Claude Opus 4.6

Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks. It is built for agents that operate across entire workflows rather than single prompts, making it especially effective for large codebases, complex refactors, and multi-step debugging that unfolds over time. The model shows deeper contextual understanding, stronger problem decomposition, and greater reliability on hard engineering tasks than prior generations.

Beyond coding, Opus 4.6 excels at sustained knowledge work. It produces near-production-ready documents, plans, and analyses in a single pass, and maintains coherence across very long outputs and extended sessions. This makes it a strong default for tasks that require persistence, judgment, and follow-through, such as technical design, migration planning, and end-to-end project execution.

For users upgrading from earlier Opus versions, see our official migration guide here

anthropic/claude-opus-4-6
Organization
Anthropic
Providers
32
Context
1,000,000
Output limit
128,000
Knowledge
2025-05-31
Release
2026-02-05
Updated
2026-03-13
Weights
Closed
Input
text, image, pdf
Output
text
Capabilities
Tools, Reasoning, Temperature

Providers

Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.

49 offers · USD per million tokens
ProviderModel IDContextOutputInput / output · 1MReasoningToolsStructuredDetails
Abacus
claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYes—
View
Amazon Bedrock
eu.anthropic.claude-opus-4-6-v1
1,000,000128,000$5.50 / $27.50YesYesYes
View
Amazon Bedrock
us.anthropic.claude-opus-4-6-v1
1,000,000128,000$5.50 / $27.50YesYesYes
View
Amazon Bedrock
au.anthropic.claude-opus-4-6-v1
1,000,000128,000$5.50 / $27.50YesYesYes
View
Amazon Bedrock
anthropic.claude-opus-4-6-v1
1,000,000128,000$5.50 / $27.50YesYesYes
View
Amazon Bedrock
global.anthropic.claude-opus-4-6-v1
1,000,000128,000$5.00 / $25.00YesYesYes
View
AnyAPI
anthropic/claude-opus-4-6
1,000,000128,000— / —YesYes—
View
Auriko
claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYes—
View
Cloudflare AI Gateway
anthropic/claude-opus-4.6
1,000,000128,000$5.00 / $25.00YesYesYes
View
Cortecs
claude-opus4-6
1,000,000128,000$5.313 / $26.561YesYesYes
View
Databricks
databricks-claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYes—
View
DevPass (LLM Gateway)
claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYes—
View
Eden AI
anthropic/claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYesYes
View
FreeModel
claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYes—
View
GMI Cloud
anthropic/claude-opus-4.6
409,600128,000$5.00 / $25.00YesYes—
View
Impossibl
anthropic/claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYes—
View
Kilo Gateway
anthropic/claude-opus-4.6
1,000,000128,000$5.00 / $25.00YesYesYes
View
LLM Gateway
azure-anthropic/claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYesYes
View
LLM Gateway
aws-bedrock/claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYesYes
View
LLM Gateway
anthropic/claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYesYes
View
LLM Gateway
vertex-anthropic/claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYesYes
View
Merge Gateway
anthropic/claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYesYes
View
NanoGPT
anthropic/claude-opus-4.6:thinking:max
1,000,000128,000$5.00 / $25.00YesYesYes
View
NanoGPT
anthropic/claude-opus-4.6:thinking:medium
1,000,000128,000$5.00 / $25.00YesYesYes
View
NanoGPT
anthropic/claude-opus-4.6
1,000,000128,000$5.00 / $25.00YesYesYes
View
NanoGPT
anthropic/claude-opus-4.6:thinking:low
1,000,000128,000$5.00 / $25.00YesYesYes
View
NanoGPT
anthropic/claude-opus-4.6:thinking
1,000,000128,000$5.00 / $25.00YesYesYes
View
NEAR AI Cloud
anthropic/claude-opus-4-6
200,000128,000$5.00 / $25.00YesYes—
View
Neon
claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYesYes
View
Ofox
anthropic/claude-opus-4.6
1,000,000128,000$5.00 / $25.00YesYes—
View
OpenCode Zen
claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYes—
View
Amazon Bedrockvia OpenRouter
anthropic/claude-opus-4.6
unknown
1,000,000128,000$5.00 / $25.00YesYesYes
View
Anthropicvia OpenRouter
anthropic/claude-opus-4.6
unknown
1,000,000128,000$5.00 / $25.00YesYesYes
View
Azurevia OpenRouter
anthropic/claude-opus-4.6
unknown
1,000,000128,000$5.00 / $25.00YesYesYes
View
Claude Platform on AWSvia OpenRouter
anthropic/claude-opus-4.6
unknown
1,000,000128,000$5.00 / $25.00YesYesYes
View
Googlevia OpenRouter
anthropic/claude-opus-4.6
unknown
1,000,000128,000$5.00 / $25.00YesYesYes
View
Googlevia OpenRouter
anthropic/claude-opus-4.6
unknown
1,000,000128,000$5.50 / $27.50YesYesYes
View
Opper
claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYesYes
View
OrcaRouter
anthropic/claude-opus-4.6
1,000,000128,000$5.00 / $25.00YesYes—
View
Pioneer
claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYes—
View
Requesty
claude-opus-4-6@eu
1,000,000128,000$5.50 / $27.50YesYesYes
View
Requesty
claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYesYes
View
SAP AI Core
anthropic--claude-4.6-opus
1,000,000128,000$5.00 / $25.00YesYes—
View
Snowflake Cortex
claude-opus-4-6
1,000,000128,000— / —YesYes—
View
Tempr Gateway
anthropic/claude-opus-4-6
1,000,000128,000$5.00 / $25.00YesYesYes
View
Venice AI
claude-opus-4-6
1,000,000128,000$6.00 / $30.00YesYesYes
View
Vercel AI Gateway
anthropic/claude-opus-4.6
1,000,000128,000$5.00 / $25.00YesYes—
View
Vertex
claude-opus-4-6@default
1,000,000128,000$5.00 / $25.00YesYes—
View
Vertex (Anthropic)
claude-opus-4-6@default
1,000,000128,000$5.00 / $25.00YesYes—
View

Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.

Pricing

Compare token rates and cache charges across available routes and hosting configurations.

USD per million tokens
Input from$5.00Per 1M input tokens
Output from$25.00Per 1M output tokens
Available offers49Across 32 providers

Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.

Endpoint price comparisonLatest observed snapshot · 6 lowest-priced configurations
InputOutput
Amazon Bedrockamazon-bedrock
$5.00$25.00
Anthropicanthropic
$5.00$25.00
Azureazure/global
$5.00$25.00
Claude Platform on AWSclaude-on-aws
$5.00$25.00
Googlegoogle-vertex/global
$5.00$25.00
Googlegoogle-vertex/europe
$5.50$27.50
anthropic/claude-opus-4.6Observed
Input
$5.00
Output
$25.00
Cache read
$0.50
Cache write
$6.25
Cache write · 1 hour
$10.00
Additional fees and pricing conditions
web search
$0.01 / search
Hosting prices through OpenRouter All endpoints and cache rates
EndpointInput / 1MOutput / 1MCache read / 1MCache write / 1M
Amazon Bedrockamazon-bedrock$5.00$25.00$0.50$6.25
Anthropicanthropic$5.00$25.00$0.50$6.25
Azureazure/global$5.00$25.00$0.50$6.25
Claude Platform on AWSclaude-on-aws$5.00$25.00$0.50$6.25
Googlegoogle-vertex/global$5.00$25.00$0.50$6.25
Googlegoogle-vertex/europe$5.50$27.50$0.55$6.875

Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.

Performance

Review reported latency and output-throughput percentiles from the latest observed rolling window.

Latest 30-minute observation

Time to first token · milliseconds

Latency distributionRolling 30-minute snapshot · 5 endpoints
P50P75P90P99
Amazon Bedrockamazon-bedrock
1,606–5,184.5 ms
Anthropicanthropic
1,756.5–10,869.35 ms
Azureazure/global
1,330–1,330 ms
Claude Platform on AWSclaude-on-aws
1,471–22,515.96 ms
Googlegoogle-vertex/global
1,787–8,304.13 ms
All latency percentiles 5 endpoints
EndpointP50P75P90P99
Amazon Bedrockamazon-bedrock1,606 ms2,151.5 ms2,800.4 ms5,184.5 ms
Anthropicanthropic1,756.5 ms2,455 ms4,187 ms10,869.35 ms
Azureazure/global1,330 ms1,330 ms1,330 ms1,330 ms
Claude Platform on AWSclaude-on-aws1,471 ms1,782 ms2,559.2 ms22,515.96 ms
Googlegoogle-vertex/global1,787 ms2,196.75 ms5,180.4 ms8,304.13 ms

Output throughput · tokens per second

Throughput distributionRolling 30-minute snapshot · 5 endpoints
P50P75P90P99
Amazon Bedrockamazon-bedrock
37–61 t/s
Anthropicanthropic
34–70.95 t/s
Azureazure/global
21–21 t/s
Claude Platform on AWSclaude-on-aws
40–93 t/s
Googlegoogle-vertex/global
39–81.5 t/s
All throughput percentiles 5 endpoints
EndpointP50P75P90P99
Amazon Bedrockamazon-bedrock37 t/s44.25 t/s57 t/s61 t/s
Anthropicanthropic34 t/s41.75 t/s52 t/s70.95 t/s
Azureazure/global21 t/s21 t/s21 t/s21 t/s
Claude Platform on AWSclaude-on-aws40 t/s53 t/s60 t/s93 t/s
Googlegoogle-vertex/global39 t/s47.25 t/s59 t/s81.5 t/s

Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.

