Claude Fable 5
AI models review
As of Aug 12, 2026, Claude Fable 5 is rated 1506 Elo on LM Arena (21,242 Arena votes).
Benchmark and price facts
- Vendor
- Anthropic
- License
- proprietary
- LM Arena text
- 1506 Elo (#1 of 20)
- LM Arena coding
- 1628 Elo (#5)
- Price (input)
- $10 / 1M tokens
- Price (output)
- $50 / 1M tokens
- Context window
- 1,000,000 tokens
- AA Intelligence Index
- 62.1 / 100
- AA Coding Index
- 76.5 / 100
- Output speed (AA)
- 67 tokens/s
- Human votes
- 21,242
Elo and votes from LM Arena; intelligence, coding and speed indices from Artificial Analysis; price and context window from OpenRouter. Real fetched values, attributed, never blended.
Is Claude Fable 5 worth it in 2026?
Claude Fable 5 holds the top text Elo of 1506 (rank 1 of 20) from about 21,000 human votes, making it the highest-rated model for general chat and conversation in this field. Its coding Elo of 1628 at rank 5 of 20 is strong but not the outright leader on that dimension. The 1,000,000-token context window supports very long document work. The Artificial Analysis Intelligence Index is 62.1 and Coding Index is 76.5, consistent with a well-rounded but coding-leaning capable model. Output speed is 67 tokens per second, which is moderate.
Users who prioritize top-ranked general conversation quality and chat, long-document work enabled by the 1M token context, and strong but not leading coding assistance.
Pricing is steep at 10 USD per 1M input tokens and 50 USD per 1M output tokens, making it a poor fit for high-volume or cost-sensitive use cases. Coding rank of 5 of 20 means there are four models ahead of it specifically for coding tasks.
AI-generated editorial synthesis of the source scores and sampled review themes, not a hands-on test. Reviewer experiences and allegations are not independently verified facts.
Rated near Claude Fable 5
Closest scores in ai models, on the same source.