Skip to content

Anthropic · Chat

Claude Sonnet 4.6

Anthropic 🇺🇸ChatClaude familyOne of its sources says the weights aren't publicly available.The maker currently plans to deprecate it within the next 180 days (10 Feb 2027). Plans can change.

Newer version: Claude Sonnet 5.5. We compared release dates and version names in our catalogue.

From Neosantara's listing:
Anthropic Claude Sonnet 4.6 is the most capable Sonnet model, upgraded for coding, computer use, long-context reasoning, agent planning, knowledge work, and design.
llminfo IQ
133.8
Maker's price
$6.00 blended / 1M
Max context
1M
Listings
125

Evidence and sources

llminfo IQ 133.8, ranked among chat models. From 469 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 108.4, 95% interval 103–114. From 29 benchmark families across 18 sources. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedPoor value (35.4). Costs more per task than its benchmark results predict. From 58 benchmarks with measured task costs. Plenty of data behind it.How Value for money worksYap score: tokens used per coding task vs. models with the same resultsChatty (71 ±8). Generates more tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. By effort: default 71 · low 48 · medium 54 · high 63 · max 56.Median 107,010 output tokens per task at default effort.How the Yap score worksReleased 17 Feb 2026Maker HQ: US (according to Anthropic's own website)
All rankings: #72 on OpenRouter this month #76 on OpenRouter today #76 on OpenRouter this week #201 on NanoGPT recentlySources: Anthropic API data, Heroku Managed Inference data, LiteLLM data, Models.dev data, Neosantara Inference data, llminfo researchPrimary data from Anthropic API from Heroku Managed Inference: primary mode from LiteLLM: deprecation date from Models.dev: weights availability from Neosantara Inference: description from llminfo research: release date
Tags from its sources:instructtoolsvisionreasoningdocumentproprietarycodingagenticlong-context

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Trade-offs

Deprecation scheduled

The maker plans to deprecate it on 10 Feb 2027, within the next 180 days.

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 8 maker's prices; 2 market median prices.

Input & output

as stated by the model's providers
Text
Images
Files
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured outputPrompt cachingPDF inputComputer useWeb searchBatch API

Listings

 Anthropicclaude-sonnet-4-6Sources: Anthropic API data, GenAI Prices data, Helicone Costs data, LiteLLM data, Models.dev data1M$3.00cache read $0.30batch $1.50$15.00cache write 5m $3.75cache write 1h $6.00batch $7.50—
 Xpersonaclaude-sonnet-4-6Sources: Models.dev data, Xpersona data200K$0.90cache read $0.09$5.55$0.001
 Tokensmindclaude-sonnet-4-6Source: TokensMindNot published by the provider$1.05$5.25—
 OpenRouteranthropic/claude-sonnet-4.6:batch · batchSource: OpenRouter data docs1M$1.50cache read $0.15$7.50cache write 5m $1.875cache write 1h $3.00—
 Vercelanthropic/claude-sonnet-4.6-batch · batchSource: Vercel AI Gateway data1M$1.50cache read $0.15$7.50—
 Outerlabsclaude-sonnet-4-6Source: Models.dev data1M$1.80cache read $0.18$9.00cache write $2.25—
 Kunavoclaude-sonnet-4-6Source: Models.dev data1M$2.10cache read $0.21$10.50cache write $2.625—
 Meganovaanthropic/claude-sonnet-4-6Source: MegaNova data1M$2.40$12.00—
 Poeanthropic/claude-sonnet-4.6Source: Models.dev data983K$2.5758cache read $0.2576$12.879cache write $3.2197—
 Amazon Web Servicesanthropic.claude-sonnet-4-6Sources: AWS Bedrock Runtime data, LiteLLM data, Models.dev data1M$3.00cache read $0.30$15.00cache write 5m $3.75cache write 1h $6.00—

11 listings without a price are hidden.

