Claude Haiku 4.5 vs Claude Opus 5.5
Claude Opus 5.5 leads Claude Haiku 4.5 on coding, 88.8 to 35.6 on the Vibe Coding Index, and Claude Haiku 4.5 costs 4.0x less per output token ($5.00 vs $20.00 per million). Updated October 22, 2026; 65 benchmarks have a result for both.
Claude Haiku 4.5 or Claude Opus 5.5: which to pick
Pick Claude Opus 5.5 for coding: it scores 88.8 on the Vibe Coding Index against 35.6 for Claude Haiku 4.5.
Pick Claude Haiku 4.5 for output tokens 4.0x cheaper ($5.00 vs $20.00 per million).
- Shared results
- 65
- Claude Haiku 4.5 only
- 30
- Claude Opus 5.5 only
- 44
- 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.
Coding
Vibe Coding Index, six public coding benchmarks
Claude Opus 5.5
Claude Opus 5.5 scores 88.8 against 35.6.
4 and 5 benchmarks behind the scores
Intelligence
AA Intelligence Index
Claude Opus 5.5
Claude Opus 5.5 leads, 57.6 against 16.9.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Claude Haiku 4.5 has a AA Agentic Index result, 8.0.
Directional only
Reasoning
GPQA Diamond
No call
Only Claude Haiku 4.5 has a GPQA Diamond result, 72.2%.
Directional only
Math
AIME
No call
Only Claude Haiku 4.5 has a AIME result, 82.7%.
Directional only
Knowledge
MMLU Pro
No call
Only Claude Haiku 4.5 has a MMLU Pro result, 78.7%.
Directional only
Web design
Design Arena: websites
Claude Opus 5.5
Claude Opus 5.5 leads, 1335 against 1122.
Same board, both measured
General
Vals Index
No call
Only Claude Opus 5.5 has a Vals Index result, 70.2%.
Directional only
Long documents
Listed context window
Claude Opus 5.5
Claude Opus 5.5 takes 1M tokens against 200K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Claude Haiku 4.5
Claude Haiku 4.5 $0.0035, Claude Opus 5.5 $0.014 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Claude Haiku 4.5 has the number, $0.56.
Directional only
At a glance
- Coding index
- Claude Haiku 4.5 35.6 from 4 benchmarks, Claude Opus 5.5 88.8 from 5 benchmarks.
- Widest category
- Agentic work, 46.5 points apart on average over 8 shared benchmarks.
- Widest single gap
- ARC-AGI-2: Claude Haiku 4.5 4.0% against Claude Opus 5.5 93.3%, 89.3 points apart on a 0 to 100 scale.
- Output price
- Claude Haiku 4.5 $5.00 against Claude Opus 5.5 $20.00 per million output tokens, 4.0x apart.
- Context window
- Claude Haiku 4.5 takes 200K tokens, Claude Opus 5.5 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.
| Category | Claude Haiku 4.5 | Bars | Claude Opus 5.5 | Ties |
|---|---|---|---|---|
| Coding | 0 | 10 | ||
| Intelligence | 0 | 2 | ||
| Agentic work | 0 | 8 | ||
| Reasoning | 0 | 8 | ||
| Math | 0 | 3 | ||
| Knowledge | 0 | 6 | ||
| Web design | 0 | 8 | ||
| General | 0 | 20 | ||
| All shared boards | 0 | 65 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index010
| Benchmark | Claude Haiku 4.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| SWE-bench Verified | 66.6% | Not measured | Only Claude Haiku 4.5 measured |
| SWE-bench Multilingual | 64.7% | Not measured | Only Claude Haiku 4.5 measured |
| LiveBench coding | 48.1% | 96.3% | Claude Opus 5.5 |
| LMArena WebDev | 1330 | 1807 | Claude Opus 5.5 |
| AA Coding Index | 43.9 | Not measured | Only Claude Haiku 4.5 measured |
| LiveCodeBench (Vals AI) | 41.2% | Not measured | Only Claude Haiku 4.5 measured |
| SWE-bench Verified (Vals AI) | 66.6% | Not measured | Only Claude Haiku 4.5 measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 65.2% | Only Claude Opus 5.5 measured |
| Vibe Code Bench (Vals AI) | 11.4% | 90.3% | Claude Opus 5.5 |
| Code migration (Vals AI) | 7.5% | 66.7% | Claude Opus 5.5 |
| ProgramBench (Vals AI) | 0.0% | 18.5% | Claude Opus 5.5 |
| IOI (Vals AI) | Not measured | 95.1% | Only Claude Opus 5.5 measured |
| Terminal-Bench 2.1 (Vals AI) | 43.8% | 87.6% | Claude Opus 5.5 |
| SWE-bench Multimodal (Vals AI) | Not measured | 70.2% | Only Claude Opus 5.5 measured |
