Where is the Claude Sonnet 4.6 documentation?

The Claude Sonnet 4.6 documentation on OpenRouter includes the Claude 4.6 Migration Guide cookbook. The quickstart covers API keys and your first request, and the API reference lists every parameter.

How much does Claude Sonnet 4.6 cost?

Claude Sonnet 4.6 costs $3.00/M input tokens and $15.00/M output tokens, with separate rates for Cache Read at $0.30/M tokens, Cache Write at $3.75/M tokens, Cache Write (1h) at $6.00/M tokens, and Web Search at $10.00/1K calls.

What is the context length of Claude Sonnet 4.6?

Claude Sonnet 4.6 has a 1,000,000 token context window. It supports up to 128,000 completion tokens.

Does Claude Sonnet 4.6 support tool calling and structured outputs?

Yes. Claude Sonnet 4.6 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

What inputs and outputs does Claude Sonnet 4.6 support?

Claude Sonnet 4.6 accepts text, images, and files such as PDFs as input and returns text.

Which providers serve Claude Sonnet 4.6?

Claude Sonnet 4.6 is served by 5 providers on OpenRouter: Claude Platform on AWS, Azure, Anthropic, Amazon Bedrock, and Google Vertex. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

When was Claude Sonnet 4.6 released?

Claude Sonnet 4.6 was released on February 17, 2026.

Anthropic: Claude Sonnet 4.6

anthropic/claude-sonnet-4.6

Compare

Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work. It excels at iterative development, complex codebase navigation, end-to-end project management with memory, polished document creation, and confident computer use for web QA and workflow automation.

Modalities

In / out price

$3 / $15per 1M

Context

1.0M

Released

Feb 17, 2026

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput


	$3.00	$15.00	$0.30	0.88s	77 tok/s	99.97%
	$3.00	$15.00	$0.30	0.59s	43 tok/s	99.97%
	$3.00	$15.00	$0.30	0.78s	41 tok/s	99.03%
	$3.00	$15.00	$0.30	0.86s	47 tok/s	99.93%
	$3.00	$15.00	$0.30	--	--	80.00%
Not used in Standard routing:Why these endpoints are not used
	$3.30	$16.50	$0.33	0.67s	70 tok/s	100.00%
	$3.30	$16.50	$0.33	0.73s	95 tok/s	100.00%
	$3.30	$16.50	$0.33	--	--	100.00%
	$3.30	$16.50	$0.33	0.74s	72 tok/s	100.00%

Input $3.00Output $15.00
77 tok/s99.97% uptime
Input $3.00Output $15.00
43 tok/s99.97% uptime
Input $3.00Output $15.00
41 tok/s99.03% uptime
Input $3.00Output $15.00
47 tok/s99.93% uptime
Input $3.00Output $15.00
-- tok/s80.00% uptime
Not used in Standard routing:Why these endpoints are not used
Input $3.30Output $16.50
70 tok/s100.00% uptime
Input $3.30Output $16.50
95 tok/s100.00% uptime
Input $3.30Output $16.50
-- tok/s100.00% uptime
Input $3.30Output $16.50
72 tok/s100.00% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$1.412/M tokens

Output

$15.07/M tokens

Claude Sonnet 4.6 — Price history

Price history

No effective price data for this selection

	$0.997	$15.00	79.8%	1.5%
	$1.161	$15.00	70.3%	22.9%
	$1.195	$15.00	71.0%	18.1%
	$1.593	$15.00	54.2%	56.9%
	$3.000	$15.00	0.0%	0.1%
Not used in Standard routing:Why these endpoints are not used
	$2.468	$16.50	31.4%	0.3%
	$2.869	$16.50	26.3%	0.1%
	$3.183	$16.50	4.6%	0.1%
	$3.382	$16.50	0.0%	0.0%

Effective in $0.997Effective out $15.00
Token share
1.5%
Effective in $1.161Effective out $15.00
Token share
22.9%
Effective in $1.195Effective out $15.00
Token share
18.1%
Effective in $1.593Effective out $15.00
Token share
56.9%
Effective in $3.000Effective out $15.00
Token share
0.1%
Not used in Standard routing:Why these endpoints are not used
Effective in $2.468Effective out $16.50
Token share
0.3%
Effective in $2.869Effective out $16.50
Token share
0.1%
Effective in $3.183Effective out $16.50
Token share
0.1%
Effective in $3.382Effective out $16.50
Token share
0.0%

