LLM Reference

Claude Sonnet 4.6

Released
2026-02-17
Last refreshed
2026-06-15
Status
Researched 6d ago
ProprietaryCommercial use: conditionalMultimodalCodingRAGAgentsLong contextVisionClassificationJSON / Tool useHighlight

Claude Sonnet 4.6 is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.

Use it for

  • Teams evaluating coding, rag, and agents
  • Workloads that can use a 1m context window
  • Buyers comparing 4 tracked provider routes

Do not use it for

  • Workloads where another current model has stronger sourced task evidence
Specifications
Released
2026-02-17
Context
1m
Max output
64,000
Architecture
Decoder Only
Knowledge cutoff
2025-08
Specialization
general
Openness
Proprietary
License
ProprietaryCommercial use: conditional
Training
Fine-tuned
Created by

Developing safe and ethical AI systems.

San Francisco, California, United States
Founded 2021
Website
Pricing
Output / 1M
$15.00
Input / 1M
$3.00

Cheapest of 6 routes · Anthropic · cache read $0.300

About

Claude Sonnet 4.6 is Anthropic's best combination of speed and intelligence. Proprietary decoder-only model with 1M-token context, 64K max output, multimodal vision, extended thinking, and function calling. Available via Anthropic API, AWS Bedrock, GCP Vertex AI, and OpenRouter at $3/1M input and $15/1M output tokens.

Claude Sonnet 4.6 is a proprietary model in the Claude 4.6 family. The structured metadata tracks a 1m-token context window, multimodal input, reasoning, function calling, tool use, structured outputs, and code execution. This page tracks provider routes through OpenRouter, Anthropic, AWS Bedrock, and 3 more, with the cheapest tracked route listed at $3 input and $15 output per 1M tokens. Headline tracked benchmarks include SWE-bench Verified 79.6, Terminal-Bench 2.0 59.1, and SWE-bench Multilingual 75.9.

Top use-case fit: coding, agents, and build tasks

Coding

Q/$ D

3 relevant benchmarks in the decision map.

RAG

Included by capability and metadata signals in the decision map.

Agents

Q/$ D

3 relevant benchmarks in the decision map.

Provider price ladder

Compare all 6

Compare API pricing across 4 providers for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MBatch in / outCacheRoute
Anthropic$3.00$15.00$1.50 / $7.50read $0.300 / 5m $3.75 / 1h $6.00
Serverless
AWS Bedrock$3.00$15.00--
Serverless
GCP Vertex AI$3.00$15.00--
Serverless
Microsoft Foundry$3.00$15.00-read $0.300 / 5m $3.75 / 1h $6.00
Serverless

Available via routers & gateways(16)

AIRouter

Router

Commercial LLM router that analyzes incoming requests and routes to the optimal model for cost/quality/latency via a drop-in OpenAI-compatible API, with a privacy-preserving embedding mode that avoids sending prompt content.

Passthrough + feeAnthropicGCP Vertex AI

Amazon Bedrock Intelligent Prompt Routing

Router

AWS Bedrock's native intelligent prompt router that routes prompts between Anthropic Claude model tiers (Haiku/Sonnet) based on predicted task complexity, with no extra per-routing charge.

PassthroughAWS Bedrock

Azure AI Foundry Model Router

Router

Microsoft Azure AI Foundry's native model router that uses a trained ML model to route each prompt in real time to the optimal Azure-hosted model, with Balanced/Cost/Quality mode selection and automatic failover.

PassthroughMicrosoft Foundry

Helicone

Gateway

Observability-first AI gateway with routing, caching, rate limiting, and request tracing; Apache 2.0 open-source core with a managed hosted tier for logging and analytics.

SubscriptionAnthropicMicrosoft FoundryGCP Vertex AI

Kong AI Gateway

Gateway

Multi-LLM AI gateway built on Kong Gateway 3.x, adding semantic routing, load balancing, guardrails, and MCP traffic analytics as plugins over Kong's existing API management platform.