| Vibe Code Bench 1-100 (Vals AI) | Not measured | 30.4% | Only Claude Opus 5.5 measured |
| CyberBench (Vals AI) | Not measured | 74.6% | Only Claude Opus 5.5 measured |
| SciCode | Not measured | 66.9% | Only Claude Opus 5.5 measured |
| Terminal-Bench | 35.5% | Not measured | Only Claude Haiku 4.5 measured |
| FrontierCode | Not measured | 54.6% | Only Claude Opus 5.5 measured |
| FrontierSWE | Not measured | 62.3% | Only Claude Opus 5.5 measured |
| WeirdML | 45.4% | Not measured | Only Claude Haiku 4.5 measured |
| MirrorCode | Not measured | 77.4% | Only Claude Opus 5.5 measured |
| ALE-Bench | Not measured | 2147.4 | Only Claude Opus 5.5 measured |
| LMArena Text: coding | 1454 | 1545 | Claude Opus 5.5 |
| LMArena Text: software and IT | 1439 | 1538 | Claude Opus 5.5 |
| LMArena WebDev: React | 1326 | 1855 | Claude Opus 5.5 |
| LMArena WebDev: HTML | 1318 | 1753 | Claude Opus 5.5 |
| LMArena WebDev: from an image | Not measured | 1753 | Only Claude Opus 5.5 measured |
| SciCode | Not measured | 18.5% | Only Claude Opus 5.5 measured |
| Neither model has a result on the other 10 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Claude Haiku 4.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| AA Intelligence Index | 16.9 | 57.6 | Claude Opus 5.5 |
| Epoch Capabilities Index | 142.2 | 168.8 | Claude Opus 5.5 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index08
| Benchmark | Claude Haiku 4.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| AA Agentic Index | 8.0 | Not measured | Only Claude Haiku 4.5 measured |
| Design Arena: agents building web apps | Not measured | 1316 | Only Claude Opus 5.5 measured |
| Design Arena: agents building mobile apps | Not measured | 1308 | Only Claude Opus 5.5 measured |
| Design Arena: spreadsheets | Not measured | 1368 | Only Claude Opus 5.5 measured |
| LiveBench research | 42.2% | 85.0% | Claude Opus 5.5 |
| LiveBench browser | 40.0% | 80.0% | Claude Opus 5.5 |
| LiveBench long agentic | 11.1% | 85.0% | Claude Opus 5.5 |
| SREBench (Vals AI) | Not measured | 33.6% | Only Claude Opus 5.5 measured |
| Finance Agent (Vals AI) | 31.0% | 58.6% | Claude Opus 5.5 |
| Tax Agent Bench (Vals AI) | 5.9% | 45.6% | Claude Opus 5.5 |
| Legal Research Bench (Vals AI) | 10.6% | 50.5% | Claude Opus 5.5 |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 9.2% | Only Claude Opus 5.5 measured |
| Excel Modeling Benchmark (Vals AI) | 22.7% | 75.9% | Claude Opus 5.5 |
| BioMysteryBench (Vals AI) | Not measured | 79.3% | Only Claude Opus 5.5 measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 47.1% | Only Claude Opus 5.5 measured |
| Vals Science Index | Not measured | 58.7% | Only Claude Opus 5.5 measured |
| Time Horizon Index (Vals AI) | Not measured | 91.3% | Only Claude Opus 5.5 measured |
| CUA-bench (Vals AI) | Not measured | 14.0% | Only Claude Opus 5.5 measured |
| Robo-bench (Vals AI) | Not measured | 14.6% | Only Claude Opus 5.5 measured |
| APEX-Agents | Not measured | 73.5% | Only Claude Opus 5.5 measured |
| Remote Labor Index | Not measured | 21.3% | Only Claude Opus 5.5 measured |
| BALROG | 31.2% | Not measured | Only Claude Haiku 4.5 measured |
| DeepResearch Bench | 45.5% | Not measured | Only Claude Haiku 4.5 measured |
| SpireClimb | Not measured | 3.3% | Only Claude Opus 5.5 measured |
| Vending-Bench 2 | $459 | $9,235 | Claude Opus 5.5 |
| LMArena Search | Not measured | 1213 | Only Claude Opus 5.5 measured |
| Berkeley Function Calling Leaderboard (BFCL) | 68.7% | Not measured | Only Claude Haiku 4.5 measured |
| Neither model has a result on the other 14 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index08
| Benchmark | Claude Haiku 4.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| GPQA Diamond | 72.2% | Not measured | Only Claude Haiku 4.5 measured |
| LiveBench reasoning | 61.7% | 92.2% | Claude Opus 5.5 |
