Tool support requires the appropriate API integration; a supported tool is not automatically active in every chat.
Supported API features and tools
Text and image input; text output
Adaptive thinking with configurable effort
Messages API tool-use workflows with an appropriate integration
Prompt caching with 5-minute and 1-hour write options
Asynchronous Message Batches processing
Before you choose
No native audio, video, or image output; accepting images does not make this an image generator.
Fable 5 is active legacy; Fable 5.1 is the current Fable model.
Thinking is always on, and hidden summaries still consume billed output tokens.
Fable 5 and the separate invitation-only Mythos 5 offering do not share access rights merely because their specifications align.
Adaptive thinking and effort
lowmediumhigh · defaultxhighmax
Adaptive thinking is always on, with high effort as the default. All five effort levels are supported. Lower effort does not turn thinking off; disabled, between_tools, and manual enabled/budget_tokens are rejected. Bound the task and reserve enough max_tokens for thinking and the answer before raising effort.
Anthropic lists this model as active legacy; released June 9, 2026. Retirement is not sooner than June 9, 2027. This is an availability commitment, not a scheduled retirement.
The displayed cutoff is Anthropic’s reliable knowledge cutoff, published to month precision. A cutoff does not replace checking current facts or supplied evidence.
Compare the current Fable 5.1 guide before migrating: cache-read pricing and thinking/tool integration behavior differ.
The documented maximum output is 128,000 tokens. Do not infer support for another model’s extended-output beta.
Put it to work
Start with a more useful prompt.
Original examples from EZ Ai Assist. Adapt these to your task and the features available in your workspace.
Workflow 01
Test the logic of a technical argument
Look for counterexamples without manufacturing evidence.
Analyze this technical argument using the supplied definitions, assumptions, and evidence. Identify unsupported inferences, hidden premises, and boundary cases that could invalidate the conclusion. For each issue, cite the exact step and describe a concrete check or counterexample to investigate. Separate a demonstrated contradiction from a possible concern, and do not claim a proof you have not established.
Workflow 02
Build a review map for a complex program
Connect dependencies and risks to explicit source material.
From these program documents, map the dependencies among milestones, teams, and external approvals. Cite the section establishing each dependency. Identify circular dependencies, missing owners, and milestones with no measurable completion criteria. Prioritize clarification questions and propose review checkpoints without inventing dates, resources, or commitments.
Workflow 03
Compare competing explanations for an outcome
Keep causation claims tied to observations and testable predictions.
Compare the proposed explanations for this outcome using only the supplied observations and study notes. List what each explanation predicts, which observations support or challenge it, and which factors remain confounded. Suggest a feasible discriminating test for unresolved alternatives. Do not convert correlation into causation or treat missing evidence as proof against an explanation.
Developer reference
Anthropic API pricing
These are Anthropic API reference prices, not EZ Ai Assist subscription prices.
Standard rates apply across the full 1M-token context window; no separate long-context surcharge is listed for this model.
Batch processing discounts input and output by 50%. Cache writes and reads have distinct rates; calculate from actual usage rather than applying a cache-read discount to the entire request.
Thinking tokens are billed as output even when their text is hidden. Tool charges and optional US-only inference (1.1× token pricing) can also affect the total.
Cache reads cost $1 per million tokens on Fable 5, compared with $0.25 on Fable 5.1. Both have the same listed $10/$50 base input/output rates.
Verify current provider pricing before estimating spend. These rates do not describe an EZ Ai Assist subscription.
Common questions
A few things worth knowing.
Why is Fable 5 marked Legacy?
Anthropic lists it as active legacy because Fable 5.1 is the current Fable model. The older model remains available in the provider API; legacy is not synonymous with retired.
Should I migrate to Fable 5.1?
Compare it on your actual tasks, including costs and integration behavior. The newer model has lower cache-read pricing and a newer cutoff, but request compatibility still needs checking.
Can I disable adaptive thinking?
No. Thinking is always on. Change supported effort levels to tune work and cost, and do not confuse hiding a thinking summary with eliminating the underlying output tokens.
Are cache prices identical to Fable 5.1?
No. Fable 5 cache reads are $1 per million tokens versus $0.25 for Fable 5.1. Base input/output and both write-duration rates are unchanged, so savings depend on actual cache usage.
Does Fable 5 include Mythos access?
No. Mythos 5 is a separate invitation-only offering for authorized participants. Matching specifications do not establish availability or permission to use that offering.
Can it use the Opus and Sonnet extended-output beta?
The reviewed Fable 5 reference establishes 128,000 output tokens and does not list that extended-output beta for this model. Use model-specific limits rather than borrowing another family’s capabilities.
Will it shut down on June 9, 2027?
No shutdown on that date is announced by the cited commitment. Anthropic says retirement will not occur sooner. Provider lifecycle and API prices also remain separate from EZ Ai Assist plans and availability.
Check the source
Official documentation
Specifications and API prices checked on . Example prompts and workflow advice are editorial guidance from EZ Ai Assist.