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

95tok/s

P50, best across providers

Latency

0.59s

P50, best provider

Reasoning effortLatency / throughput

AutoExacto benchmarks

GPQA DiamondTAU-BenchAnthropic82.1%74.9%Google Vertex (US)82.8%73.9%Azure82.6%73.1%Google Vertex82.0%73.7%Amazon Bedrock82.0%73.7%auto-routing81.9%73.1%Amazon Bedrock (US)81.5%73.4%Amazon Bedrock (EU)81.5%73.4%Google Vertex (EU)81.5%71.7%

+1 more providers

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)

100.00%

Availability (3d)

99.90%

Availability over the last 3 days

Last 72 hours

Availability 99.90%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.93%

Without Routing

97.56%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for Anthropic: Claude Sonnet 4.6 (Artificial Analysis, Design Arena and Vals AI)
Source	Benchmark	Score
Artificial Analysis	Claude Sonnet 4.6 (Max) Coding Index	63.0
Artificial Analysis	Claude Sonnet 4.6 (Max) Agentic Index	31.8
Artificial Analysis	Claude Sonnet 4.6 (Max) GPQA Diamond	87.5%
Artificial Analysis	Claude Sonnet 4.6 (Max) HLE	33.6%
Artificial Analysis	Claude Sonnet 4.6 (Max) IFBench	56.6%
Artificial Analysis	Claude Sonnet 4.6 (Max) τ²-Bench Telecom	75.7%
Artificial Analysis	Claude Sonnet 4.6 (Max) AA-LCR	80.0%
Artificial Analysis	Claude Sonnet 4.6 (Max) τ-Bench Banking	34.4%
Artificial Analysis	Claude Sonnet 4.6 (Max) GDPval-AA	37.0%
Artificial Analysis	Claude Sonnet 4.6 (Max) CritPt	3.1%
Artificial Analysis	Claude Sonnet 4.6 (Max) SciCode	50.1%
Artificial Analysis	Claude Sonnet 4.6 (Max) Terminal-Bench Hard	53.0%
Artificial Analysis	Claude Sonnet 4.6 (Max) Terminal-Bench 4.0	3.0%
Artificial Analysis	Claude Sonnet 4.6 (Max) Terminal-Bench 2.1	71.2%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, High) GPQA Diamond	79.9%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, High) HLE	13.3%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, High) IFBench	41.2%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, High) τ²-Bench Telecom	79.5%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, High) AA-LCR	68.3%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, High) CritPt	0.9%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, High) Terminal-Bench Hard	46.2%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, Low) GPQA Diamond	79.7%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, Low) HLE	11.2%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, Low) IFBench	42.4%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, Low) τ²-Bench Telecom	78.9%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, Low) AA-LCR	69.3%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, Low) CritPt	0.9%
Artificial Analysis	Claude Sonnet 4.6 (Non-reasoning, Low) Terminal-Bench Hard	42.4%
Design Arena	Claude Sonnet 4.6 Agents Arena Agenticgamedev Elo	1114
Design Arena	Claude Sonnet 4.6 Agents Arena Androidnative Elo	1183
Design Arena	Claude Sonnet 4.6 Agents Arena Full Stack Elo	1185
Design Arena	Claude Sonnet 4.6 Agents Arena Godotgamedev Elo	1226
Design Arena	Claude Sonnet 4.6 Agents Arena Mobile Apps Elo	1204
Design Arena	Claude Sonnet 4.6 Agents Arena Webapps Elo	1167
Design Arena	Claude Sonnet 4.6 Models Arena 3D Elo	1214
Design Arena	Claude Sonnet 4.6 Models Arena Asciiart Elo	1230
Design Arena	Claude Sonnet 4.6 Models Arena Code Categories Elo	1265
Design Arena	Claude Sonnet 4.6 Models Arena Data Visualization Elo	1275
Design Arena	Claude Sonnet 4.6 Models Arena Game Development Elo	1248
Design Arena	Claude Sonnet 4.6 Models Arena SVG Elo	1186
Design Arena	Claude Sonnet 4.6 Models Arena UI Component Elo	1262
Design Arena	Claude Sonnet 4.6 Models Arena Website Elo	1277
Vals AI	Claude Sonnet 4.6 AIME	92.3%
Vals AI	Claude Sonnet 4.6 CaseLaw v2	64.0%
Vals AI	Claude Sonnet 4.6 Code Migration	39.9%
Vals AI	Claude Sonnet 4.6 CorpFin v2	65.3%
Vals AI	Claude Sonnet 4.6 EMB	60.2%
Vals AI	Claude Sonnet 4.6 Finance Agent (v2)	51.0%
Vals AI	Claude Sonnet 4.6 GPQA Diamond	85.6%
Vals AI	Claude Sonnet 4.6 Harvey's Legal Agent Benchmark v1	5.0%
Vals AI	Claude Sonnet 4.6 LiveCodeBench	82.1%
Vals AI	Claude Sonnet 4.6 LegalBench	82.1%
Vals AI	Claude Sonnet 4.6 Legal Research Bench	38.5%
Vals AI	Claude Sonnet 4.6 MedQA	92.1%
Vals AI	Claude Sonnet 4.6 MMLU Pro	87.3%
Vals AI	Claude Sonnet 4.6 MMMU Pro	83.6%
Vals AI	Claude Sonnet 4.6 MortgageTax	67.7%
Vals AI	Claude Sonnet 4.6 ProgramBench	0.5%
Vals AI	Claude Sonnet 4.6 ProofBench v1	45.0%
Vals AI	Claude Sonnet 4.6 Public Benefits Bench v1.1	62.4%
Vals AI	Claude Sonnet 4.6 SAGE v1	46.6%
Vals AI	Claude Sonnet 4.6 SkillsBench	49.1%
Vals AI	Claude Sonnet 4.6 SWE-bench	77.4%
Vals AI	Claude Sonnet 4.6 Tax Agent Bench	25.8%
Vals AI	Claude Sonnet 4.6 TaxEval v2	77.1%
Vals AI	Claude Sonnet 4.6 Terminal-Bench 2.1	57.3%
Vals AI	Claude Sonnet 4.6 Vals Index	50.6%
Vals AI	Claude Sonnet 4.6 Vals Multimodal Index	60.6%
Vals AI	Claude Sonnet 4.6 Vibe Code Bench v1.1	51.5%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

