Enter opens the highlighted result. Esc closes.

Claude Haiku 4.5 vs Claude Opus 5

Claude Opus 5 leads Claude Haiku 4.5 on coding, 83.0 to 35.6 on the Vibe Coding Index, and Claude Haiku 4.5 costs 5.0x less per output token ($5.00 vs $25.00 per million). Updated October 22, 2026; 84 benchmarks have a result for both.

Replace Claude Haiku 4.5 Replace Claude Opus 5
Claude Haiku 4.5Anthropic
Coding
35.6
Overall
41.0
Output $/M
$5.00
0
Claude Opus 5Anthropic
Coding
83.0
Overall
81.6
Output $/M
$25.00
84
vs boards won of 84 shared

Claude Haiku 4.5 or Claude Opus 5: which to pick

Pick Claude Opus 5 for coding: it scores 83.0 on the Vibe Coding Index against 35.6 for Claude Haiku 4.5.

Pick Claude Haiku 4.5 for output tokens 5.0x cheaper ($5.00 vs $25.00 per million) and a lower cost per solved task ($0.56 against $1.33).

Model AClaude Haiku 4.5Anthropic35.6/1004 independent benchmarks$0.56 per solved taskModel BClaude Opus 5Anthropic83.0/1006 independent benchmarks$1.33 per solved task
Shared results
84
Claude Haiku 4.5 only
11
Claude Opus 5 only
47
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Claude Haiku 4.5 35.6 from 4 benchmarks, Claude Opus 5 83.0 from 6 benchmarks.
Widest category
Agentic work, 45.2 points apart on average over 10 shared benchmarks.
Widest single gap
ARC-AGI-2: Claude Haiku 4.5 4.0% against Claude Opus 5 90.4%, 86.4 points apart on a 0 to 100 scale.
Output price
Claude Haiku 4.5 $5.00 against Claude Opus 5 $25.00 per million output tokens, 5.0x apart.
Context window
Claude Haiku 4.5 takes 200K tokens, Claude Opus 5 1M.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryClaude Haiku 4.5BarsClaude Opus 5Ties
Coding014
Intelligence02
Agentic work010
Reasoning010
Math03
Knowledge011
Web design09
General025
All shared boards084
CodingLike-for-likeSix of these boards feed the Vibe Coding Index014
Coding benchmarks for Claude Haiku 4.5 and Claude Opus 5
BenchmarkClaude Haiku 4.5Claude Opus 5Leader
SWE-bench Verified66.6%Not measuredOnly Claude Haiku 4.5 measured
SWE-bench Multilingual64.7%Not measuredOnly Claude Haiku 4.5 measured
LiveBench coding48.1%90.8%Claude Opus 5
LMArena WebDev13301689Claude Opus 5
AA Coding Index43.978.0Claude Opus 5
LiveCodeBench (Vals AI)41.2%89.0%Claude Opus 5
SWE-bench Verified (Vals AI)66.6%97.0%Claude Opus 5
Terminal-Bench 4 (Vals AI)Not measured53.5%Only Claude Opus 5 measured
Vibe Code Bench (Vals AI)11.4%88.4%Claude Opus 5
Code migration (Vals AI)7.5%57.5%Claude Opus 5
ProgramBench (Vals AI)0.0%3.0%Claude Opus 5
IOI (Vals AI)Not measured84.3%Only Claude Opus 5 measured
Terminal-Bench 2.1 (Vals AI)43.8%84.6%Claude Opus 5
Vibe Code Bench 1-100 (Vals AI)Not measured28.5%Only Claude Opus 5 measured
SkillsBench (Vals AI)Not measured60.4%Only Claude Opus 5 measured
CyberBench (Vals AI)Not measured65.4%Only Claude Opus 5 measured
SciCodeNot measured56.4%Only Claude Opus 5 measured
Terminal-Bench35.5%Not measuredOnly Claude Haiku 4.5 measured
FrontierCodeNot measured53.4%Only Claude Opus 5 measured
FrontierSWENot measured52.0%Only Claude Opus 5 measured
WeirdML45.4%91.8%Claude Opus 5
ALE-BenchNot measured2164.6Only Claude Opus 5 measured
LMArena Text: coding14541533Claude Opus 5
LMArena Text: software and IT14391527Claude Opus 5
LMArena WebDev: React13261711Claude Opus 5
LMArena WebDev: HTML13181680Claude Opus 5
LMArena WebDev: from an imageNot measured1662Only Claude Opus 5 measured
SciCodeNot measured13.8%Only Claude Opus 5 measured
Neither model has a result on the other 11 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Claude Haiku 4.5 and Claude Opus 5
BenchmarkClaude Haiku 4.5Claude Opus 5Leader
AA Intelligence Index16.950.8Claude Opus 5
