Anthropic / anthropic/claude-fable-5.2

Claude Fable 5.2 - access through LLMTR

Claude Fable 5.2 is Anthropic's most capable widely released model for long-running agent sessions, multistep research, and hard work in large codebases. It offers a 1M context window, 128K max output, always-on adaptive thinking, vision input, tool calling, and prompt caching. Effort is configurable from low through max and defaults to high. It costs less than Fable 5.1. `tool_choice` accepts only `auto` and `none`. Some cybersecurity and biology requests may be declined; a declined response can carry empty or partial content, so do not use the content when `content_filter` or `refusal` comes back.

Technical specifications

Canonical IDanthropic/claude-fable-5.2
ProviderAnthropic
Context window1,000,000 tokens
OperationsCHAT_COMPLETIONS
Modalitiestext, image

Pricing

An 8% platform margin applies to credit top-ups; model usage prices are not separately marked up.

OperationMetricUnitPrice
CHAT_COMPLETIONSINPUT_TEXTPER_1M_TOKENS$8.00
CHAT_COMPLETIONSCACHE_READPER_1M_TOKENS$0.200000
CHAT_COMPLETIONSOUTPUT_TEXTPER_1M_TOKENS$40.00
CHAT_COMPLETIONSCACHE_WRITEPER_1M_TOKENS$10.00
CHAT_COMPLETIONSCACHE_WRITEPER_1M_TOKENS$16.00

Example usage

With existing OpenAI SDK flows, change only the base URL and model identifier.

curl https://llmtr.com/v1/chat/completions   -H "Authorization: Bearer llmtr-your_key"   -H "Content-Type: application/json"   -d '{"model":"anthropic/claude-fable-5.2","messages":[{"role":"user","content":"Hello"}]}'

Related models