsMuse Spark OpenClaude Fable 5vsGrok 4.7Claude Fable 5vsDeepSeek-V4.1-Flash-1017Claude Fable 5vsMistral Large 4Claude Fable 5vsKimi K3Claude Fable 5vsGLM-5.3-FlashClaude Fable 5vsQwen3.8-Max-0902Claude Fable 5vsNemotron 3.5 Super VL

About

Claude Fable 5 was announced on June 9, 2026 as Anthropic's generally available Mythos-class model. It shared its underlying model with Claude Mythos 5, with additional safeguards for wider use. Anthropic positioned Fable for long software-engineering tasks, analytical work and multimodal reasoning.

At launch, some requests could be routed to Claude Opus 4.8 when the safeguards intervened. This means an evaluation of the deployed Fable service could include fallback behavior as well as the main model. Fable and Mythos were introduced together with different access conditions, rather than as interchangeable public API choices.

Frequently asked questions

Claude Fable 5 was released by Anthropic on Tuesday, Jun 9 2026.

All Anthropic releases

30 tracked