Skip to main content

AnthropicLLM matchup

Claude Opus 5vs.Claude Sonnet 5

Is the higher price worth it?

Claude Opus 5 performs better than Claude Sonnet 5, but costs 2.5 times as much.

Opus 5 ahead
IntelligenceOverall index, 100 points = average
117.3

Claude Opus 5

Rank 10

92.0

Claude Sonnet 5

Rank 45

about as fast
SpeedTokens per second
68

Claude Opus 5

75

Claude Sonnet 5

Sonnet 5 cheaper
Price$ per 1M tokens
$10

Claude Opus 5

$4

Claude Sonnet 5

about as popular
Popularity0 to 100
81

Claude Opus 5

Rank 18 of 508

80

Claude Sonnet 5

Rank 20 of 508

Choose Claude Opus 5 for

HealthKnowledgeGeneralMultimodalCodingMathematicsLegalReasoningAgentsFinance

No clear advantage for Claude Sonnet 5.

Where is the difference?

Strengths at a glance

Opus 5Sonnet 5Median of all ranked models

9511513575CodingOpus 5ReasoningOpus 5MathematicsOpus 5KnowledgeOpus 5AgentsOpus 5GeneralOpus 5MultimodalOpus 5FinanceOpus 5LegalOpus 5HealthOpus 5
CodingOpus 5 ahead
Opus 5118.3Rank 9
Sonnet 594.1Rank 48
Median of ranked models87.5

95 % range: Opus 5 114.6 to 122.1, Sonnet 5 89.7 to 98.4.

AreaOpus 5Sonnet 5

The same values and verdicts as the bars below. In the close-up the net starts at 75 points, just below the weaker model's lowest value; the reference group average is 100.

Opus 5 is ahead in 10 of 10 areas, Sonnet 5 in none.

Opus 5 ahead: 10

Areas in the overall index
KnowledgeOpus 5 +38121.9 to 84.3
GeneralOpus 5 +29116.4 to 87.5
MultimodalOpus 5 +26115.5 to 89.1
CodingOpus 5 +24118.3 to 94.1
MathematicsOpus 5 +21113.5 to 92.8
ReasoningOpus 5 +17107.6 to 90.2
AgentsOpus 5 +17112.0 to 94.7
Specialist fields
HealthOpus 5 +40130.0 to 89.7
LegalOpus 5 +18116.3 to 97.8
FinanceOpus 5 +17119.0 to 102.0

Bar length shows the gap in index points (full width 15 points, one standard deviation of the reference group). Solid means clearly ahead, hatched not certain yet, a thin grey bar level.

Show 3 more areas that do not count toward the overall index
Productivitytoo little data132.5 to 99.1
Cybersecuritytoo little data112.7 to 90.6
Searchonly Opus 5 has a score

Bar length shows the gap in index points (full width 15 points, one standard deviation of the reference group). Solid means clearly ahead, hatched not certain yet, a thin grey bar level.

Claude Opus 5 vs. Claude Sonnet 5 at the same effort

Both models measured at the same reasoning efforts. This shows whether a lead comes from the model or only from more compute.

maxOpus 5 ahead
18.1 points gap across 3 shared test series
Overall index (max vs. max)Opus 5 ahead
Opus 5 117.3 vs. Sonnet 5 92.0

The gap compares only the Artificial Analysis test series in which both models are measured at that effort. Tied below 3 points, ahead only beyond two standard errors.

Which model for which task?

Everyday tasks, translated from the area values above. Each row names the values it rests on.

Writing and debugging codeOpus 5Coding 118.3 to 94.1
Maths and data analysisOpus 5Mathematics 113.5 to 92.8, Reasoning 107.6 to 90.2
Facts and researchOpus 5Knowledge 121.9 to 84.3, Search without a score for Sonnet 5
Agents and workflowsOpus 5Agents 112.0 to 94.7
Fast chat repliesabout the same68 to 75 tokens per second
Security analysisopenCybersecurity 112.7 to 90.6 (too little data)
Many requests on a small budgetSonnet 5$10 to $4 per 1M tokens
Finance, legal and medical textsOpus 5Finance 119.0 to 102.0, Legal 116.3 to 97.8, Health 130.0 to 89.7
Images, screenshots and PDFsOpus 5Multimodal 115.5 to 89.1

Claude Opus 5 vs. Claude Sonnet 5 in our own test

The same prompt for both models. We show the latest attempt without fixes; it does not count toward the overall index.

