# Claude Sonnet 4.6 Claude Sonnet 4.6 is an Anthropic model built for coding, computer use, document analysis, and agent planning. It expands on Sonnet 4.5 with improvements in instruction following and reasoning across long inputs. Its published context window is 1,000,000 tokens. The advertised input rate is $3.00 per million input tokens. The advertised output rate is $15.00 per million output tokens. - Model identifier: ` anthropic/claude-sonnet-4-6 ` - Creator: Anthropic - Model type: Text generation - Canonical model page: [Claude Sonnet 4.6]() - API availability: Listed as serving through Impossibl. - Creator release date: 2026-02-17 ## Capabilities and limits - Context window: 1,000,000 tokens - Maximum output tokens: Not published in the public catalog; this is separate from the context window. - Inputs: text, image - Outputs: text - Tool calling: Not published ## API access Use ` anthropic/claude-sonnet-4-6 ` as the model identifier through the Impossibl API. Published API endpoints: - ` /v1/chat/completions ` [API quickstart]() · [Models and billing documentation]() ## Published API prices All rates are in USD. An unpublished rate is unknown, not zero. Workspace-specific pricing, taxes, balance-purchase fees, billing adjustments, and tool charges can affect the amount paid. | Usage | Published base rate | Billing unit | | --- | --- | --- | | Input | $3.00 | per million input tokens | | Output | $15.00 | per million output tokens | | Cache read | $0.30 | per million cached input tokens | | Cache write (default retention) | $3.75 | per million tokens | | Cache write (one-hour retention) | $6.00 | per million tokens | | Audio input | Not published | per million audio input tokens | ## Independent benchmarks Source: [Artificial Analysis](). Last checked: 2026-09-11T10:50:14.034Z. Configurations are evaluated separately. API scores retain their original units. Unmeasured scores are omitted. ### Claude Sonnet 4.6 (Non-reasoning, High Effort) Source: [Evaluation source]() | Benchmark | Score | | --- | --- | | Intelligence Index | 24.69 | | Omniscience Index | -3.55 | | GPQA | 79.90% | | Humanity’s Last Exam | 13.35% | | CritPt | 0.86% | | Terminal-Bench Hard | 46.21% | | AA-LCR | 68.33% | | Multilingual LCR | 20.00% | | Harvey LAB | 84.44% | | MMMU-Pro | 70.58% | | IFBench | 41.16% | | Tau2 | 79.53% | | intelligence index (API) | 24.7 | ### Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) Source: [Evaluation source]() | Benchmark | Score | | --- | --- | | Intelligence Index | 30.45 | | Omniscience Index | 12.22 | | GPQA | 87.47% | | Humanity’s Last Exam | 33.60% | | SciCode | 50.12% | | CritPt | 3.14% | | Terminal-Bench v2.1 | 71.16% | | Terminal-Bench v4.0 | 3.03% | | Terminal-Bench Hard | 53.03% | | AA-LCR | 80.00% | | Multilingual LCR | 24.44% | | Harvey LAB | 85.97% | | GDPval Elo | 1295.41 | | GDP.pdf All-pass | 15.80% | | AutomationBench Partial | 20.10% | | AnalystAgent | 20.00% | | MMMU-Pro | 73.29% | | IFBench | 56.60% | | Tau2 | 75.73% | | Tau Banking | 34.43% | | APEX-Agents | 28.02% | | ITBench | 39.80% | | coding index (API) | 63 | | agentic index (API) | 33.1 | | intelligence index (API) | 30.5 | ### Claude Sonnet 4.6 (Non-reasoning, Low Effort) Source: [Evaluation source]() | Benchmark | Score | | --- | --- | | Intelligence Index | 23.30 | | Omniscience Index | -2.10 | | GPQA | 79.70% | | Humanity’s Last Exam | 11.21% | | CritPt | 0.86% | | Terminal-Bench Hard | 42.42% | | AA-LCR | 69.33% | | Multilingual LCR | 20.56% | | MMMU-Pro | 69.19% | | IFBench | 42.38% | | Tau2 | 78.95% | | intelligence index (API) | 23.3 | Full benchmark data and breakdowns: [Complete benchmark snapshot](). ## Availability and retirement The model is listed as serving. No retirement is announced in the current public catalog. ## Frequently asked questions ### What is Claude Sonnet 4.6? Claude Sonnet 4.6 is an Anthropic model built for coding, computer use, document analysis, and agent planning. It expands on Sonnet 4.5 with improvements in instruction following and reasoning across long inputs. Sources: [Claude Sonnet 4.6 release announcement](); [Claude Sonnet 4.6 model documentation]() ### How much does Claude Sonnet 4.6 cost? Claude Sonnet 4.6 costs $3.00 per million input tokens and $15.00 per million output tokens through the Impossibl API. Cache reads cost $0.30 per million tokens; cache writes cost $3.75 per million tokens; cache writes with one-hour retention cost $6.00 per million tokens. All rates are in USD. ### What is the context length of Claude Sonnet 4.6? Claude Sonnet 4.6 has a 1,000,000-token context window through the Impossibl API. ### What inputs and outputs does Claude Sonnet 4.6 support? Claude Sonnet 4.6 accepts text and image as input. Claude Sonnet 4.6 returns text. ### How do I use Claude Sonnet 4.6 through an API? Claude Sonnet 4.6 is available through the Impossibl API. Use anthropic/claude-sonnet-4-6 as the model identifier with /v1/chat/completions. The quickstart explains how to create an API key and send your first request. ### When was Claude Sonnet 4.6 released? Anthropic released Claude Sonnet 4.6 on February 17, 2026. Sources: [Creator release announcement]() ## Sources and related links Prices, limits, capabilities, endpoints, and availability come from the public Impossibl catalog. Creator documentation supplies descriptions and release dates where available; it does not override the published API offering. - [Public Impossibl model catalog]() - [Full model directory as JSON]() - [Claude Sonnet 4.6: model details and pricing]() - [Model directory]() - [Claude Sonnet 4.6 release announcement]() - [Claude Sonnet 4.6 model documentation]()