Uptime

Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.

Successful requests by window
Availability by endpointCurrent overlapping windows · 6 endpoints
Endpoint5 minutes30 minutes24 hours
Amazon Bedrockamazon-bedrock
Anthropicanthropic
Azureazure/global
Claude Platform on AWSclaude-on-aws
Googlegoogle-vertex/global
Googlegoogle-vertex/europe

Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.

Benchmarks

Review attributed evaluation results on each source's original scale.

External evaluations
anthropic/claude-opus-4.6Observed
Design ArenaCategoryEloWin rateRank
agentsagenticgamedev1,21350%8
agentsandroidnative1,21563.8%12
agentsfullstack1,21859.8%13
agentsmobileapps1,20459.2%15
agentswebapps1,19957%18
models3d1,29062.5%17
modelsasciiart1,26661.9%14
modelscodecategories1,29561.2%15
modelsdataviz1,28958.7%15
modelsgamedev1,29061%18
modelssvg1,22760.9%17
modelsuicomponent1,29159.8%20
modelswebsite1,29861.2%13

Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.

Detailed evaluations

Evaluation scoresAccuracy or item F1 · higher is better
tau bench verified airlineAnthropic: Claude Opus 4.6
74.741%
gpqa diamondAnthropic: Claude Opus 4.6
86.027%
EvaluationConfigurationSourceResultsDetails
models · websiteClaude Opus 4.6anthropic/claude-4.6-opus-20260205Design Arena1,298 Elo
Details
models · asciiartClaude Opus 4.6anthropic/claude-4.6-opus-20260205Design Arena1,264 Elo
Details
models · datavizClaude Opus 4.6anthropic/claude-4.6-opus-20260205Design Arena1,289 Elo
Details
models · gamedevClaude Opus 4.6anthropic/claude-4.6-opus-20260205Design Arena1,285 Elo
Details
models · datavizClaude Opus 4.6 (Thinking)anthropic/claude-4.6-opus-20260205Design Arena1,279 Elo
Details
tau bench verified airlineAnthropic: Claude Opus 4.6anthropic/claude-4.6-opus-20260205OpenRouter74.741%
Details
models · websiteClaude Opus 4.6 (Thinking)anthropic/claude-4.6-opus-20260205Design Arena1,293 Elo
Details
models · 3dClaude Opus 4.6anthropic/claude-4.6-opus-20260205Design Arena1,290 Elo
Details
models · codecategoriesClaude Opus 4.6 (Thinking)anthropic/claude-4.6-opus-20260205Design Arena1,291 Elo
Details
models · uicomponentClaude Opus 4.6anthropic/claude-4.6-opus-20260205Design Arena1,291 Elo
Details
models · asciiartClaude Opus 4.6 (Thinking)anthropic/claude-4.6-opus-20260205Design Arena1,266 Elo
Details
models · svgClaude Opus 4.6 (Thinking)anthropic/claude-4.6-opus-20260205Design Arena1,227 Elo
Details
models · gamedevClaude Opus 4.6 (Thinking)anthropic/claude-4.6-opus-20260205Design Arena1,290 Elo
Details
models · uicomponentClaude Opus 4.6 (Thinking)anthropic/claude-4.6-opus-20260205Design Arena1,288 Elo
Details
gpqa diamondAnthropic: Claude Opus 4.6anthropic/claude-4.6-opus-20260205OpenRouter86.027%
Details
models · codecategoriesClaude Opus 4.6anthropic/claude-4.6-opus-20260205Design Arena1,295 Elo
Details
models · 3dClaude Opus 4.6 (Thinking)anthropic/claude-4.6-opus-20260205Design Arena1,287 Elo
Details
models · svgClaude Opus 4.6anthropic/claude-4.6-opus-20260205Design Arena1,214 Elo
Details

Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 4, 2026.

Apps & session costs

Compare observed median session costs for applications using this model, grouped by conversation length.