Pelican on a bicycle

Prompt

Generate an SVG of a pelican riding a bicycle

Task origin: Simon Willison

Claude Opus 5

Claude Opus 5: Pelican on a bicycle
A community classic for free-form SVG drawing.

Oct 1, 2026 · 63 s · $0.14

Test details
Run date
Oct 1, 2026
API provider
openrouter
Requested API model
anthropic/claude-opus-5
Reasoning setting
high
Token limit including reasoning
8,192 tokens
Output tokens
5,423 tokens
Cost of the attempt
$0.1357
Response time
63 s
API model reported in response
anthropic/claude-opus-5
Pinned endpoint
Anthropic | anthropic/claude-opus-5-20260723
Temperature
Provider default
Test protocol
community-visual-tests-v1
Technical checks
Passed, not a quality rating

Visually reviewed. This individual test does not establish overall model quality.

Earlier attempts (1)

  • Sep 7, 2026API request failedReasoning setting: high

Claude Sonnet 5

Claude Sonnet 5: Pelican on a bicycle
A community classic for free-form SVG drawing.

Sep 7, 2026 · 48 s · $0.03

Test details
Run date
Sep 7, 2026
API provider
openrouter
Requested API model
anthropic/claude-sonnet-5
Reasoning setting
high
Token limit including reasoning
8,192 tokens
Output tokens
3,394 tokens
Cost of the attempt
$0.0340
Response time
47.9 s
API model reported in response
anthropic/claude-sonnet-5
Pinned endpoint
Anthropic | anthropic/claude-sonnet-5-20260630
Temperature
Provider default
Test protocol
community-visual-tests-v1
Technical checks
Passed, not a quality rating

Visually reviewed. This individual test does not establish overall model quality.

SVG short test

SVG short test

One SVG task with the same fixed settings for every model, without hidden thinking.

Prompt

Return only one standalone SVG of a pelican riding a bicycle. Use viewBox="0 0 800 600", simple shapes, two clearly visible wheels and a recognizable pelican with a long beak. At most 30 SVG elements and 1800 characters. No Markdown, explanations, comments, scripts, animation, CSS or external resources.

Claude Opus 5

Unavailable in this round

Claude Sonnet 5

Unavailable in this round

Claude Opus 5 and Claude Sonnet 5 in the overall index

Every model with a place in the overall index, both models of this matchup highlighted. You can add or hide other models.

Overall index (Gradually Meta-Score v6)

20 of 95 models

Claude Fable 5.2max130.9
Gemini 4 Carbonhigh127.0
Claude Opus 5.5max125.7
Claude Fable 5.1max124.1
Claude Sonnet 5.5max121.9
Gemini 4 Argonhigh121.2
GPT-6 Astramax119.6
Muse Wonder Fullmax118.4
GPT-6.1 Solmax117.5
Claude Opus 5max117.3
Muse Wondermax113.7
Claude Fable 5max112.8
Kimi K3.1 Promax112.6
Muse Ember 1.0max110.5
GPT-5.6 Solmax109.3
GPT-6 Solmax108.4
Gemini 3.7 Flashhigh107.7
DeepSeek V4.1 Flash 1017max107.1
GPT-6.1 Terramax106.5
Claude Sonnet 5max92.0

Popularity compared

How often the models are searched for, used and mentioned, on a scale from 0 to 100.

Both are about as popular: Claude Opus 5 81, Claude Sonnet 5 80 out of 100.

Over time

Weekly values from October 18, 2026 to November 16, 2026

Claude Opus 5Claude Sonnet 5

All data side by side

The better value is bold. For the overall index only when the gap is clear.