| MysteryMechanism (Vals AI) | Not measured | 49.5% | Only Claude Opus 5.5 measured |
| MMMU Pro (Vals AI) | 46.1% | Not measured | Only Claude Haiku 4.5 measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 61.4% | Only Claude Opus 5.5 measured |
| GPQA Diamond (Epoch AI run) | 71.2% | 90.6% | Claude Opus 5.5 |
| ARC-AGI | 47.7% | 98.5% | Claude Opus 5.5 |
| ARC-AGI-2 | 4.0% | 93.3% | Claude Opus 5.5 |
| Chess puzzles | 8.0% | 67.0% | Claude Opus 5.5 |
| Mystery game puzzles | Not measured | 71.0% | Only Claude Opus 5.5 measured |
| EBR-bench | Not measured | 71.4% | Only Claude Opus 5.5 measured |
| LMArena Text: hard prompts | 1421 | 1536 | Claude Opus 5.5 |
| LMArena Text: expert | 1443 | 1549 | Claude Opus 5.5 |
| LMArena Text: hard prompts in English | 1435 | 1535 | Claude Opus 5.5 |
| LMArena Vision: diagrams | Not measured | 1345 | Only Claude Opus 5.5 measured |
| LMArena Vision: homework | Not measured | 1362 | Only Claude Opus 5.5 measured |
| Neither model has a result on the other 6 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | Claude Haiku 4.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| AIME | 82.7% | Not measured | Only Claude Haiku 4.5 measured |
| MGSM (Vals AI) | 92.2% | Not measured | Only Claude Haiku 4.5 measured |
| OTIS Mock AIME | 66.7% | 100.0% | Claude Opus 5.5 |
| FrontierMath, tiers 1 to 3 | Not measured | 91.2% | Only Claude Opus 5.5 measured |
| FrontierMath, tier 4 | Not measured | 95.0% | Only Claude Opus 5.5 measured |
| ProofBench | Not measured | 100.0% | Only Claude Opus 5.5 measured |
| MATH level 5 | 96.4% | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text: math | 1398 | 1518 | Claude Opus 5.5 |
| LMArena Text: mathematical work | 1430 | 1516 | Claude Opus 5.5 |
| Neither model has a result on the other 4 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | Claude Haiku 4.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| MMLU Pro | 78.7% | Not measured | Only Claude Haiku 4.5 measured |
| LegalBench (Vals AI) | 81.2% | Not measured | Only Claude Haiku 4.5 measured |
| CaseLaw (Vals AI) | 56.5% | Not measured | Only Claude Haiku 4.5 measured |
| CorpFin (Vals AI) | 60.6% | Not measured | Only Claude Haiku 4.5 measured |
| TaxEval (Vals AI) | 67.5% | Not measured | Only Claude Haiku 4.5 measured |
| MortgageTax (Vals AI) | 62.2% | Not measured | Only Claude Haiku 4.5 measured |
| MedQA (Vals AI) | 79.6% | Not measured | Only Claude Haiku 4.5 measured |
| MedCode (Vals AI) | 32.7% | 49.8% | Claude Opus 5.5 |
| SimpleQA Verified | 13.2% | 72.2% | Claude Opus 5.5 |
| LMArena Text: science | 1406 | 1524 | Claude Opus 5.5 |
| LMArena Text: medicine and healthcare | 1394 | 1509 | Claude Opus 5.5 |
| LMArena Text: legal and government | 1400 | 1518 | Claude Opus 5.5 |
| LMArena Text: business and finance | 1400 | 1494 | Claude Opus 5.5 |
| FACTS Benchmark Suite | Not measured | 65.3% | Only Claude Opus 5.5 measured |
| Neither model has a result on the other 4 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index08
| Benchmark | Claude Haiku 4.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| Design Arena: websites | 1122 | 1335 | Claude Opus 5.5 |
| Design Arena: UI components | 1097 | 1385 | Claude Opus 5.5 |
| Design Arena: game dev | 1097 | 1461 | Claude Opus 5.5 |
| Design Arena: data viz | 1124 | 1397 | Claude Opus 5.5 |
| Design Arena: 3D | 1071 | 1476 | Claude Opus 5.5 |
| Design Arena: all code categories | 1114 | 1390 | Claude Opus 5.5 |
| Design Arena: SVG | 1025 | 1329 | Claude Opus 5.5 |
| Design Arena: ASCII art | 1140 | Not measured | Only Claude Haiku 4.5 measured |
| LiveBench design | 59.9% | 86.9% | Claude Opus 5.5 |
GeneralLike-for-likePublic boards, shown as published, kept out of the index020
| Benchmark | Claude Haiku 4.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| Vals Index | Not measured | 70.2% | Only Claude Opus 5.5 measured |
