Loading model…正在加载模型详情…
Work through complex code and research with Claude Fable 5, Anthropic's model for sustained reasoning and careful multi-step analysis, in chat or by API.
- From $4.20 in / $21.00 out per 1M tokensSave up to 58%
- OpenAI- and Anthropic-compatible API
- Reasoning · Coding · Research
What can I help you with?
Use it inClaude desktop app
What is Claude Fable 5?
Anthropic describes Claude Fable 5 as a Mythos-class model for advanced general use. Its focus is sustained reasoning across demanding coding and knowledge tasks, keeping goals, constraints and intermediate findings in view while developing an answer.
On UniRoute, send a message in the workspace or use the model ID claude-fable-5 through Chat Completions or Anthropic Messages. The API section provides request examples for both formats.
Fable is a separate model line within Claude. Anthropic positions Fable 5 for long-running coding, research and complex knowledge work; the published benchmark table below gives a view of the reported results.
What you can do
What Claude Fable 5 is built for.
Codebase reasoning
Analyze dependencies, plan migrations and review proposed changes with the wider project context in mind.
Research synthesis
Compare supplied sources, organize evidence and draft findings that distinguish assumptions from supported conclusions.
Detailed document analysis
Work through dense material, compare requirements and prepare structured review notes.
This page accepts text messages; file inputs need the API, and the current model configuration does not expose function calling, search or structured output.
Specs & variants
The variants you can call, with their options and prices.
| Variant | Model ID | Modes | Inputs | Output | Price |
|---|---|---|---|---|---|
| Claude Fable 5 | claude-fable-5 | Chat and text generation | Chat messages | Streaming; Thinking; Context window: Not stated on this page | $4.20 input / $21.00 output per 1M tokens |
Pricing
Priced per token. Input and output are billed separately.
| Usage | Official API | UniRoute | Save |
|---|---|---|---|
| Input | $10.00 | $4.20 | Save 58% |
| Output | $50.00 | $21.00 | Save 58% |
| Cache read | $1.00 | $0.42 | Save 58% |
| Cache write (5 min) | $12.50 | $5.25 | Save 58% |
| Cache write (1 hour) | $20.00 | $8.40 | Save 58% |
Official price: platform.claude.com · 2026-11-07
Prices updated 2026-11-08.
Pay as you go from a single USD wallet balance. You see the quote before each request.
API
Call Claude Fable 5 through OpenAI- or Anthropic-compatible endpoints.
curl 'https://api.uniroute.app/v1/chat/completions' \ -H 'Authorization: Bearer YOUR_API_KEY' \ -H 'Content-Type: application/json' \ -d '{ "model": "claude-fable-5", "messages": [ { "role": "user", "content": "Hello" } ], "stream": false}'curl 'https://api.uniroute.app/v1/messages' \ -H 'Authorization: Bearer YOUR_API_KEY' \ -H 'anthropic-version: 2023-06-01' \ -H 'Content-Type: application/json' \ -d '{ "model": "claude-fable-5", "max_tokens": 1024, "messages": [ { "role": "user", "content": "Hello" } ]}'Parameters
| Name | In | Description |
|---|---|---|
model | Body | Model ID to run. |
messages | Body | The conversation as a list of role and content messages. |
max_tokens | Body | Upper limit on generated tokens (Anthropic Messages). |
stream | Body | Set to true to receive the response as a stream. |
anthropic-version | Header | Protocol version header for the Anthropic Messages API. |
Guides
Benchmarks & ranking
Published scores next to the models they were compared with.
| Benchmark Area | Claude Mythos 5 / Fable 5 | Claude Mythos Preview | Claude Opus 4.8 | GPT 5.5 | Gemini 3.1 Pro |
|---|---|---|---|---|---|
| Agentic coding SWE-Bench Pro | 80.30% | 77.80% | 69.20% | 58.60% | 54.20% |
| Agentic coding FrontierCode Diamond | 29.30% | — | 13.40% | 5.70% | — |
| Knowledge work GDPval-AA | 1932 | — | 1890 | 1769 | 1314 |
| Knowledge work vision GDP.pdf | 29.80% | — | 22.50% | 24.90% | 16.70% |
| Spatial reasoning Blueprint-Bench 2 | 38.60% | — | 14.50% | 36.20% | 26.50% |
| Tool use AutomationBench | 17.40% | — | 15.50% | 12.90% | 9.60% |
| Computer use OSWorld-Verified | 85.00% | 85.40% | 83.40% | 78.70% | 76.20% |
| Legal Legal Agent Benchmark | 13.30% | — | 10.40% | 2.10% | 0.00% |
| Multidisciplinary reasoning Humanity’s Last Exam, no tools | 59.0%* | 56.80% | 49.80% | 41.40% | 44.40% |
| Multidisciplinary reasoning Humanity’s Last Exam, with tools | 64.5%* | 64.70% | 57.90% | 52.20% | 51.40% |
| Biology BioMysteryBench, hard | 46.1%* | 29.60% | 40.00% | — | — |
| Biology BioMysteryBench, human solved | 83.9%* | 82.60% | 80.40% | — | — |
| Agentic coding Terminal-Bench 2.1 | 88.0%* | — | 82.70% | 83.40% | 70.70% |
| Cybersecurity ExploitBench(Cap%) | 78.0%* | 69.00% | 40.00% | 34.00% | — |
| Health HealthBench Professional | 66.0%* | 64.70% | 56.90% | 51.80% | — |
Figures published by Anthropic from its own evaluations. Results depend on each benchmark’s configuration.
Use cases
Typical ways to put Claude Fable 5 to work.
Migration planning
Build a staged migration plan that tracks dependencies, constraints and checks before implementation.
Technical report drafts
Turn supplied research notes and technical material into a coherent report with open questions made explicit.
Business document review
Compare supplied reports or policy text and prepare a review draft grounded in their contents.
FAQ
Common questions about Claude Fable 5.
What is Claude Fable 5?
Anthropic describes Claude Fable 5 as a Mythos-class model for advanced general use. Its focus is sustained reasoning across demanding coding and knowledge tasks, keeping goals, constraints and intermediate findings in view while developing an answer.
How much does Claude Fable 5 cost?
From $4.20 input / $21.00 output per 1M tokens. Input and output are billed separately according to usage. The pricing table and worked examples show how the rates apply (prices updated 2026-11-08).
Which API formats can I use?
Use the model ID claude-fable-5 with Chat Completions or Anthropic Messages. The API section shows both request formats and their parameters.
What can I send in the chat box?
The workspace on this page accepts text messages. File inputs need the API; they cannot be attached in this chat box.
Does it support function calling?
The current UniRoute model configuration does not expose function calling for Claude Fable 5. Use it for conversation and analysis without assuming it can run application tools.
Is there a free trial or free credits?
No. Requests are paid from your wallet balance. You can read this page and review the prices without paying, and you need an account with funds to send a message.
Related models
Other versions, models from the same provider and alternatives.