MetricClaude Opus 5Claude Sonnet 5
Intelligence
Overall indexpoints, 100 = reference group average, higher is better117.392.0
RankRank 10 of 95Rank 45 of 95
Price per 1M tokens
Inputwhat you send$5$2
Outputwhat the model writes$25$10
Cost per task (own tests)average of 3 original-protocol tasks both models passed (SVG short test not included)$0.14$0.057
Speed
Tokens per second6875
Popularity
Popularitysearch, usage and mentions, 0 to 10081Rank 18 of 50880Rank 20 of 508
Technical
Context windowhow much text the model reads at once1 M tokens≈ 1,500 A4 pages1 M tokens≈ 1,500 A4 pages
Knowledge cutoffMay 2026January 2026
ReleasedJul 24, 2026Jun 30, 2026
AvailabilityActiveActive
Open weightsnono

Prices without caching or batch discounts · Page count as a rule of thumb from Artificial Analysis

Is there a better choice?

Successor of Opus 5Claude Opus 5.5

Overall index 125.7, rank 3$4 input and $20 output per 1M tokens

Sonnet 5 vs. Claude Opus 5.5
Successor of Sonnet 5Claude Sonnet 5.5

Overall index 121.9, rank 5$2 input and $10 output per 1M tokens

Opus 5 vs. Claude Sonnet 5.5
As cheap as Sonnet 5Gemini 4 Carbon

Overall index 127.0, rank 2$2 input and $10 output per 1M tokens

Opus 5 vs. Gemini 4 Carbon

Shared benchmarks

Single results from the same benchmark version, task and metric. A value is bold only with the same test setup and a clear gap.