Epoch Capabilities Index142.2163.9Claude Opus 5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index010
Agentic work benchmarks for Claude Haiku 4.5 and Claude Opus 5
BenchmarkClaude Haiku 4.5Claude Opus 5Leader
AA Agentic Index8.056.5Claude Opus 5
Design Arena: agents building web appsNot measured1266Only Claude Opus 5 measured
Design Arena: agents building full-stack appsNot measured1306Only Claude Opus 5 measured
Design Arena: agents building mobile appsNot measured1348Only Claude Opus 5 measured
Design Arena: agents building Android appsNot measured1257Only Claude Opus 5 measured
Design Arena: agents building Godot gamesNot measured1279Only Claude Opus 5 measured
Design Arena: agents building gamesNot measured1261Only Claude Opus 5 measured
Design Arena: PowerPoint slidesNot measured1240Only Claude Opus 5 measured
LiveBench research42.2%88.9%Claude Opus 5
LiveBench browser40.0%84.0%Claude Opus 5
LiveBench long agentic11.1%62.1%Claude Opus 5
SREBench (Vals AI)Not measured12.2%Only Claude Opus 5 measured
Finance Agent (Vals AI)31.0%58.6%Claude Opus 5
Tax Agent Bench (Vals AI)5.9%45.5%Claude Opus 5
Legal Research Bench (Vals AI)10.6%55.3%Claude Opus 5
Excel Modeling Benchmark (Vals AI)22.7%73.6%Claude Opus 5
BioMysteryBench (Vals AI)Not measured79.3%Only Claude Opus 5 measured
Terminal-Bench-Science (Vals AI)Not measured27.1%Only Claude Opus 5 measured
Vals Science IndexNot measured47.9%Only Claude Opus 5 measured
Time Horizon Index (Vals AI)Not measured18.8%Only Claude Opus 5 measured
CUA-bench (Vals AI)Not measured9.0%Only Claude Opus 5 measured
APEX-AgentsNot measured65.8%Only Claude Opus 5 measured
OSWorld 2.0Not measured31.4%Only Claude Opus 5 measured
BALROG31.2%63.4%Claude Opus 5
DeepResearch Bench45.5%Not measuredOnly Claude Haiku 4.5 measured
PostTrainBenchNot measured35.0%Only Claude Opus 5 measured
Vending-Bench 2$459$11,182Claude Opus 5
LMArena SearchNot measured1242Only Claude Opus 5 measured
Berkeley Function Calling Leaderboard (BFCL)68.7%Not measuredOnly Claude Haiku 4.5 measured
Neither model has a result on the other 12 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index010
Reasoning benchmarks for Claude Haiku 4.5 and Claude Opus 5
BenchmarkClaude Haiku 4.5Claude Opus 5Leader
GPQA Diamond72.2%93.4%Claude Opus 5
LiveBench reasoning61.7%91.2%Claude Opus 5
MysteryMechanism (Vals AI)Not measured37.4%Only Claude Opus 5 measured
MMMU Pro (Vals AI)46.1%89.9%Claude Opus 5
GPQA Diamond (Artificial Analysis run)Not measured93.7%Only Claude Opus 5 measured
Humanity's Last Exam (Artificial Analysis run)Not measured54.9%Only Claude Opus 5 measured
GPQA Diamond (Epoch AI run)71.2%93.9%Claude Opus 5
ARC-AGI47.7%97.5%Claude Opus 5
ARC-AGI-24.0%90.4%Claude Opus 5
SimpleBenchNot measured80.6%Only Claude Opus 5 measured
Chess puzzles8.0%43.0%Claude Opus 5
Mystery game puzzlesNot measured59.0%Only Claude Opus 5 measured
EBR-benchNot measured45.7%Only Claude Opus 5 measured
LMArena Text: hard prompts14211527Claude Opus 5
LMArena Text: expert14431554Claude Opus 5
LMArena Text: hard prompts in English14351523Claude Opus 5
LMArena Vision: diagramsNot measured1348Only Claude Opus 5 measured
LMArena Vision: homeworkNot measured1338Only Claude Opus 5 measured
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index03
Math benchmarks for Claude Haiku 4.5 and Claude Opus 5
BenchmarkClaude Haiku 4.5Claude Opus 5Leader
AIME82.7%Not measuredOnly Claude Haiku 4.5 measured
MGSM (Vals AI)92.2%Not measuredOnly Claude Haiku 4.5 measured
OTIS Mock AIME66.7%98.9%Claude Opus 5
FrontierMath, tiers 1 to 3Not measured85.6%Only Claude Opus 5 measured
FrontierMath, tier 4Not measured73.2%Only Claude Opus 5 measured
ProofBenchNot measured99.0%Only Claude Opus 5 measured
MATH level 596.4%Not measuredOnly Claude Haiku 4.5 measured