| LiveBench data analysis | 59.3% | 80.3% | Claude Opus 5.5 |
| Public Benefits Bench (Vals AI) | 54.3% | 70.6% | Claude Opus 5.5 |
| SAGE (Vals AI) | 31.8% | 45.8% | Claude Opus 5.5 |
| MedScribe (Vals AI) | 85.2% | 91.4% | Claude Opus 5.5 |
| Vals Multimodal Index | 42.9% | Not measured | Only Claude Haiku 4.5 measured |
| Long context reasoning (AA-LCR) | Not measured | 84.7% | Only Claude Opus 5.5 measured |
| DTBench | 73.6% | 98.9% | Claude Opus 5.5 |
| LMCA | 30.9% | 68.2% | Claude Opus 5.5 |
| Furniture assembly | Not measured | 83.3% | Only Claude Opus 5.5 measured |
| ExploitBench | 13.7% | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text | 1397 | 1516 | Claude Opus 5.5 |
| LMArena Text: instruction following | 1414 | 1537 | Claude Opus 5.5 |
| LMArena Text: creative writing | 1371 | 1535 | Claude Opus 5.5 |
| LMArena Vision | Not measured | 1321 | Only Claude Opus 5.5 measured |
| LMArena Text: English | 1414 | 1519 | Claude Opus 5.5 |
| LMArena Text: non-English | 1377 | 1507 | Claude Opus 5.5 |
| LMArena Text: Chinese | 1418 | 1582 | Claude Opus 5.5 |
| LMArena Text: German | 1374 | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text: Japanese | 1338 | 1534 | Claude Opus 5.5 |
| LMArena Text: Korean | 1346 | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text: French | 1411 | 1493 | Claude Opus 5.5 |
| LMArena Text: Spanish | 1421 | 1508 | Claude Opus 5.5 |
| LMArena Text: Russian | 1381 | 1517 | Claude Opus 5.5 |
| LMArena Text: Polish | 1343 | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text: multi-turn | 1409 | 1501 | Claude Opus 5.5 |
| LMArena Text: longer queries | 1427 | 1532 | Claude Opus 5.5 |
| LMArena Text: writing and language | 1386 | 1532 | Claude Opus 5.5 |
| LMArena Text: entertainment and media | 1369 | 1515 | Claude Opus 5.5 |
| LMArena Vision: English | Not measured | 1317 | Only Claude Opus 5.5 measured |
| LMArena Vision: reading text | Not measured | 1326 | Only Claude Opus 5.5 measured |
| LMArena Vision: creative writing | Not measured | 1320 | Only Claude Opus 5.5 measured |
| LMArena Document | 1420 | Not measured | Only Claude Haiku 4.5 measured |
| Neither model has a result on the other 13 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Haiku 4.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| 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 costs less
Repository review
50K input + 3K output tokens
Claude Haiku 4.5 costs less
Agent loop
200K input + 10K output tokens
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.
| Specification | Anthropic Claude Haiku 4.5 | Anthropic Claude Opus 5.5 |
|---|---|---|
| Maker | Anthropic | Anthropic |
| Weights | No public weights listed | No public weights listed |
| Released | October 15, 2025 | September 22, 2026 |
| Context window | 200K | 1M |
| Input $/Mtok | $1.00 | $4.00 |
| Output $/Mtok | $5.00 | $20.00 |
| Coding index | 35.6 | 88.8 |
| Independent measurements behind it | 4 | 5 |
| Cost per solved task | $0.56 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Haiku 4.5 better than Claude Opus 5.5 for coding?
Pick Claude Opus 5.5 for coding: it scores 88.8 on the Vibe Coding Index against 35.6 for Claude Haiku 4.5. Pick Claude Haiku 4.5 for output tokens 4.0x cheaper ($5.00 vs $20.00 per million).
Which is cheaper, Claude Haiku 4.5 or Claude Opus 5.5?
Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens; Claude Opus 5.5 costs $4.00 and $20.00. Cost per solved task is known for Claude Haiku 4.5 only.
Which has the bigger context window, Claude Haiku 4.5 or Claude Opus 5.5?
Claude Opus 5.5, with 1M tokens against 200K.
How many benchmarks is the Claude Haiku 4.5 vs Claude Opus 5.5 comparison based on?
65 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 5.5's on 5. Where only one model has a result, the page says so instead of naming a winner.