30-day sample · through Sep 27, 2026
ApplicationModel routeTurnsMedian cost / sessionDetails
Hermes Agentanthropic/claude-4.6-opus-202602051 turn
$0.002461
Details
Claude Codeanthropic/claude-4.6-opus-202602051 turn
$0.068548
Details
Claude Codeanthropic/claude-4.6-opus-2026020550+ turns
$9.06267
Details
Hermes Agentanthropic/claude-4.6-opus-202602052–9 turns
$0.11591
Details
Hermes Agentanthropic/claude-4.6-opus-2026020510–49 turns
$0.874375
Details
Hermes Agentanthropic/claude-4.6-opus-2026020550+ turns
$5.968968
Details
Claude Codeanthropic/claude-4.6-opus-2026020510–49 turns
$1.356056
Details
Claude Codeanthropic/claude-4.6-opus-202602052–9 turns
$0.360704
Details

Source: OpenRouter, as of Oct 2, 2026. Published under CC BY 4.0. These are observed costs per session, not token prices or a forecast for your workload. Apps are compared separately.

Activity

Daily prompt and completion tokens reported for this model's routes in OpenRouter's top 50. Missing days have no published value.

Last 90 complete UTC days
Daily token usage1,151,989,284,580 reported tokens · 42 days with observations
Jul 6, 2026Oct 3, 2026
Daily observations · 42 days
Date (UTC)Reported tokensReported routes
Aug 26, 202623,706,947,3621
Aug 24, 202629,715,111,2221
Aug 18, 202632,662,978,3091
Aug 17, 202627,628,781,7131
Aug 15, 202623,996,987,8311
Aug 14, 202646,134,608,0291
Aug 13, 202625,975,183,5561
Aug 12, 202625,034,375,3921
Aug 11, 202626,610,249,0191
Aug 10, 202622,796,165,6871
Aug 9, 202617,077,739,6641
Aug 8, 202615,341,030,6171
Aug 7, 202622,545,795,9471
Aug 6, 202624,433,010,6861
Aug 5, 202624,123,403,4921
Aug 3, 202623,145,169,2841
Aug 2, 202617,148,916,0741
Aug 1, 202617,167,380,6761
Jul 31, 202627,969,195,0041
Jul 30, 202627,827,003,3161
Jul 29, 202628,094,517,7841
Jul 28, 202626,554,394,5011
Jul 27, 202625,269,604,1611
Jul 26, 202618,476,344,0061
Jul 24, 202629,546,962,5131
Jul 23, 202633,637,522,9721
Jul 22, 202631,209,025,7761
Jul 21, 202633,704,072,2881
Jul 20, 202634,272,676,3141
Jul 19, 202621,514,587,4271
Jul 18, 202623,039,745,0891
Jul 17, 202631,021,230,7881
Jul 16, 202632,391,966,1461
Jul 15, 202636,781,315,0991
Jul 14, 202636,342,232,5441
Jul 13, 202631,805,933,2261
Jul 12, 202626,131,472,6541
Jul 11, 202621,000,217,4301
Jul 10, 202628,879,551,8561
Jul 9, 202634,140,237,2471
Jul 8, 202634,359,314,9411
Jul 7, 202632,776,326,9381

Source: OpenRouter (openrouter.ai/rankings), as of Oct 4, 2026. CC BY 4.0. Totals include only individually published routes. Tokenizers differ by provider; the aggregated “other” category is never assigned to a model.

Technical details & API

Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.

Current route specifications
anthropic/claude-opus-4.6Observed
Context
1,000,000
Maximum output
128,000
Tokenizer
Claude
Input
text, image, file
Output
text
Moderation
Yes
Added to OpenRouter
2026-02-04
Base URLhttps://openrouter.ai/api/v1
Model IDanthropic/claude-opus-4.6
Version IDanthropic/claude-4.6-opus-20260205

Reasoning

Required
No
Enabled by default
No
Default effort
high
Supported efforts
max, high, medium, low

Supported parameters

  • include_reasoning
  • max_completion_tokens
  • max_tokens
  • reasoning
  • reasoning_effort
  • response_format
  • stop
  • structured_outputs
  • temperature
  • tool_choice
  • tools
  • top_k
  • top_p
  • verbosity

Frequently asked questions

Answers based on the model metadata and serving offers currently in the catalog.

How much does Claude Opus 4.6 cost?

Listed rates start at $5.00 per million input tokens and $25.00 per million output tokens. Prices vary by provider and configuration; see Pricing for details.

Which providers offer Claude Opus 4.6?

There are 49 cataloged offers across 32 providers. Compare model IDs, prices and limits in Providers.

What is the context window?

The cataloged model context is 1,000,000 tokens. Each serving endpoint may apply a different limit.

Does it support tools and structured output?

Tool calling: Yes. Structured output: Not reported. Support can vary by endpoint.

Catalog history · 2 recorded revisions

Metadata observations since this model was first cataloged. These are not model release versions.

  • · 6ccf6800f233
  • · d22a5f7ce3c1