Complete guide to using Claude Fable 5 API
claude-fable-5. The model is designed for frontier coding, research, vision, knowledge work, and complex agent workflows.The official launch positions Fable 5 above prior generally available Anthropic models, with targeted safeguards that route some sensitive requests to Claude Opus 4.8.
01
Anthropic says Claude Fable 5 brings capabilities beyond its previous generally available models. The launch materials emphasize broad gains across coding, knowledge work, vision, scientific research, and complex tasks.
02
The official charts highlight strong results on agentic coding evaluations, including SWE-bench Pro and FrontierCode. This makes the model relevant for repository-level debugging, migration planning, and autonomous code workflows.
03
Anthropic calls out vision and scientific research as key capability areas. The release includes scientific evaluation visuals such as viral capsid property prediction, making the landing page assets directly tied to the model.
04
Fable 5 is described as a Mythos-class model made safe for general use. Anthropic says safeguards route some topics to Claude Opus 4.8 and trigger in less than 5% of sessions on average.
05
The official release includes cyber and misuse evaluations, so production teams can reason about how Fable 5 is intended to be used and where additional policy checks may be needed.
Debug cross-file changes, plan migrations, explain unfamiliar code, and produce structured implementation steps.
Turn long research notes, specs, contracts, and internal docs into clear summaries and decisions.
Analyze technical material, benchmark evidence, scientific notes, and vision inputs with detailed reasoning.
Run multi-step planning and tool-backed assistant tasks for complex assistant workflows.
Fable 5 is the newer Mythos-class model. Opus 4.8 remains the fallback target for selected safeguard routing and an established Claude model option.
| Capability | Claude Fable 5 | Claude Opus 4.8 |
|---|---|---|
| Model ID | claude-fable-5 | claude-opus-4-8 |
| Class | Mythos-class model for general use | Claude Opus model |
| Best fit | Frontier coding, research, vision, complex tasks | Long-context Claude workflows and safeguard routing |
| Pricing | Input 2142.8 / output 10714 credits per 1M tokens | Input 800 / output 4000 credits per 1M tokens |
Source: Anthropic official Claude Fable 5 and Claude Mythos 5 launch post, published June 9, 2026.
Create a PoYo API key
Get API Key
Set the model ID
Use claude-fable-5 as the model ID in your request.
Open API docs
View API docs
| Model | Input | Output |
|---|---|---|
| Claude Fable 5 | $8 / 1M tokens 2142.8 credits | $40 / 1M tokens 10714 credits |
claude-fable-5.