A: Claude Opus 5B: Claude Sonnet 5

AA-AnalystAgentOverall · higher is betterClaude Opus 5: 53.8%Claude Sonnet 5: 46.3%Artificial AnalysisAs of 11/07/2026
Code MigrationOverall · higher is betterClaude Opus 5: 57.5%Claude Sonnet 5: 44.4%Vals AIAs of 10/29/2026
CorpFin v2Overall · higher is betterTest setup not documented as identicalClaude Opus 5: 73.2%Claude Sonnet 5: 67.0%Vals AIAs of 08/12/2026
CyberBench v1.1Overall · higher is betterClaude Opus 5: 65.4%Claude Sonnet 5: 61.9%Vals AIAs of 11/05/2026
EMBOverall · higher is betterClaude Opus 5: 73.6%Claude Sonnet 5: 66.3%Vals AIAs of 10/20/2026
Finance Agent (v2)Overall · higher is betterClaude Opus 5: 58.6%Claude Sonnet 5: 53.9%Vals AIAs of 10/29/2026
GDPval-AAOverall · higher is betterClaude Opus 5: 1,729.9 eloClaude Sonnet 5: 1,470.4 eloArtificial AnalysisAs of 11/07/2026
GPQA DiamondOverall · higher is betterTest setup not documented as identicalClaude Opus 5: 93.4%Claude Sonnet 5: 88.9%Vals AIAs of 09/01/2026
IOIOverall · higher is betterClaude Opus 5: 84.3%Claude Sonnet 5: 45.0%Vals AIAs of 10/29/2026
Legal Research BenchOverall · All-pass · higher is betterClaude Opus 5: 55.3%Claude Sonnet 5: 41.8%Vals AIAs of 11/05/2026
Show 54 more benchmarks
LegalBenchOverall · higher is betterClaude Opus 5: 87.0%Claude Sonnet 5: 83.9%Vals AIAs of 10/13/2026
LiveCodeBenchOverall · higher is betterClaude Opus 5: 89.0%Claude Sonnet 5: 82.4%Vals AIAs of 09/01/2026
LMArena Code Arena WebDevOverall · higher is betterTest setup not documented as identicalClaude Opus 5: 1,691Claude Sonnet 5: 1,539LMArenaAs of 11/15/2026
LMArena Text, Style ControlOverall · higher is betterTest setup not documented as identicalClaude Opus 5: 1,489Claude Sonnet 5: 1,461LMArenaAs of 11/11/2026
LMArena Vision, Style ControlOverall · higher is betterClaude Opus 5: 1,285Claude Sonnet 5: 1,261LMArenaAs of 11/09/2026
MedCodeOverall · higher is betterClaude Opus 5: 63.6%Claude Sonnet 5: 47.5%Vals AIAs of 10/29/2026
MedScribeOverall · higher is betterClaude Opus 5: 91.0%Claude Sonnet 5: 76.1%Vals AIAs of 10/29/2026
MMLU ProOverall · higher is betterClaude Opus 5: 91.6%Claude Sonnet 5: 87.5%Vals AIAs of 09/01/2026
MMMU ProOverall · higher is betterClaude Opus 5: 89.9%Claude Sonnet 5: 83.0%Vals AIAs of 09/01/2026
MMMU-ProOverall · higher is betterClaude Opus 5: 84.7%Claude Sonnet 5: 77.3%Artificial AnalysisAs of 11/07/2026
MortgageTaxOverall · higher is betterTest setup not documented as identicalClaude Opus 5: 72.1%Claude Sonnet 5: 70.0%Vals AIAs of 09/01/2026
ProgramBenchOverall · higher is betterClaude Opus 5: 3%Claude Sonnet 5: 0%Vals AIAs of 11/05/2026
ProofBench v1.1Overall · higher is betterClaude Opus 5: 99%Claude Sonnet 5: 77%Vals AIAs of 10/29/2026
Public Benefits Bench v1.1Overall · higher is betterClaude Opus 5: 76.9%Claude Sonnet 5: 66.0%Vals AIAs of 10/12/2026
SkillsBenchOverall · higher is betterClaude Opus 5: 60.4%Claude Sonnet 5: 46.5%Vals AIAs of 10/29/2026
SRE BenchOverall · higher is betterClaude Opus 5: 12.2%Claude Sonnet 5: 0.4%Vals AIAs of 11/05/2026
SWE-bench VerifiedOverall · higher is betterTest setup not documented as identicalClaude Opus 5: 97.0%Claude Sonnet 5: 79.6%Vals AIAs of 09/01/2026
Taste-BenchOverall · higher is betterTest setup not documented as identicalClaude Opus 5: 55.5%Claude Sonnet 5: 51.6%Taste-Bench paperAs of 09/23/2026
Tax Agent BenchControversy & Precedence Analysis · higher is betterClaude Opus 5: 23.3%Claude Sonnet 5: 10.0%Vals AIAs of 11/05/2026
TaxEval v2Overall · higher is betterTest setup not documented as identicalClaude Opus 5: 75.1%Claude Sonnet 5: 75.6%Vals AIAs of 09/01/2026
Terminal-Bench 4.0Overall · higher is betterClaude Opus 5: 53.5%Claude Sonnet 5: 10.6%Vals AIAs of 11/05/2026
Terminal-Bench 4.0Overall · higher is betterClaude Opus 5: 51.8%Claude Sonnet 5: 12.4%Terminal-BenchAs of 10/06/2026
Terminal-Bench-ScienceOverall · higher is betterClaude Opus 5: 27.1%Claude Sonnet 5: 5.7%Vals AIAs of 11/05/2026
