claude-fable-5-1

claude-fable-5-1 is Anthropic's frontier-tier reasoning model in the Claude 5 family, priced at $10.00 per million input tokens and $50.00 per million output tokens. Prompt caching reduces repeated input to $0.25/1M — a 97.5% discount that makes long shared system prompts and document Q&A workloads significantly cheaper. It supports a 1M-token context window, vision inputs, function calling, and streaming. claude-fable-5-1 is a proprietary model available exclusively through Anthropic's API.

claude-fable-5-1 is priced at $10.00 per million input tokens and $50.00 per million output tokens, with prompt caching at $0.25/1M. The 5:1 output-to-input ratio reflects the sequential cost of autoregressive generation. At 1,000 input + 500 output tokens per call, the cost is ($10.00 × 1,000 + $50.00 × 500) ÷ 1,000,000 = $0.035 per call, or approximately $35.00 per 1,000 API calls. For workloads with long repeated system prompts, the $0.25/1M cached input rate makes the effective cost substantially lower — a 1M-token cached system prompt costs $0.25 per call at the cached rate versus $10.00 at the standard rate.

claude-fable-5-1's 1M-token context window is among the largest available from any production model, enabling full-codebase analysis, book-length document processing, and extended multi-turn conversations in a single API call. Median latency is approximately 1,800ms TTFT (Estimate). Throughput is approximately 68 tokens per second (Estimate). Maximum output is 65,536 tokens. As a reasoning model, it applies extended chain-of-thought processing before producing a final answer, which improves accuracy on complex multi-step tasks at the cost of higher latency compared to standard models.

claude-fable-5-1 is best suited to tasks requiring frontier reasoning capability with very long context: complex codebase analysis, multi-document research synthesis, advanced mathematical reasoning, and agentic workflows requiring sustained multi-step execution. Its 1M-token context window and reasoning capability make it particularly well-suited for tasks that require both breadth (large context) and depth (complex reasoning). For cost-sensitive workloads, claude-sonnet-5-5 or claude-haiku-5-5 offer lower per-token rates at reduced capability.

claude-fable-5-1 sits at the frontier tier of Anthropic's Claude 5 lineup alongside claude-mythos-5-1, with both priced at $10/$50. The choice between them depends on the specific task profile — evaluate both on your workload before committing to production. Its main competitors are frontier reasoning models from OpenAI and Google at similar price points. Anthropic's Constitutional AI training approach is reflected in strong safety alignment and instruction following, which is relevant for enterprise deployments with strict content requirements.