Claude Code

Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.

60Btokens

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

58.1Btokens

OpenClaw

OpenClaw is an open-source AI agent that connects to your messaging apps and takes real actions on your behalf, from running commands and browsing the web to managing files and sending emails.

46.4Btokens

KoalaBear

new

12.9Btokens

There are many coding agents, but this one is yours.

10.5Btokens

Rest of OpenRouter

10.7B

Top 5 Apps

2.54B

Anthropic: Claude Sonnet versions

Available through EU in-region routing

Anthropic: Claude Sonnet 4.6 is available through EU in-region routing. Requests sent to eu.openrouter.ai are decrypted in the EU and routed only to shared provider endpoints in Europe (the EU/EEA, the UK, and Switzerland).

Providers in Europe

Amazon Bedrock, Google Vertex

Prices in Europe

Amazon Bedrock: $3.30/M tokens input, $16.50/M tokens output
Google Vertex: $3.30/M tokens input, $16.50/M tokens output

Base URL

https://eu.openrouter.ai/api/v1

Plan

Requires Business or Enterprise. Self-serve on Business: no contract, no monthly minimum. OpenRouter Business for EU in-region routing

Fails closed: if no shared provider endpoint in Europe is available, the request errors instead of leaving the region. Requests served by your own BYOK key go to your deployment’s region, which you control. In-region routing docs

More models from Anthropic

Claude Fable 5.2

Claude Fable 5.2 is Anthropic's most capable model for the most demanding technical work, with gains over Fable 5.1 in coding, reasoning, and vision. In prerelease testing it completed otherwise-infeasible software migrations, designed working electronic devices from scratch, and made progress on open problems in physics and mathematics. It communicates more clearly than earlier Fable models and generates output up to 30% faster than Fable 5.1. Adaptive extended thinking is always on, with reasoning effort levels from low to max (default high) and preserved thinking across turns. Anthropic recommends Opus 5.5 as the daily driver for most coding and knowledge work, with Fable 5.2 for the most ambitious, open-ended problems.

Text1M context$8 / $40

Claude Fable 5.2 (batch)

Text1M context$4 / $20

Claude Haiku 5.5

Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and knowledge work, and accepts text and image input with a 1M-token context window.

It is the first Haiku model with adjustable effort. Thinking is adaptive and on by default, with effort as the main lever for trading off depth, latency, and cost; it can be turned off at low, medium, and high effort.

Text1M context$0.10 / $0.50

Claude Haiku 5.5 (batch)

Text1M context$0.05 / $0.25

Claude Sonnet 5.5

Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a direct upgrade. It is especially strong at building features, fixing bugs, and producing polished documents, slides, and spreadsheets, and it writes and communicates more clearly than its predecessor.

Thinking is always on, so effort is the main lever for trading off depth, latency, and cost, and lower effort settings keep it responsive for everyday agentic loops.