SubscriptionAnthropicGCP Vertex AIMicrosoft Foundry

LiteLLM

Gateway

Open-source Python SDK and proxy server that unifies 100+ LLM APIs behind a single OpenAI-compatible interface, with load balancing, cost tracking, and configurable failover.

Free OSSAnthropicGCP Vertex AIMicrosoft Foundry

Capabilities

VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode ExecutionPrompt CachingBatch API

Benchmark peer barsfor Coding

Benchmark scores(18)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionSource
SWE-bench Verified79.6SWE-bench Verifiedhttps://www.nxcode.io/resources/news/claude-sonnet-4-6-complete-guide-benchmarks-pricing-2026
Terminal-Bench 2.059.1Terminal-Bench 2.0https://www.datacamp.com/blog/claude-sonnet-4-6
SWE-bench Multilingual75.9SWE-bench Multilingualhttps://www.nxcode.io/resources/news/claude-sonnet-4-6-complete-guide-benchmarks-pricing-2026
Google-Proof Q&A89.9diamondhttps://www-cdn.anthropic.com/78073f739564e986ff3e28522761a7a0b4484f84.pdf
MMLU PRO87.3https://huggingface.co/spaces/TIGER-Lab/MMLU-Pro
τ-bench87.5τ-benchhttps://taubench.com/
MultiChallenge57.1MultiChallengehttps://labs.scale.com/leaderboard/multichallenge
Chatbot Arena1459.0https://arena.ai/leaderboard
MMMU Pro75.6official Anthropic system card, adaptive thinking, max effort, with image cropping toolhttps://www.anthropic.com/news/claude-sonnet-4-6
SWE-rebench60.7pass@1 (best of 5 runs)https://swe-rebench.com/leaderboard
AIME 202594.0AIME 2025 (accuracy)https://automatio.ai/models/claude-sonnet-4-6
ARC-AGI-258.3llm-stats shows 0 (accuracy%)https://llm-stats.com/benchmarks/arc-agi-v2
Humanity's Last Exam33.2HLE without tools (accuracy)https://automatio.ai/models/claude-sonnet-4-6
HumanEval98.0HumanEval (pass@1)https://automatio.ai/models/claude-sonnet-4-6
LiveCodeBench80.0LiveCodeBench score (accuracy)https://automatio.ai/models/claude-sonnet-4-6
MCP-Atlas61.3llm-stats shows 0 (accuracy%)https://llm-stats.com/benchmarks/mcp-atlas
Massive Multitask Language Understanding89.3MMLU (accuracy)https://automatio.ai/models/claude-sonnet-4-6
Massive Multi-discipline Multimodal Understanding83.6MMMU (accuracy)https://automatio.ai/models/claude-sonnet-4-6