Vals Multimodal IndexOverall · higher is betterTest setup not documented as identicalClaude Opus 5: 73.9%Claude Sonnet 5: 68.8%Vals AIAs of 08/11/2026
Vibe Code Bench 1-100Overall · higher is betterClaude Opus 5: 28.5%Claude Sonnet 5: 13.8%Vals AIAs of 10/29/2026
Vibe Code Bench v1.1Overall · higher is betterTest setup not documented as identicalClaude Opus 5: 88.4%Claude Sonnet 5: 81.3%Vals AIAs of 11/05/2026
Code MigrationCLI · higher is betterClaude Opus 5: 53.3%Claude Sonnet 5: 36.3%Vals AIAs of 10/29/2026
CorpFin v2Exact Pages · higher is betterTest setup not documented as identicalClaude Opus 5: 68.6%Claude Sonnet 5: 63.9%Vals AIAs of 08/12/2026
CyberBench v1.1Patch · higher is betterClaude Opus 5: 85.7%Claude Sonnet 5: 82.1%Vals AIAs of 11/05/2026
EMBComps · higher is betterClaude Opus 5: 75.2%Claude Sonnet 5: 69.4%Vals AIAs of 10/20/2026
Finance Agent (v2)Adjustments · higher is betterClaude Opus 5: 54.6%Claude Sonnet 5: 48.7%Vals AIAs of 10/29/2026
GPQA DiamondFew-Shot CoT · higher is betterTest setup not documented as identicalClaude Opus 5: 92.4%Claude Sonnet 5: 87.4%Vals AIAs of 09/01/2026
IOIIOI 2024 · higher is betterClaude Opus 5: 100.0%Claude Sonnet 5: 50.5%Vals AIAs of 10/29/2026
Legal Research BenchAdministrative / Regulatory · All-pass · higher is betterClaude Opus 5: 65.2%Claude Sonnet 5: 54.5%Vals AIAs of 11/05/2026
LegalBenchConclusion Tasks · higher is betterClaude Opus 5: 91.6%Claude Sonnet 5: 89.6%Vals AIAs of 10/13/2026
LiveCodeBenchEasy · higher is betterClaude Opus 5: 99.1%Claude Sonnet 5: 98.4%Vals AIAs of 09/01/2026
LMArena Code Arena WebDevHTML development · higher is betterTest setup not documented as identicalClaude Opus 5: 1,679Claude Sonnet 5: 1,526LMArenaAs of 11/15/2026
LMArena Text, Style ControlBusiness, management and finance · higher is betterTest setup not documented as identicalClaude Opus 5: 1,483Claude Sonnet 5: 1,466LMArenaAs of 11/11/2026
LMArena Vision, Style ControlChinese · higher is betterClaude Opus 5: 1,364Claude Sonnet 5: 1,319LMArenaAs of 11/09/2026
MMLU ProBiology · higher is betterClaude Opus 5: 95.7%Claude Sonnet 5: 93.6%Vals AIAs of 09/01/2026
MortgageTaxNumerical Extraction · higher is betterTest setup not documented as identicalClaude Opus 5: 81.4%Claude Sonnet 5: 78.7%Vals AIAs of 09/01/2026
ProgramBenchAlmost Resolved · higher is betterClaude Opus 5: 41.5%Claude Sonnet 5: 13.5%Vals AIAs of 11/05/2026
SkillsBenchNo Skills · higher is betterClaude Opus 5: 45.2%Claude Sonnet 5: 35.1%Vals AIAs of 10/29/2026
SRE BenchCapability Score · higher is betterClaude Opus 5: 31.1%Claude Sonnet 5: 7.7%Vals AIAs of 11/05/2026
SWE-bench Verified<15 min fix · higher is betterTest setup not documented as identicalClaude Opus 5: 98.5%Claude Sonnet 5: 83.5%Vals AIAs of 09/01/2026
Tax Agent BenchFact-Pattern Analysis · higher is betterClaude Opus 5: 48.8%Claude Sonnet 5: 27.1%Vals AIAs of 11/05/2026
TaxEval v2Correctness · higher is betterTest setup not documented as identicalClaude Opus 5: 62%Claude Sonnet 5: 64%Vals AIAs of 09/01/2026
Terminal-Bench 4.0Hardware · higher is betterClaude Opus 5: 33.3%Claude Sonnet 5: 0.0%Vals AIAs of 11/05/2026
Terminal-Bench-ScienceEarth Sciences · higher is betterClaude Opus 5: 12.5%Claude Sonnet 5: 0.0%Vals AIAs of 11/05/2026
Vals Multimodal IndexCorpFin v2 · higher is betterTest setup not documented as identicalClaude Opus 5: 74.9%Claude Sonnet 5: 67.9%Vals AIAs of 08/11/2026
Code MigrationCOBOL · higher is betterClaude Opus 5: 70.0%Claude Sonnet 5: 68.6%Vals AIAs of 10/29/2026
CorpFin v2Max Fitting Context · higher is betterTest setup not documented as identicalClaude Opus 5: 76.0%Claude Sonnet 5: 69.1%Vals AIAs of 08/12/2026
CyberBench v1.1PoC · higher is betterClaude Opus 5: 45.0%Claude Sonnet 5: 41.7%Vals AIAs of 11/05/2026
EMBDataroom · higher is betterClaude Opus 5: 87.7%Claude Sonnet 5: 84.9%Vals AIAs of 10/20/2026