LMArena Text: math13981533Claude Opus 5
LMArena Text: mathematical work14301544Claude Opus 5
Neither model has a result on the other 4 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index011
Knowledge benchmarks for Claude Haiku 4.5 and Claude Opus 5
BenchmarkClaude Haiku 4.5Claude Opus 5Leader
MMLU Pro78.7%91.6%Claude Opus 5
LegalBench (Vals AI)81.2%87.0%Claude Opus 5
CaseLaw (Vals AI)56.5%Not measuredOnly Claude Haiku 4.5 measured
CorpFin (Vals AI)60.6%73.2%Claude Opus 5
TaxEval (Vals AI)67.5%75.1%Claude Opus 5
MortgageTax (Vals AI)62.2%72.1%Claude Opus 5
MedQA (Vals AI)79.6%Not measuredOnly Claude Haiku 4.5 measured
MedCode (Vals AI)32.7%63.6%Claude Opus 5
SimpleQA Verified13.2%59.9%Claude Opus 5
LMArena Text: science14061531Claude Opus 5
LMArena Text: medicine and healthcare13941525Claude Opus 5
LMArena Text: legal and government14001530Claude Opus 5
LMArena Text: business and finance14001501Claude Opus 5
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index09
Web design benchmarks for Claude Haiku 4.5 and Claude Opus 5
BenchmarkClaude Haiku 4.5Claude Opus 5Leader
Design Arena: websites11221309Claude Opus 5
Design Arena: UI components10971346Claude Opus 5
Design Arena: game dev10971337Claude Opus 5
Design Arena: data viz11241329Claude Opus 5
Design Arena: 3D10711334Claude Opus 5
Design Arena: all code categories11141323Claude Opus 5
Design Arena: SVG10251311Claude Opus 5
Design Arena: ASCII art11401352Claude Opus 5
LiveBench design59.9%83.0%Claude Opus 5
GeneralLike-for-likePublic boards, shown as published, kept out of the index025
General benchmarks for Claude Haiku 4.5 and Claude Opus 5
BenchmarkClaude Haiku 4.5Claude Opus 5Leader
LiveBench data analysis59.3%74.6%Claude Opus 5
Public Benefits Bench (Vals AI)54.3%76.9%Claude Opus 5
SAGE (Vals AI)31.8%49.4%Claude Opus 5
MedScribe (Vals AI)85.2%91.0%Claude Opus 5
Vals Multimodal Index42.9%73.9%Claude Opus 5
Long context reasoning (AA-LCR)Not measured82.0%Only Claude Opus 5 measured
DTBench73.6%97.9%Claude Opus 5
LMCA30.9%64.5%Claude Opus 5
Furniture assemblyNot measured60.8%Only Claude Opus 5 measured
ExploitBench13.7%Not measuredOnly Claude Haiku 4.5 measured
LMArena Text13971509Claude Opus 5
LMArena Text: instruction following14141518Claude Opus 5
LMArena Text: creative writing13711493Claude Opus 5
LMArena VisionNot measured1320Only Claude Opus 5 measured
LMArena Text: English14141508Claude Opus 5
LMArena Text: non-English13771502Claude Opus 5
LMArena Text: Chinese14181573Claude Opus 5
LMArena Text: German13741527Claude Opus 5
LMArena Text: Japanese13381520Claude Opus 5
LMArena Text: Korean13461517Claude Opus 5
LMArena Text: French14111519Claude Opus 5
LMArena Text: Spanish14211523Claude Opus 5
LMArena Text: Russian13811507Claude Opus 5
LMArena Text: Polish13431522Claude Opus 5
LMArena Text: multi-turn14091500Claude Opus 5
LMArena Text: longer queries14271517Claude Opus 5
LMArena Text: writing and language13861499Claude Opus 5
LMArena Text: entertainment and media13691481Claude Opus 5
LMArena Vision: EnglishNot measured1315Only Claude Opus 5 measured
LMArena Vision: ChineseNot measured1400Only Claude Opus 5 measured
LMArena Vision: reading textNot measured1329Only Claude Opus 5 measured
LMArena Vision: humorNot measured1293Only Claude Opus 5 measured
LMArena Vision: creative writingNot measured1313Only Claude Opus 5 measured
LMArena Document14201516Claude Opus 5
Neither model has a result on the other 12 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Haiku 4.5 and Claude Opus 5
BenchmarkClaude Haiku 4.5Claude Opus 5Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Claude Haiku 4.5$0.0035Fits in one request
Claude Opus 5$0.018Fits in one request