Migration checks

Show all 79 popular comparisonssorted by 7-day search impressions
Claude Sonnet 4.6 vs Kimi K2.64KClaude Sonnet 4.6 vs Composer 2.53KClaude Sonnet 4.6 vs GPT-5.5 Pro3KClaude Sonnet 4.6 vs Gemini 3.5 Flash2KClaude Sonnet 4.6 vs GLM-5.12KClaude Sonnet 4.6 vs Claude Opus 4.71KClaude Sonnet 4.6 vs Qwen3.6-27B1KClaude Sonnet 4.6 vs GPT-5.51KClaude Sonnet 4.6 vs Kimi K2 Thinking972Claude Sonnet 4.6 vs Gemini 2.5 Pro783Claude Sonnet 4.6 vs Xiaomi MiMo-V2.5-Pro645Claude Sonnet 4.6 vs Qwen3.6-35B-A3B587Claude Sonnet 4.6 vs Kimi K2.5504Claude Sonnet 4.6 vs Hunyuan Hy3 Preview493Claude Sonnet 4.6 vs Gemini 2.5 Flash483Claude Sonnet 4.6 vs DeepSeek R1460Claude Sonnet 4.6 vs Step 3.5 Flash431Claude Sonnet 4.6 vs Claude 3.7 Sonnet410Claude Sonnet 4.6 vs Xiaomi MiMo-V2.5355Claude Sonnet 4.6 vs Grok-3349Claude Sonnet 4.6 vs Qwen3.5-27B286Claude Sonnet 4.6 vs Ling-2.6-1T243Claude Sonnet 4.6 vs GPT-4o-mini Search Preview207Claude Sonnet 4.6 vs DeepSeek V3201Claude Sonnet 4.6 vs Llama 3 70B Instruct181Claude Sonnet 4.6 vs GPT-5.4173Claude Sonnet 4.6 vs Llama 3.1 70B Instruct149Claude Sonnet 4.6 vs Qwen3-235B-A22B137Claude Sonnet 4.6 vs Claude Opus 4.5135Claude Sonnet 4.6 vs Qwen3.5-35B-A3B122Claude Sonnet 4.6 vs GPT-5.2 Codex119Claude Sonnet 4.6 vs Qwen3.5-397B-A17B111Claude Sonnet 4.6 vs GLM-5 Turbo103Claude Sonnet 4.6 vs Qwen3.6 Max Preview102Claude Sonnet 4.6 vs Mistral Large 291Claude Sonnet 4.6 vs Llama 3 8B Instruct90Claude Sonnet 4.6 vs DeepSeek R1 Distill Llama 70B89Claude Sonnet 4.6 vs Qwen2.5-72B-Instruct89Claude Sonnet 4.6 vs StepFun Step-284Claude Sonnet 4.6 vs Claude Mythos Preview84Claude Sonnet 4.6 vs o3 Mini81Claude Sonnet 4.6 vs Qwen2.5-72B76Claude Sonnet 4.6 vs Qwen3.5-122B-A10B67Claude Sonnet 4.6 vs Claude Opus 4.866Claude Sonnet 4.6 vs Trinity-Large-Thinking63Claude Sonnet 4.6 vs Ling-2.6-Flash56Claude Sonnet 4.6 vs Xiaomi MiMo-V2.5-TTS-Series54Claude Sonnet 4.6 vs DeepSeek V3.153Claude Sonnet 4.6 vs GPT-4o (2024-11-20)52Claude Sonnet 4.6 vs Qwen2.5-7B-Instruct49Claude Sonnet 4.6 vs GLM-5V-Turbo36Claude Sonnet 4.6 vs Mistral Nemotron36Claude Sonnet 4.6 vs GPT-4o (11-20)33Claude Sonnet 4.6 vs DeepSeek R1 052832Claude Sonnet 4.6 vs GPT-5.5 Instant32Claude Sonnet 4.6 vs Phi 3.5 Mini Instruct26Claude Sonnet 4.6 vs Llama 3.2 1B Instruct25Claude Sonnet 4.6 vs GLM-5 9B23Claude Sonnet 4.6 vs Together AI - Llama 3 8B Lite22Claude Sonnet 4.6 vs Qwen3-9B17Claude Sonnet 4.6 vs DeepSeek R1 Zero17Claude Sonnet 4.6 vs Mixtral 8x22B v0.115Claude Sonnet 4.6 vs Llama 2 13B Chat14Claude Sonnet 4.6 vs Llama Guard 4 12B12Claude Sonnet 4.6 vs Mixtral 8x7B12Claude Sonnet 4.6 vs Qwen2-7B-Instruct12Claude Sonnet 4.6 vs Gemma 7B Instruct11Claude Sonnet 4.6 vs Code Davinci 0018Claude Sonnet 4.6 vs o3 Deep Research8Claude Sonnet 4.6 vs Trinity-Large-Preview7Claude Sonnet 4.6 vs Phi-3 Mini 4k6Claude Sonnet 4.6 vs Gemini 2.5 Flash Live API6Claude Sonnet 4.6 vs Gemma 2 9B SahabatAI Instruct5Claude Sonnet 4.6 vs Phi-4 Mini Flash Reasoning5Claude Sonnet 4.6 vs Mixtral 8x22B Instruct v0.34Claude Sonnet 4.6 vs Code Cushman 0024Claude Sonnet 4.6 vs ShieldGemma 9B2Claude Sonnet 4.6 vs Mistral Large 3 675B Instruct1Claude Sonnet 4.6 vs GLM-50