{ "model": "claude-sonnet-5", "context_tokens": 1000000, "max_output_tokens": 128000, "input_microusd_per_token": 4, "output_microusd_per_token": 10, "observed_on": "2026-09-27", "source": "https://platform.claude.com/docs/en/models/sonnet-5/whats-new-sonnet-5", "pricing_source": "https://platform.claude.com/docs/en/about-claude/pricing", "scope": "First-party global standard Messages; input bound covers one-hour cache creation. No inference_geo, service_tier, speed, server tools or other extra-fee request fields admitted by transport. Snapshot for review, not authority or a future tariff guarantee.", "observed_cli_max_tokens": 64000, "required_request_betas": [ "claude-code-20250219", "interleaved-thinking-2025-05-14", "thinking-token-count-2026-05-13", "context-management-2025-06-27", "prompt-caching-scope-2026-01-05", "mid-conversation-system-2026-04-07", "effort-2025-11-24" ], "cli_shape_evidence": "2026-09-27-sonnet5-runtime-proof.json" }