Enter opens the highlighted result. Esc closes.

Claude Haiku 4.5 vs Claude Opus 4.5

Claude Opus 4.5 leads Claude Haiku 4.5 on coding, 44.0 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; 75 benchmarks have a result for both.

Replace Claude Haiku 4.5 Replace Claude Opus 4.5
Claude Haiku 4.5Anthropic
Coding
35.6
Overall
41.0
Output $/M
$5.00
0
Claude Opus 4.5Anthropic
Coding
44.0
Overall
55.7
Output $/M
$25.00
75
vs boards won of 75 shared

Claude Haiku 4.5 or Claude Opus 4.5: which to pick

Pick Claude Opus 4.5 for coding: it scores 44.0 on the Vibe Coding Index against 35.6 for Claude Haiku 4.5, though on 3 benchmarks against 4.

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 $1.37).

Model AClaude Haiku 4.5Anthropic35.6/1004 independent benchmarks$0.56 per solved taskModel BClaude Opus 4.5Anthropic44.0/1003 independent benchmarks$1.37 per solved task
Shared results
75
Claude Haiku 4.5 only
20
Claude Opus 4.5 only
33
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 4.5 44.0 from 3 benchmarks.
Widest category
Coding, 18.5 points apart on average over 12 shared benchmarks.
Widest single gap
LiveCodeBench (Vals AI): Claude Haiku 4.5 41.2% against Claude Opus 4.5 83.7%, 42.5 points apart on a 0 to 100 scale.
Output price
Claude Haiku 4.5 $5.00 against Claude Opus 4.5 $25.00 per million output tokens, 5.0x apart.
Context window
Both take 200K tokens.