Included in plans

Copilot ProgithubSubscription$10/moextra use billed past the allowancePlan page
Copilot Pro+githubSubscription$39/moextra use billed past the allowancePlan page
Copilot MaxgithubSubscription$100/moextra use billed past the allowancePlan page

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Human preference

Arena ELO67 related benchmarks · 67 results
Source: LMArena data · Nov 2026
Arena ELO
Chinese
Show 61 more resultsHide 61 results
Design Arena EloSource: Design Arena Leaderboards data · Nov 2026

Agentic

BrowseComp6 results
Source: Steel.dev Agent Leaderboards data · Jun 2026
Setting not stated
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
single agent · no thinking · max effort
launch-page figure · web search/fetch + code · no thinking
multi-agent orchestrator · subagents with tools · compaction · max effort
Design Arena EloSource: Design Arena Leaderboards data · Nov 2026
FinanceAgent v1.1Vals AI · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
GAIA4 results
Source: GAIA Leaderboard data · Jul 2026
Difficulty level
Difficulty not reported
GDPval-AA3 related benchmarks · 4 results
GDPval-AA
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
220 tasks · shell + web browsing
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.5 Flash Model Card · May 2026
Published Score
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M2.7 official benchmark overview · exact harness/version not stated
v2
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Artificial Analysis · as of 2026-06-17
MCP Atlas3 results
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.5 Flash Model Card · May 2026
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
MLE-Bench LiteMiniMax M2.7 official benchmark overview · exact harness/version not stated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OpenHandsSource: OpenHands Index data
OSWorld3 related benchmarks · 6 results
OSWorld
Source: Steel.dev Agent Leaderboards data · Jun 2026
OSWorld 2
Source: Epoch AI Benchmarking Hub data
Verified
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
1080p · 100 steps · pass@1 · 5 runs
zoom-tool fix + 128K per-turn output · 361 tasks · 5 runs
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.5 Flash Model Card · May 2026
SWE-Bench ProMiniMax M2.7 official benchmark overview · exact harness/version not stated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Tau2 Retail3 related benchmarks · 8 results
Tau2 Retail
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Tau2 Telecom
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
tau2-bench airline
Source: OpenRouter benchmark exports · Oct 2026
17 Feb 2026 version · 90-day results
Across 5 providers lowest 73.1% median 73.6% highest 74.3%OpenRouter's automatic routing 73.7%
Show results from 5 providers and OpenRouter's automatic routingHide results from 5 providers and OpenRouter's automatic routing
  • Amazon Bedrock
  • Azure
  • OpenRouter's automatic routing
  • Claude Platform on AWS
  • Anthropic
  • Google
Terminal-Bench4 related benchmarks · 4 results
2.0 · Terminus-2
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Terminus-2 · max effort · no thinking · 5 runs
Terminal-Bench 2.0
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Terminal-Bench 2.1
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
mini-SWE-agent · high effort · Sonnet 5 card infrastructure
WebArenasingle policy · screenshot + DOM navigation · general prompts · Average@5 · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
full set · official prompt/grader · pass@1 · Average@5
hard subset · 258 tasks · official prompt/grader · pass@1 · Average@5

Reasoning

ARC-AGI4 related benchmarks · 9 results
ARC-AGI
Source: Epoch AI Benchmarking Hub data
Reasoning effort
ARC-AGI 2
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.5 Flash Model Card · May 2026
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
private set · 120K thinking tokens · high effort
Verified · adaptive thinking · max effort
ARC-AGI-1
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
private set · 120K thinking tokens · high effort
v2
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
AutomationBenchZapier private held-out set · max effort · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
BioMysteryBenchbash/code/package tools · no extended thinking · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
ChartMuseumSonnet 5 card rerun · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSearchQA2 results
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
multi-agent orchestrator · 3M tokens per agent · max effort
single agent · 10M total tokens · search/fetch/code · adaptive thinking · max effort
GDP.pdfSonnet 5 card rerun · 128K max output · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.5 Flash Model Card · May 2026 · full text + multimodal set
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
no tools · adaptive thinking · max effort
no tools · Sonnet 5 card rerun
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
web search + web fetch + code · 3M total tokens · adaptive thinking
with tools · max effort · Sonnet 5 card rerun
MILU12 results
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
adaptive thinking · max effort
average · English + 10 Indic languages · adaptive thinking · max effort
OfficeQAextracted-text sandbox + code · max effort · Sonnet 5 card rerun · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OfficeQA Proextracted-text sandbox + code · max effort · Sonnet 5 card rerun · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
internal pinned harness · pass@1 · 3 trials · max effort
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M2.7 official benchmark overview · exact harness/version not stated

