Claude Fable 5
Anthropic's most capable widely released model, built for agents that run for hours rather than seconds. Adaptive thinking is always on and cannot be disabled, which is part of why it tops the Artificial Analysis Intelligence Index — and why its cost per task runs roughly triple GPT-5.6 Sol's for a one-point intelligence lead.
Pricing checked against provider documentation on . How we verify
claude-fable-5 reasoning Input
$10.00
/ 1M tokens
Output
$50.00
/ 1M tokens
Context
1M
1,000,000 tokens
Max output
128K
tokens per request
Pricing caveat: 5-minute cache writes $12.50/MTok, 1-hour $20/MTok. Requires 30-day data retention, so not available under zero-data-retention terms.
What it costs in practice
A workload of 10M input and 2M output tokens per month — roughly a moderately busy production assistant — runs $200.00/month on Claude Fable 5. Routing that same volume through the batch API brings it to $100.00. The next cheaper option, Claude Opus 5 , would cost $100.00.
Where Claude Fable 5 fits
- Multi-hour autonomous coding runs
- Large codebase migrations
- Hard scientific workflows
- High-stakes legal and medical analysis
Published benchmarks
| Benchmark | Score |
|---|---|
| Artificial Analysis Intelligence Index | 60 |
| Artificial Analysis Coding Agent Index | 77.2 |
Scores are self-reported by providers or drawn from independent indices. Treat them as directional and validate on your own evals.
Specifications
- API model ID
- claude-fable-5
- Provider
- Anthropic
- Model family
- Claude 5
- Released
- June 2026
- Knowledge cutoff
- January 2026
- Context window
- 1,000,000 tokens
- Max output
- 128,000 tokens
- Native reasoning
- Yes
- Relative latency
- high
- Input modalities
- text, image
- Output modalities
- text
- Open weights
- No
- Cached input
- $1.00 / 1M tokens
- Batch discount
- 50%
Frequently asked
How much does Claude Fable 5 cost?
Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens, with cached input at $1.00. Batch processing is 50% cheaper.
What is Claude Fable 5's context window?
Claude Fable 5 accepts up to 1,000,000 tokens of context and can generate up to 128,000 output tokens per request.
Is Claude Fable 5 a reasoning model?
Yes. Claude Fable 5 performs native chain-of-thought before answering. Those thinking tokens are billed at the output rate, so budget above the sticker price.
What is Claude Fable 5 best for?
Long-horizon agentic engineering. Anthropic's most capable widely released model, built for agents that run for hours rather than seconds. Adaptive thinking is always on and cannot be disabled, which is part of why it tops the Artificial Analysis Intelligence Index — and why its cost per task runs roughly triple GPT-5.6 Sol's for a one-point intelligence lead.