Claude Haiku 4.5 costs less

Repository review

50K input + 3K output tokens

Claude Haiku 4.5$0.065Fits in one request
Claude Opus 5$0.325Fits in one request

Claude Haiku 4.5 costs less

Agent loop

200K input + 10K output tokens

Claude Haiku 4.5$0.250Exceeds the context window
Claude Opus 5$1.25Fits in one request

Claude Haiku 4.5 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Claude Haiku 4.5 and Claude Opus 5
SpecificationAnthropic Claude Haiku 4.5Anthropic Claude Opus 5
MakerAnthropicAnthropic
WeightsNo public weights listedNo public weights listed
ReleasedOctober 15, 2025July 24, 2026
Context window200K1M
Input $/Mtok$1.00$5.00
Output $/Mtok$5.00$25.00
Coding index35.683.0
Independent measurements behind it46
Cost per solved task$0.56$1.33
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Haiku 4.5 better than Claude Opus 5 for coding?

Pick Claude Opus 5 for coding: it scores 83.0 on the Vibe Coding Index against 35.6 for Claude Haiku 4.5. Pick Claude Haiku 4.5 for output tokens 5.0x cheaper ($5.00 vs $25.00 per million) and a lower cost per solved task ($0.56 against $1.33).

Which is cheaper, Claude Haiku 4.5 or Claude Opus 5?

Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens; Claude Opus 5 costs $5.00 and $25.00. Per solved SWE-bench Verified task, Claude Haiku 4.5 costs $0.56 and Claude Opus 5 $1.33.

Which has the bigger context window, Claude Haiku 4.5 or Claude Opus 5?

Claude Opus 5, with 1M tokens against 200K.

How many benchmarks is the Claude Haiku 4.5 vs Claude Opus 5 comparison based on?

84 benchmarks have a result for both models as of October 22, 2026. Claude Haiku 4.5's coding index rests on 4 independent benchmarks and Claude Opus 5's on 6. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/claude-haiku-4-5-vs-claude-opus-5" title="Claude Haiku 4.5 vs Claude Opus 5, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/claude-haiku-4-5-vs-claude-opus-5">LLMBenchmark.ai, Claude Haiku 4.5 vs Claude Opus 5</a></p>