Text1M context$2 / $10

Claude Sonnet 5.5 (batch)

Thinking is always on, so effort is the main lever for trading off depth, latency, and cost, and lower effort settings keep it responsive for everyday agentic loops.

Text1M context$1 / $5

Claude Opus 5.5

Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code review and bug finding, financial and scientific analysis, and reading dense charts, diagrams, and screenshots, and it is more careful than its predecessor about only stating figures and citing sources it can back up.

The model completes comparable tasks in fewer steps and with fewer tokens than Opus 5, and reports on its work in plainer language, with clear updates on what it did, what it found, and what it needs from the user. Thinking is always adaptive, so effort is the main lever for trading off depth, latency, and cost, and lower effort settings remain effective for latency-sensitive workloads.

Text1M context$4 / $20

Claude Opus 5.5 (batch)

Text1M context$2 / $10

Claude Fable 5.1

Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows, and knowledge work: long code refactors, front-end and visual code generation, and finance and analysis tasks in particular. It also tends to be more concise than Fable 5 in its plans and summaries. We recommend testing it as a direct upgrade wherever you use Fable 5 today, and alongside Opus 5 on reasoning-heavy tasks.

Text1M context$10 / $50

Claude Fable 5.1 (batch)

Text1M context$5 / $25

Claude Opus 5

Fast-mode variant of Opus 5 - identical capabilities with higher output speed at 2x pricing relative to regular Opus 5.

Learn more in Anthropic's docs: https://platform.claude.com/docs/en/build-with-claude/fast-mode

Note: As of September 1, 2026, this dedicated fast model is deprecated. Fast mode is now served by the fast service tier endpoint on Claude Opus 5 — request it with service_tier: "fast" or speed: "fast". Requests to this model keep working and are served by the same fast tier capacity, so no action is required, but new integrations should target the regular model. Learn more in our service tier docs: https://openrouter.ai/docs/guides/features/service-tiers

Text1M context

Claude Opus 5

Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis of charts and documents, complex office deliverables, and coordinating parallel subagents.

The model maintains strong instruction following and tool use across extended tasks, while remaining effective at lower effort settings for workloads that prioritize latency and token efficiency.

Text1M context$5 / $25

Claude Opus 5 (batch)

The model maintains strong instruction following and tool use across extended tasks, while remaining effective at lower effort settings for workloads that prioritize latency and token efficiency.

Text1M context$2.50 / $12.50

Claude Sonnet 5

Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max, and x-high), a 1M-token context window, and text, image, and file inputs. Sonnet 5 uses an updated tokenizer and includes real-time cyber safeguards that block certain high-risk dual-use activities.

Text1M context$2 / $10

Claude Sonnet 5 (batch)

Text1M context$1 / $5

Claude Fable Latest

This model always redirects to the latest model in the Claude Fable family.

Text1M context

Claude Fable 5

Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token context window. It is suited for long-running, complex, and asynchronous tasks that previously required frequent human check-ins.

It is particularly strong at end-to-end work that would otherwise take a person hours, days, or weeks - taking on problems that are long-running, ambiguous, or highly multi-step. It executes well-scoped tasks with few mistakes, automatically self-correcting through verification loops, and ships with robust safeguards.

Text1M context$10 / $50

Claude Fable 5 (batch)

Text1M context$5 / $25

Claude Opus 4.8

Fast-mode variant of Opus 4.8 - identical capabilities with higher output speed at 2x pricing relative to regular Opus 4.8.

Learn more in Anthropic's docs: https://platform.claude.com/docs/en/build-with-claude/fast-mode

Note: As of September 1, 2026, this dedicated fast model is deprecated. Fast mode is now served by the fast service tier endpoint on Claude Opus 4.8 — request it with service_tier: "fast" or speed: "fast". Requests to this model keep working and are served by the same fast tier capacity, so no action is required, but new integrations should target the regular model. Learn more in our service tier docs: https://openrouter.ai/docs/guides/features/service-tiers

Text1M context

Claude Opus 4.8

Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token context window. It is suited for highly autonomous agents, long-horizon agentic work, knowledge work, and memory-driven tasks where coherence over extended sessions matters.

It is particularly strong on multi-step reasoning, complex coding, and end-to-end project orchestration - large codebases, multi-stage debugging, and long-running asynchronous agent pipelines. Beyond coding, it handles knowledge work such as drafting documents, building presentations, and analyzing data, maintaining quality across very long outputs.

Text1M context$5 / $25

Anthropic: Claude Sonnet 4.6

Providers