More comparisons

Compare Claude Opus 5 and Claude Sonnet 5 with the models people compare them with most often.

Compare another matchup

The most popular comparisons

  1. 1Claude Sonnet 5.5 vs. Claude Opus 5
  2. 2Claude Sonnet 5 vs. DeepSeek-V4.1-Flash
  3. 3Claude Opus 5.5 vs. Claude Sonnet 5
  4. 4Claude Sonnet 5 vs. GPT-6 Luna
  5. 5Claude Opus 5 vs. DeepSeek-V4.1-Flash
  6. 6Claude Opus 5 vs. Gemini 3.8 Flash
  7. 7Claude Opus 5 vs. Kimi K3
  8. 8Claude Haiku 5.5 vs. Claude Sonnet 5
  9. 9Claude Opus 5 vs. GPT-6 Astra
Show all 54 comparisons
  1. 10Claude Opus 5 vs. GLM-5.3-Flash
  2. 11Claude Opus 5 vs. DeepSeek-V4-Pro
  3. 12Claude Opus 5 vs. Qwen 3.8 27B
  4. 13Claude Opus 5 vs. GLM-5.3
  5. 14Claude Sonnet 5 vs. DeepSeek V4.1 Flash 1017
  6. 15Claude Opus 5 vs. Grok 4.7
  7. 16Claude Opus 5 vs. GPT-5.6 Terra
  8. 17Claude Opus 5 vs. Grok 4.6
  9. 18Claude Opus 5 vs. DeepSeek V4.1 Flash 1017
  10. 19Claude Opus 5 vs. GPT-5.6 Luna
  11. 20Claude Sonnet 5 vs. GPT-6 Astra
  12. 21Claude Haiku 5.5 vs. Claude Opus 5
  13. 22Claude Opus 5 vs. Muse Spark 1.3
  14. 23Claude Sonnet 5 vs. Gemini 3.8 Flash
  15. 24Claude Sonnet 5 vs. Kimi K3
  16. 25Claude Sonnet 5 vs. GPT-5.6 Luna
  17. 26Claude Sonnet 5 vs. GLM-5.3
  18. 27Claude Opus 5 vs. Mistral Large 4
  19. 28Claude Sonnet 5 vs. DeepSeek-V4-Flash
  20. 29Claude Sonnet 5 vs. Qwen 3.8 27B
  21. 30Claude Sonnet 5 vs. GPT-5.6 Sol
  22. 31Claude Sonnet 5 vs. GPT-6 Sol
  23. 32Claude Sonnet 5 vs. Qwen 3.8 Max 0902
  24. 33Claude Fable 5.2 vs. Claude Opus 5
  25. 34Claude Sonnet 5 vs. DeepSeek-V4-Pro
  26. 35Claude Opus 5 vs. GPT-6 Sol
  27. 36Claude Opus 5 vs. Qwen 3.8 Max 0902
  28. 37Claude Sonnet 5 vs. MiMo-V2.5
  29. 38Claude Sonnet 5 vs. Mistral Large 4
  30. 39Claude Opus 5 vs. Claude Sonnet 4.5
  31. 40Claude Fable 5.1 vs. Claude Sonnet 5
  32. 41Claude Sonnet 5 vs. Gemini 3.7 Flash
  33. 42Claude Sonnet 5 vs. Claude Haiku 4.5
  34. 43Claude Sonnet 5 vs. Kimi K3.1 Pro
  35. 44Claude Opus 5 vs. Kimi K3.1 Pro
  36. 45Claude Sonnet 5 vs. GPT-5.5
  37. 46Claude Sonnet 5 vs. Gemini 4 Argon
  38. 47Claude Opus 5 vs. Gemini 4 Carbon
  39. 48Claude Sonnet 5 vs. GPT-6.1 Sol
  40. 49Claude Opus 5 vs. GPT-6.1 Sol
  41. 50Claude Opus 5 vs. Gemini 4 Argon
  42. 51Claude Sonnet 5 vs. GPT-6.1 Terra
  43. 52Claude Sonnet 5 vs. Gemini 4 Carbon
  44. 53Claude Opus 5 vs. GPT-6.1 Terra
  45. 54Claude Opus 5 vs. Muse Wonder Full