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 4.5Ties
Coding012
Intelligence02
Agentic work04
Reasoning09
Math05
Knowledge013
Web design08
General022
All shared boards075
CodingLike-for-likeSix of these boards feed the Vibe Coding Index012
Coding benchmarks for Claude Haiku 4.5 and Claude Opus 4.5
BenchmarkClaude Haiku 4.5Claude Opus 4.5Leader
SWE-bench Verified66.6%79.2%Claude Opus 4.5
SWE-bench (full test set)Not measured52.6%Only Claude Opus 4.5 measured
SWE-bench Multilingual64.7%70.7%Claude Opus 4.5
LiveBench coding48.1%Not measuredOnly Claude Haiku 4.5 measured
LMArena WebDev13301494Claude Opus 4.5
AA Coding Index43.9Not measuredOnly Claude Haiku 4.5 measured
LiveCodeBench (Vals AI)41.2%83.7%Claude Opus 4.5
SWE-bench Verified (Vals AI)66.6%76.4%Claude Opus 4.5
Vibe Code Bench (Vals AI)11.4%20.6%Claude Opus 4.5
Code migration (Vals AI)7.5%Not measuredOnly Claude Haiku 4.5 measured
ProgramBench (Vals AI)0.0%Not measuredOnly Claude Haiku 4.5 measured
Terminal-Bench 2.1 (Vals AI)43.8%Not measuredOnly Claude Haiku 4.5 measured
LiveCodeBench (Artificial Analysis run)Not measured87.1%Only Claude Opus 4.5 measured
Terminal-Bench Hard (Artificial Analysis run)Not measured47.0%Only Claude Opus 4.5 measured
Terminal-Bench35.5%63.1%Claude Opus 4.5
GSO-BenchNot measured26.5%Only Claude Opus 4.5 measured
WeirdML45.4%63.7%Claude Opus 4.5
CybenchNot measured82.0%Only Claude Opus 4.5 measured
LMArena Text: coding14541503Claude Opus 4.5
LMArena Text: software and IT14391483Claude Opus 4.5
LMArena WebDev: React13261478Claude Opus 4.5
LMArena WebDev: HTML13181510Claude Opus 4.5
Neither model has a result on the other 17 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Claude Haiku 4.5 and Claude Opus 4.5
BenchmarkClaude Haiku 4.5Claude Opus 4.5Leader
AA Intelligence Index16.929.1Claude Opus 4.5
Epoch Capabilities Index142.2150.1Claude Opus 4.5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index04
Agentic work benchmarks for Claude Haiku 4.5 and Claude Opus 4.5
BenchmarkClaude Haiku 4.5Claude Opus 4.5Leader
AA Agentic Index8.0Not measuredOnly Claude Haiku 4.5 measured
Design Arena: agents building web appsNot measured1159Only Claude Opus 4.5 measured
Design Arena: agents building full-stack appsNot measured1156Only Claude Opus 4.5 measured
Design Arena: agents building mobile appsNot measured1189Only Claude Opus 4.5 measured
Design Arena: agents building Android appsNot measured1176Only Claude Opus 4.5 measured
LiveBench research42.2%Not measuredOnly Claude Haiku 4.5 measured
LiveBench browser40.0%Not measuredOnly Claude Haiku 4.5 measured
LiveBench long agentic11.1%Not measuredOnly Claude Haiku 4.5 measured
Finance Agent (Vals AI)31.0%Not measuredOnly Claude Haiku 4.5 measured
Tax Agent Bench (Vals AI)5.9%Not measuredOnly Claude Haiku 4.5 measured
Legal Research Bench (Vals AI)10.6%Not measuredOnly Claude Haiku 4.5 measured
Excel Modeling Benchmark (Vals AI)22.7%Not measuredOnly Claude Haiku 4.5 measured
τ²-Bench TelecomNot measured89.5%Only Claude Opus 4.5 measured
METR time horizonsNot measured75.0%Only Claude Opus 4.5 measured
Remote Labor IndexNot measured3.8%Only Claude Opus 4.5 measured
GDPvalNot measured45.5%Only Claude Opus 4.5 measured
BALROG31.2%43.5%Claude Opus 4.5
DeepResearch Bench45.5%54.8%Claude Opus 4.5
OSWorldNot measured66.3%Only Claude Opus 4.5 measured
Vending-Bench 2$459$4,967Claude Opus 4.5
Berkeley Function Calling Leaderboard (BFCL)68.7%77.5%Claude Opus 4.5
Neither model has a result on the other 20 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index09
Reasoning benchmarks for Claude Haiku 4.5 and Claude Opus 4.5
BenchmarkClaude Haiku 4.5Claude Opus 4.5Leader
GPQA Diamond72.2%85.9%Claude Opus 4.5
LiveBench reasoning61.7%Not measuredOnly Claude Haiku 4.5 measured
MMMU Pro (Vals AI)46.1%83.0%Claude Opus 4.5
GPQA Diamond (Artificial Analysis run)Not measured86.6%Only Claude Opus 4.5 measured
Humanity's Last Exam (Artificial Analysis run)Not measured30.1%Only Claude Opus 4.5 measured
GPQA Diamond (Epoch AI run)71.2%86.0%Claude Opus 4.5
ARC-AGI47.7%80.0%Claude Opus 4.5
ARC-AGI-24.0%37.6%Claude Opus 4.5
SimpleBenchNot measured62.0%Only Claude Opus 4.5 measured
Chess puzzles8.0%14.0%Claude Opus 4.5
Mystery game puzzlesNot measured22.0%Only Claude Opus 4.5 measured
EBR-benchNot measured14.3%Only Claude Opus 4.5 measured
VPCTNot measured40.0%Only Claude Opus 4.5 measured
LMArena Text: hard prompts14211476Claude Opus 4.5
LMArena Text: expert14431487Claude Opus 4.5
LMArena Text: hard prompts in English14351482Claude Opus 4.5
Kagi LLM BenchmarkNot measured80.2%Only Claude Opus 4.5 measured
Neither model has a result on the other 5 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index05
Math benchmarks for Claude Haiku 4.5 and Claude Opus 4.5
BenchmarkClaude Haiku 4.5Claude Opus 4.5Leader
AIME82.7%95.4%Claude Opus 4.5
MGSM (Vals AI)92.2%95.2%Claude Opus 4.5
AA Math IndexNot measured91.3Only Claude Opus 4.5 measured
AIME 2025 (Artificial Analysis run)Not measured91.3%Only Claude Opus 4.5 measured
OTIS Mock AIME66.7%86.1%Claude Opus 4.5
FrontierMath, tiers 1 to 3Not measured34.4%Only Claude Opus 4.5 measured
FrontierMath, tier 4Not measured4.9%Only Claude Opus 4.5 measured
ProofBenchNot measured36.0%Only Claude Opus 4.5 measured
