Claude Fable 5 Benchmark vs Gemini 3.1, GPT-5.5 and Grok 4

Claude Fable 5, released by Anthropic on June 9, 2026, is a Mythos-class AI model designed for autonomous coding and complex workflows, featuring a context window exceeding one million tokens. It outperforms its predecessors and competitors like GPT-5.5 and Gemini 3.1 Pro in specific benchmarks, notably scoring 80.3% on SWE-Bench Pro, highlighting its capability for long, multi-step engineering tasks, demonstrated by Stripe's rapid migration using the model. However, it does not lead in all areas, such as the GPQA Diamond for scientific reasoning, where Gemini 3.1 Pro ranks higher. Available via Claude API, AWS Bedrock, and GitHub Copilot, Claude Fable 5's pricing is $10 per million input tokens and $50 per million output tokens, making it suitable for tasks requiring higher reliability despite the cost. It is particularly strong in analytical workflows, financial analysis, and agentic computer use, while its broad context window and capabilities make it ideal for large-scale document and codebase analysis.