Anthropic Long-context knowledge work
Claude Fable 5.1
Reserve deep reasoning for work that must remain coherent over time.
Claude Fable 5.1 is an Anthropic route for demanding knowledge work, coding projects and multi-step agent tasks. TokenHot lists text, image and document inputs, a 1M context value, reasoning and tool capabilities. Use it for cross-file analysis and sustained workflows where reliable completion matters more than speed or a minimal request cost.
- Reasoning
- Text + image + document input
- Long context
The route name and catalog capabilities do not imply a particular Anthropic product tier or a hosted workspace. Verify the exact endpoint, accepted document formats and tool contract before making a production promise.
AI Chat
Send a prompt and see the response here.
claude-fable-5-1Start a conversation with this model and ask follow-up questions.
USD /1M tokens
Pricing
| Group | Input | Output | Cache Read | Cache Write |
|---|---|---|---|---|
| default | $10 | $50 | $0.25 | $20 |
| claude-0.34 | $3.4 | $17 | $0.085 | $6.8 |
Public list prices are shown in USD. Final charges may vary by account group and usage tier.
Overview
What is Claude Fable 5.1?
Fable 5.1 is suited to tasks that need several source types to stay aligned: a project specification and code, a PDF packet and a decision, or a sequence of tool calls with a durable final artifact.
Because the work is expensive to redo, define the final deliverable and review points before sending a large request. Ask the model to distinguish sourced facts, assumptions and recommendations.
Keep claims traceable
Name documents and dates, then require section references and a separate unknowns list.
Coordinate many files
Give the model a map of the work and ask for checkpoints, dependencies and reviewable outputs.
Protect the side effects
Return structured arguments and stop states. Let your application decide whether a tool call is authorized.
From brief to deliverable
Put Claude Fable 5.1 to work
Strategy and research teams
Make a research packet decision-ready
Compare a long set of documents and produce a memo that keeps disagreements and missing evidence visible.
- Tools you need
- Document retrieval, text extraction and a human reviewer.
- Keep in mind
- The model cannot verify that a source is authentic, current or complete without external evidence.
View a task brief
Use the supplied dated pdfs or text files, comparison questions and decision criteria. to plan make a research packet decision-ready. Return a source-linked memo with recommendation and open questions., state the limits of the evidence and identify the checks required before accepting the result. Do not claim that an output or interaction was verified without direct evidence.
Product and engineering teams
Coordinate a multi-file project brief
Turn a specification, existing code and meeting notes into a staged implementation plan with owners and acceptance tests.
- Tools you need
- Repository search, issue tracking and test tooling for verification.
- Keep in mind
- Do not turn inferred ownership or dates into commitments without confirmation.
View a task brief
Use the supplied a product brief, repository map, notes, constraints and delivery milestones. to plan coordinate a multi-file project brief. Return a dependency map, work plan and acceptance matrix., state the limits of the evidence and identify the checks required before accepting the result. Do not claim that an output or interaction was verified without direct evidence.
Operations and platform teams
Build a document-processing agent contract
Design the input schema, extraction steps and review queue for a document-heavy workflow.
- Tools you need
- Document storage, OCR or parsing and a workflow engine.
- Keep in mind
- Document input support does not guarantee OCR quality or access to every file format.
View a task brief
Use the supplied document types, fields, validation rules, escalation policy and output schema. to plan build a document-processing agent contract. Return a schema-first agent design with review states., state the limits of the evidence and identify the checks required before accepting the result. Do not claim that an output or interaction was verified without direct evidence.
Capabilities in context
Key features
| Feature | What you get | Why it matters |
|---|---|---|
| Context capacity | 1M tokens shown in the catalog | Use the window for connected documents, but keep retrieval, file validation and source citations explicit. |
| Document input | Text, image and document inputs are listed | Confirm accepted file formats and size limits for the endpoint before uploading a packet. |
| Reasoning depth | Reasoning is listed by the catalog | Use checkpoints and budgets so long work remains observable and interruptible. |
| Tool contract | Function calling and tool use are listed | Validate every argument and keep credentials and side effects in your application. |
| Version identity | TokenHot model ID: claude-fable-5-1 | Record the exact route and protocol for reproducible runs. |
Modalities and capabilities come from TokenHot's current catalog. Confirm document formats, provider identity and protocol-specific limits before integrating.
Audience & fit
Where Claude Fable 5.1 fits
Research teams with complex source packets
Use the route to organize documents and disagreements into an accountable memo, not to replace source review.
Product and engineering leads
Keep requirements, code and decisions together while using checkpoints to make a long plan reviewable.
Workflow platform builders
Design explicit extraction and review states so document inputs do not become silent automation.
Cost context
Plan the cost of a useful result
Count document preparation
Parsing, OCR, retrieval and human review can dominate the model charge. Track those steps alongside token usage.
Keep a source packet index
A manifest with names, dates and hashes makes reruns and disagreement analysis much cheaper.
Escalate high-impact decisions
Use a reviewer and an evidence threshold for legal, financial, security or operational decisions.
Use the current pricing table for available rates and account groups. Model IDs and prices come from TokenHot’s catalog; provider pricing and subscriptions are separate.
API
Code examples
curl --location --request POST 'https://api.tokenhot.ai/v1/messages' \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data-raw '{
"model": "claude-fable-5-1",
"messages": [
{
"role": "user",
"content": "What can you do?"
}
],
"max_tokens": 100
}'FAQ
Frequently asked questions
What is Claude Fable 5.1 best suited for?
Claude Fable 5.1 is suited to the input, output, and capability types listed on this page. Test production workloads before choosing it for a critical system.
How is Claude Fable 5.1 priced?
Pricing is shown in the pricing section above. Actual cost depends on usage volume, account group, and the request payload.
How can I call Claude Fable 5.1?
Use the model ID shown above with one of the supported API protocols. Code examples are provided when usage data is available.
What does document input mean on this page?
The catalog lists document input for the route. File formats, size, parsing and endpoint limits still need to be verified for your integration.
Can Fable 5.1 run an agent on its own?
No. Your application supplies tools, permissions and execution. The model can propose plans or arguments but cannot claim side effects it did not perform.
How should I use a 1M context responsibly?
Keep a source manifest, retrieve relevant material, require citations and set a clear output contract. Large context is not a substitute for evidence discipline.
How is this route different from Claude Opus 5.5?
They are separate TokenHot model IDs. Compare them on the long-running knowledge and coding tasks you actually need instead of relying on names alone.
Get started
Build with Claude Fable 5.1
API