Multilingual

EKI EstonianSource: Keelemudelite mõõdupuu data docs
MMMLU2 results
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
14 non-English languages · adaptive thinking · max effort
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
NepNewsBench v2 draft40 related benchmarks · 40 results
entity F1 (all questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
core fact recall (all questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
core fact recall (English-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
core fact recall (Nepali-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
coverage precision proxy (all questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
Show 34 more resultsHide 34 results
coverage precision proxy (English-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
coverage precision proxy (Nepali-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
effective score (all questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
effective score (English-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
effective score (Nepali-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
English prose quality (all questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
English prose quality (Nepali-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
entity F1 (English-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
entity F1 (Nepali-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
entity precision (all questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
entity precision (English-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
entity precision (Nepali-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
entity recall (all questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
entity recall (English-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
entity recall (Nepali-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
Nepali prose quality (all questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
Nepali prose quality (English-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
Nepali prose quality (Nepali-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
supporting fact recall (all questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
supporting fact recall (English-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
supporting fact recall (Nepali-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
valid-answer quality (all questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
valid-answer quality (English-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off
valid-answer quality (Nepali-source questions)
Source: NepNewsBench v2 (draft candidate release) data · Sep 2026
Anthropic via openrouter, reasoning off

Coding

Arena ELOSource: LMArena data · Nov 2026
CyberGymtargeted vulnerability reproduction · pass@1 · no thinking · default effort · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Multi-SWE-BenchMiniMax M2.7 official benchmark overview · exact harness/version not stated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OpenRCAReasoning effort · author harness · adaptive thinking · 3-run mean · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
SWE-bench Verified7 related benchmarks · 9 results
SWE-bench Verified
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
adaptive thinking · max effort · 10 trials
prompt-modified ablation · >100 tool calls · own tests first
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
SWE-bench Multilingual
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
adaptive thinking · max effort · 10 trials
SWE-bench-Live MultiLang Go
Source: SWE-bench-Live leaderboard data · Jul 2026
scaffold: ami-agent, set: go, instances: 138
Show 1 more resultHide 1 result
ValsSource: Vals AI
VIBE-ProMiniMax M2.7 official benchmark overview · exact harness/version not stated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Long context

Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
256K bin · mean match ratio · 64K extended thinking · 5 trials
256K bin · mean match ratio · adaptive thinking · max effort · 5 trials
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.5 Flash Model Card · May 2026 · 128K cumulative average

Vision

Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.5 Flash Model Card · May 2026 · reasoning · no tools
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
adaptive thinking · max effort · 5 runs · original 4.6 card harness
Sonnet 5 card rerun with Sonnet 4.6 grader
MMMU-Pro4 results
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.5 Flash Model Card · May 2026 · no tools
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
adaptive thinking · max effort
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

General

Science

GPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
GPQA Diamond11 results
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
adaptive thinking · max effort · 10 trials
Source: OpenRouter benchmark exports
17 Feb 2026 version · 90-day results
Across 5 providers lowest 73.5% median 73.7% highest 74.1%OpenRouter's automatic routing 73.7%
Show results from 5 providers and OpenRouter's automatic routingHide results from 5 providers and OpenRouter's automatic routing
  • Azure
  • Amazon Bedrock
  • OpenRouter's automatic routing
  • Google
  • Claude Platform on AWS
  • Anthropic
LAB-Bench FigQAadaptive thinking · max effort · 5 runs · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Knowledge

Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
adaptive thinking · max effort
overall average · all languages · adaptive thinking · max effort
overall average · Sonnet 5 card rerun

OCR

OmniDocBenchSource: Nanonets IDP data

Math

AIME 2025no tools · adaptive thinking · max effort · 10 trials · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Factuality

