Claude Fable 5.1
Claude Fable 5.1 is Anthropic's generally available frontier model for demanding coding, research, and long-running agent workflows, with lower cache-read pricing and several API migration changes from Fable 5.
Platform: Replicate
Reasoning AICoding AssistantLong-horizon AgentsMultimodal AI
Input: $10 per million tokens; Cache read: $0.25 per million tokens; Output: $50 per million tokens
Commercial🚀Function Overview
A high-capability text-and-image input model with adaptive thinking, a 1M-token context window, and controls for long-running agent work.
Key Features
- Stable Claude API model ID claude-fable-5-1
- 1M-token context window and 128K-token maximum output
- Adaptive thinking is always on, with high effort as the documented API default
- Text and image input with text output
- Per-message effort, turn-scoped system messages, and progress display are available as beta controls
- Cache reads cost $0.25 per million tokens, below Fable 5's cache-read rate
Use Cases
- •Long-running repository-scale coding and difficult debugging
- •Multistep research and document-heavy analysis
- •Spreadsheet, presentation, and structured artifact creation
- •Agent workflows that need large context and visible progress updates
- •Defensive vulnerability discovery within Anthropic's safeguard boundaries
⚙️Input Parameters
prompt
stringText instructions and context for the model
image
stringOptional image input for multimodal analysis
system_prompt
stringSystem-level instructions; turn-scoped system messages are available in beta
effort
stringReasoning effort control for adaptive thinking
max_tokens
integerMaximum output length, up to 128K tokens
💡Usage Examples
Example 1
Input Parameters
{
"prompt": "Investigate this intermittent repository failure, identify the root cause, implement the smallest safe fix, and summarize the evidence.",
"system_prompt": "Work methodically. Preserve unrelated changes and verify every claimed fix.",
"effort": "high",
"max_tokens": 8192
}Output Results
I will first reproduce the failure and map the affected execution path.
I will distinguish the root cause from secondary symptoms, make the narrowest bounded change, and run the relevant tests.
The final summary will separate observed evidence, the implemented fix, verification results, and remaining risk.
Quick Actions
Technical Specifications
- Hardware Type
- Commercial Use
- Supported
- Pricing
- Input: $10 per million tokens; Cache read: $0.25 per million tokens; Output: $50 per million tokens
- Platform
- Replicate