MATH level 596.4%Not measuredOnly Claude Haiku 4.5 measured
LMArena Text: math13981463Claude Opus 4.5
LMArena Text: mathematical work14301468Claude Opus 4.5
Neither model has a result on the other 2 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index013
Knowledge benchmarks for Claude Haiku 4.5 and Claude Opus 4.5
BenchmarkClaude Haiku 4.5Claude Opus 4.5Leader
MMLU Pro78.7%87.3%Claude Opus 4.5
LegalBench (Vals AI)81.2%84.6%Claude Opus 4.5
CaseLaw (Vals AI)56.5%62.6%Claude Opus 4.5
CorpFin (Vals AI)60.6%65.1%Claude Opus 4.5
TaxEval (Vals AI)67.5%74.9%Claude Opus 4.5
MortgageTax (Vals AI)62.2%68.7%Claude Opus 4.5
MedQA (Vals AI)79.6%95.9%Claude Opus 4.5
MedCode (Vals AI)32.7%49.2%Claude Opus 4.5
MMLU Pro (Artificial Analysis run)Not measured89.5%Only Claude Opus 4.5 measured
SimpleQA Verified13.2%45.7%Claude Opus 4.5
GeoBenchNot measured75.0%Only Claude Opus 4.5 measured
LMArena Text: science14061454Claude Opus 4.5
LMArena Text: medicine and healthcare13941444Claude Opus 4.5
LMArena Text: legal and government14001459Claude Opus 4.5
LMArena Text: business and finance14001448Claude Opus 4.5
FACTS Benchmark SuiteNot measured43.9%Only Claude Opus 4.5 measured
Neither model has a result on the other 2 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index08
Web design benchmarks for Claude Haiku 4.5 and Claude Opus 4.5
BenchmarkClaude Haiku 4.5Claude Opus 4.5Leader
Design Arena: websites11221247Claude Opus 4.5
Design Arena: UI components10971235Claude Opus 4.5
Design Arena: game dev10971228Claude Opus 4.5
Design Arena: data viz11241243Claude Opus 4.5
Design Arena: 3D10711213Claude Opus 4.5
Design Arena: all code categories11141242Claude Opus 4.5
Design Arena: SVG10251180Claude Opus 4.5
Design Arena: ASCII art11401190Claude Opus 4.5
LiveBench design59.9%Not measuredOnly Claude Haiku 4.5 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index022
General benchmarks for Claude Haiku 4.5 and Claude Opus 4.5
BenchmarkClaude Haiku 4.5Claude Opus 4.5Leader
LiveBench data analysis59.3%Not measuredOnly Claude Haiku 4.5 measured
Public Benefits Bench (Vals AI)54.3%Not measuredOnly Claude Haiku 4.5 measured
SAGE (Vals AI)31.8%52.1%Claude Opus 4.5
MedScribe (Vals AI)85.2%85.3%Claude Opus 4.5
Vals Multimodal Index42.9%Not measuredOnly Claude Haiku 4.5 measured
IFBenchNot measured58.0%Only Claude Opus 4.5 measured
Long context reasoning (AA-LCR)Not measured77.3%Only Claude Opus 4.5 measured
DTBench73.6%89.9%Claude Opus 4.5
LMCA30.9%44.5%Claude Opus 4.5
CL-benchNot measured21.1%Only Claude Opus 4.5 measured
Furniture assemblyNot measured28.3%Only Claude Opus 4.5 measured
ExploitBench13.7%Not measuredOnly Claude Haiku 4.5 measured
LMArena Text13971451Claude Opus 4.5
LMArena Text: instruction following14141478Claude Opus 4.5
LMArena Text: creative writing13711445Claude Opus 4.5
LMArena Text: English14141461Claude Opus 4.5
LMArena Text: non-English13771438Claude Opus 4.5
LMArena Text: Chinese14181470Claude Opus 4.5
LMArena Text: German13741449Claude Opus 4.5
LMArena Text: Japanese13381417Claude Opus 4.5
LMArena Text: Korean13461424Claude Opus 4.5
LMArena Text: French14111471Claude Opus 4.5
LMArena Text: Spanish14211457Claude Opus 4.5
LMArena Text: Russian13811447Claude Opus 4.5
LMArena Text: Polish13431439Claude Opus 4.5
LMArena Text: multi-turn14091466Claude Opus 4.5
LMArena Text: longer queries14271479Claude Opus 4.5
LMArena Text: writing and language13861452Claude Opus 4.5
LMArena Text: entertainment and media13691434Claude Opus 4.5
LMArena Document14201462Claude Opus 4.5
Neither model has a result on the other 16 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 4.5
BenchmarkClaude Haiku 4.5Claude Opus 4.5Leader
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 4.5$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 4.5$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 4.5$1.25Exceeds the context window

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 4.5
SpecificationAnthropic Claude Haiku 4.5Anthropic Claude Opus 4.5
MakerAnthropicAnthropic
WeightsNo public weights listedNo public weights listed
ReleasedOctober 15, 2025November 24, 2025
Context window200K200K
Input $/Mtok$1.00$5.00
Output $/Mtok$5.00$25.00
Coding index35.644.0
Independent measurements behind it43
Cost per solved task$0.56$1.37
SourceOpenRouter listingOpenRouter listing

Questions

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

Pick Claude Opus 4.5 for coding: it scores 44.0 on the Vibe Coding Index against 35.6 for Claude Haiku 4.5, though on 3 benchmarks against 4. 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 $1.37).

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

Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens; Claude Opus 4.5 costs $5.00 and $25.00. Per solved SWE-bench Verified task, Claude Haiku 4.5 costs $0.56 and Claude Opus 4.5 $1.37.

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

Neither: both take 200K tokens.

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

75 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 4.5's on 3. 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-4-5" title="Claude Haiku 4.5 vs Claude Opus 4.5, 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-4-5">LLMBenchmark.ai, Claude Haiku 4.5 vs Claude Opus 4.5</a></p>