Safety

PhareSource: Phare

Instruction following

Speech

Other

Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
1,000-file subset · no tools · Sonnet 5 card rerun
Python + image crop · Sonnet 5 card rerun
BioPipelineBenchbash/code/package tools · no extended thinking · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Claw-Eval v1.1Reported by Claw-Eval · via BenchBoard docs
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
256K API-reproducible subset of 1M · F1
1M set · F1 · 64K extended thinking · 5 trials
1M set · F1 · adaptive thinking · max effort · 5 trials
GraphWalks Parents256K API-reproducible subset of 1M · F1 · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
HealthBenchlength-adjusted · Opus 4.8 grader · no tools · Sonnet 5 card rerun · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
length-adjusted · Opus 4.8 grader · no tools · Sonnet 5 card rerun
Source: Clinical Benchmarks data docs · Reported by the model's developer. The source relaying it did not verify it. claim · Jun 2026
Setting not stated
Horangi36 related benchmarks · 36 results
MedCalc-Bench VerifiedPython REPL · code-augmented agent loop · max effort · 5 runs · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MM-ClawBenchMiniMax M2.7 official benchmark overview · exact harness/version not stated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
<1M API-reproducible subset · 29 problems · max effort
<1M API-reproducible subset · 54 problems · 64K extended thinking
1M bin · full internal set · 64K extended thinking · 5 trials
1M bin · full internal set · adaptive thinking · max effort · 5 trials
Nejumi 428 related benchmarks · 28 results
ALT_バイアス
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
ALT_倫理・道徳
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
ALT_制御性
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
ALT_堅牢性
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
ALT_毒性
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
ALT_真実性
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
Show 22 more resultsHide 22 results
AVG_jaster_0shot
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
AVG_jaster_2shots
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_アプリケーション開発
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_コーディング
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_一般的知識
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_基礎的言語性能
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_専門的知識
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_応用的言語性能
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_情報検索
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_意味解析
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_抽象的推論
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_推論能力
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_数学的推論
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_構文解析
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_知識・質問応答
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_翻訳
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_表現
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_論理的推論
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
GLP_関数呼び出し
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
TOTAL_SCORE
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
アラインメント(ALT)_AVG
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
汎用的言語性能(GLP)_AVG
Source: Nejumi Leaderboard 4 data docs
anthropic/claude-sonnet-4.6: extended-thinking
Organic Chemistrybash/code/package tools · no extended thinking · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Phylogeneticsquantitative + visual interpretation · no extended thinking · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Sayf16 related benchmarks · 16 results
Show 10 more resultsHide 10 results
SkillsBench9 related benchmarks · 9 results
1.1 · With Skills
Reported by SkillsBench / BenchFlow · via BenchBoard docs
OpenHands · 87 tasks · 3 trials/task · max reasoning effort · with Skills
1.1 · Without Skills
Reported by SkillsBench / BenchFlow · via BenchBoard docs
OpenHands · 87 tasks · 3 trials/task · max reasoning effort · without Skills
Copilot resolution-rate difference
Reported by GitHub · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
GitHub 2026-06-20 weekly suite · Copilot CLI vs Claude Code with the same model · medium reasoning effort · five independent runs, best-scored run · 2h timeout · no web tools/tool search/MCP servers · adjusted task set, exact task count and version not stated
Vals · With Skills · Accuracy
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=anthropic/claude-sonnet-4-6 · provider=Anthropic · compute_effort=max · reasoning=true · temperature=1 · max_output_tokens=128000
Vals · With Skills · Mean cost / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=anthropic/claude-sonnet-4-6 · provider=Anthropic · compute_effort=max · reasoning=true · temperature=1 · max_output_tokens=128000
Vals · With Skills · Mean duration / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=anthropic/claude-sonnet-4-6 · provider=Anthropic · compute_effort=max · reasoning=true · temperature=1 · max_output_tokens=128000
Show 3 more resultsHide 3 results
Vals · Without Skills · Accuracy
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=anthropic/claude-sonnet-4-6 · provider=Anthropic · compute_effort=max · reasoning=true · temperature=1 · max_output_tokens=128000
Vals · Without Skills · Mean cost / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=anthropic/claude-sonnet-4-6 · provider=Anthropic · compute_effort=max · reasoning=true · temperature=1 · max_output_tokens=128000
Vals · Without Skills · Mean duration / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=anthropic/claude-sonnet-4-6 · provider=Anthropic · compute_effort=max · reasoning=true · temperature=1 · max_output_tokens=128000
Structural Biologybasic tools · no extended thinking · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
τ²-bench Retailadaptive thinking · max effort · 10 trials · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
τ²-bench Telecomadaptive thinking · max effort · 10 trials · Reported by Anthropic · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · Data by LLM Stats · LMArena, CC BY 4.0 (licence) · Steel.dev Agent Leaderboards, MIT (licence) · BullshitBench by Peter Gostev, MIT (licence) · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence) · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · CyberGym leaderboard (Wang, Shi, He, Cai, Zhang, Song; UC Berkeley), ICLR 2026. Results evaluated and submitted by the listed teams. No licence is stated for the result data; the repository's Apache-2.0 licence covers its code only. · Keelemudelite mõõdupuu: Independent LLM leaderboard for Estonian. Eesti Keele Instituut. https://moodupuu.eki.ee, CC BY 4.0. Latest per-benchmark results selected with partial-run flags retained. (licence) · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · HalluHard: Fan, Delsad, Flammarion and Andriushchenko (EPFL), arXiv:2602.01031 · Horangi 4, Weights & Biases Korea; public publisher-selected numeric evaluation results · JMLE2026-Bench and IgakuQA119 by Naoto Iwase; results licence not stated (code MIT, question data CC BY 4.0). · LLM Chess by Maxim Saplin, Apache-2.0 (Copyright [2024] [Maxim Saplin]) (licence) · MineBench, MIT (licence) · Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results · Outback Yak Research, NepNewsBench v2 research draft. Public numerical aggregates from a candidate preview that the publisher says is not evidence of publication; no release licence assigned and none asserted here. · OpenHands Index (All Hands AI), Apache-2.0 (licence) · Phare (Giskard), CC BY 4.0 (licence) · Sayf-eval. Copyright (c) 2026 Qatar Computing Research Institute (QCRI). MIT License. (licence) · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence) · SWE-bench-Live (Microsoft and contributors), swe-bench-live.github.io. Resolve rates computed from the published report counts. · UzLiB, Tahrirchi (Abror Shopulatov). Public LEADERBOARD.md numbers; repository MIT covers software and documentation, no separate results licence stated and none asserted. · Vectara Hallucination Leaderboard, Apache-2.0 (licence)

Also known as: anthropic-claude-4.6-sonnet, anthropic.claude-sonnet-4-6, anthropic.claude-sonnet-4-6-v1, anthropic.claude-sonnet-4-6-v1:0, anthropic/claude-sonnet-4-6, anthropic/claude-sonnet-4.6, anthropic/claude-sonnet-4.6-batch, anthropic/claude-sonnet-4.6:batch, anthropic/claude-sonnet-4.6:thinking, apac.anthropic.claude-sonnet-4-6, au.anthropic.claude-sonnet-4-6, aws-bedrock/claude-sonnet-4-6, azure_ai/claude-sonnet-4-6, bedrock/claude-sonnet-4-6, bedrock/claude-sonnet-4-6@ap-northeast-1, bedrock/claude-sonnet-4-6@eu-central-1, bedrock/claude-sonnet-4-6@eu-north-1, bedrock/claude-sonnet-4-6@eu-west-1, bedrock/claude-sonnet-4-6@eu-west-3, bedrock/claude-sonnet-4-6@us-east-1, bedrock/claude-sonnet-4-6@us-east-2, bedrock/claude-sonnet-4-6@us-west-2, claude sonnet 4.6, claude-4-6-sonnet, claude-4.6-sonnet, claude-sonnet-4-6, claude-sonnet-4-6-20260217, claude-sonnet-4-6-think, claude-sonnet-4-6-thinking, claude-sonnet-4-6-v1, claude-sonnet-4-6@default, claude-sonnet-4-6@eu, claude-sonnet-4.6-search, databricks-claude-sonnet-4-6, databricks/databricks-claude-sonnet-4-6, deepinfra/anthropic/claude-sonnet-4-6, eu.anthropic.claude-sonnet-4-6, eu.anthropic.claude-sonnet-4-6-v1, eu.anthropic.claude-sonnet-4-6-v1:0, github_copilot/claude-sonnet-4.6, global.anthropic.claude-sonnet-4-6, global.anthropic.claude-sonnet-4-6-v1:0, jp.anthropic.claude-sonnet-4-6, perplexity/anthropic/claude-sonnet-4-6, regional.anthropic.claude-sonnet-4-6-v1:0, replicate/anthropic/claude-sonnet-4.6, snowflake/claude-sonnet-4-6, us.anthropic.claude-sonnet-4-6, us.anthropic.claude-sonnet-4-6-v1, us.anthropic.claude-sonnet-4-6-v1:0, vertex-anthropic/claude-sonnet-4-6, vertex/claude-sonnet-4-6, vertex/claude-sonnet-4-6@europe-west1, vertex/claude-sonnet-4-6@us-east5, vertex_ai/claude-sonnet-4-6, vertex_ai/claude-sonnet-4-6@default

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works