Claude Haiku 5.5 vs Claude Opus 5.5
Claude Opus 5.5 leads Claude Haiku 5.5 on coding, 88.8 to 70.7 on the Vibe Coding Index, and Claude Haiku 5.5 costs 40.0x less per output token ($0.50 vs $20.00 per million). Updated October 22, 2026; 63 benchmarks have a result for both.
Claude Haiku 5.5 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 70.7 for Claude Haiku 5.5.
Pick Claude Haiku 5.5 for output tokens 40.0x cheaper ($0.50 vs $20.00 per million) and work measured by Vibe Code Bench (Vals AI) or CyberBench (Vals AI), where it leads.
- Shared results
- 63
- Claude Haiku 5.5 only
- 0
- Claude Opus 5.5 only
- 46
- 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 70.7.
5 benchmarks behind each score
Intelligence
AA Intelligence Index
Claude Opus 5.5
Claude Opus 5.5 leads, 57.6 against 43.4.
Same board, both measured
Web design
Design Arena: websites
Claude Opus 5.5
Claude Opus 5.5 leads, 1335 against 1262.
Same board, both measured
General
Vals Index
Claude Opus 5.5
Claude Opus 5.5 leads, 70.2% against 58.7%.
Same board, both measured
Long documents
Listed context window
No call
Both take 1M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Claude Haiku 5.5
Claude Haiku 5.5 $0.0003, Claude Opus 5.5 $0.014 per turn.
List prices
At a glance
- Coding index
- Claude Haiku 5.5 70.7 from 5 benchmarks, Claude Opus 5.5 88.8 from 5 benchmarks.
- Widest category
- Knowledge, 25.1 points apart on average over 2 shared benchmarks.
- Widest single gap
- FrontierMath, tier 4: Claude Haiku 5.5 46.3% against Claude Opus 5.5 95.0%, 48.7 points apart on a 0 to 100 scale.
- Output price
- Claude Haiku 5.5 $0.50 against Claude Opus 5.5 $20.00 per million output tokens, 40.0x apart.
- Context window
- Both take 1M 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.
| Category | Claude Haiku 5.5 | Bars | Claude Opus 5.5 | Ties |
|---|---|---|---|---|
| Coding | 2 | 14 | ||
| Intelligence | 0 | 2 | ||
| Agentic work | 0 | 13 | 1 tie | |
| Reasoning | 0 | 9 | ||
| Math | 0 | 4 | ||
| Knowledge | 1 | 1 | ||
| Web design | 0 | 7 | ||
| General | 0 | 9 | ||
| All shared boards | 3 | 59 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index214
| Benchmark | Claude Haiku 5.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| LiveBench coding | 91.4% | 96.3% | Claude Opus 5.5 |
| LMArena WebDev | 1586 | 1807 | Claude Opus 5.5 |
| Terminal-Bench 4 (Vals AI) | 36.4% | 65.2% | Claude Opus 5.5 |
| Vibe Code Bench (Vals AI) | 90.4% | 90.3% | Claude Haiku 5.5 |
| Code migration (Vals AI) | 44.6% | 66.7% | Claude Opus 5.5 |
| ProgramBench (Vals AI) | 2.5% | 18.5% | Claude Opus 5.5 |
| IOI (Vals AI) | 47.3% | 95.1% | Claude Opus 5.5 |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 87.6% | Only Claude Opus 5.5 measured |
| SWE-bench Multimodal (Vals AI) | 44.0% | 70.2% | Claude Opus 5.5 |
| Vibe Code Bench 1-100 (Vals AI) | 20.4% | 30.4% | Claude Opus 5.5 |
| CyberBench (Vals AI) | 75.5% | 74.6% | Claude Haiku 5.5 |
| SciCode | 55.0% | 66.9% | Claude Opus 5.5 |
| FrontierCode | 46.4% | 54.6% | Claude Opus 5.5 |
| FrontierSWE | 43.8% | 62.3% | Claude Opus 5.5 |
| 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 | Not measured | 1545 | Only Claude Opus 5.5 measured |
| LMArena Text: software and IT | Not measured | 1538 | Only Claude Opus 5.5 measured |
| LMArena WebDev: React | 1605 | 1855 | Claude Opus 5.5 |
| LMArena WebDev: HTML | 1569 | 1753 | Claude Opus 5.5 |
| LMArena WebDev: from an image | Not measured | 1753 | Only Claude Opus 5.5 measured |
| SciCode | 9.2% | 18.5% | Claude Opus 5.5 |
| Neither model has a result on the other 17 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Claude Haiku 5.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| AA Intelligence Index | 43.4 | 57.6 | Claude Opus 5.5 |
| Epoch Capabilities Index | 159.3 | 168.8 | Claude Opus 5.5 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index013
| Benchmark | Claude Haiku 5.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| 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 | 81.9% | 85.0% | Claude Opus 5.5 |
| LiveBench browser | 80.0% | 80.0% | Tie |
| LiveBench long agentic | 57.6% | 85.0% | Claude Opus 5.5 |
| SREBench (Vals AI) | 5.7% | 33.6% | Claude Opus 5.5 |
| Finance Agent (Vals AI) | 54.1% | 58.6% | Claude Opus 5.5 |
| Tax Agent Bench (Vals AI) | 31.9% | 45.6% | Claude Opus 5.5 |
| Legal Research Bench (Vals AI) | 43.3% | 50.5% | Claude Opus 5.5 |
| Harvey's Legal Agent Benchmark (Vals AI) | 4.2% | 9.2% | Claude Opus 5.5 |
| Excel Modeling Benchmark (Vals AI) | 59.0% | 75.9% | Claude Opus 5.5 |
| BioMysteryBench (Vals AI) | 70.4% | 79.3% | Claude Opus 5.5 |
| Terminal-Bench-Science (Vals AI) | 10.0% | 47.1% | Claude Opus 5.5 |
| Vals Science Index | 38.5% | 58.7% | Claude Opus 5.5 |
| 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 | 56.1% | 73.5% | Claude Opus 5.5 |
| Remote Labor Index | Not measured | 21.3% | Only Claude Opus 5.5 measured |
| SpireClimb | Not measured | 3.3% | Only Claude Opus 5.5 measured |
| Vending-Bench 2 | $3,602 | $9,235 | Claude Opus 5.5 |
| LMArena Search | Not measured | 1213 | Only Claude Opus 5.5 measured |
| Neither model has a result on the other 18 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index09
| Benchmark | Claude Haiku 5.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| LiveBench reasoning | 82.4% | 92.2% | Claude Opus 5.5 |
| MysteryMechanism (Vals AI) | 35.1% | 49.5% | Claude Opus 5.5 |
| Humanity's Last Exam (Artificial Analysis run) | 44.4% | 61.4% | Claude Opus 5.5 |
| GPQA Diamond (Epoch AI run) | 89.6% | 90.6% | Claude Opus 5.5 |
| ARC-AGI | 89.0% | 98.5% | Claude Opus 5.5 |
| ARC-AGI-2 | 61.3% | 93.3% | Claude Opus 5.5 |
| Chess puzzles | 55.0% | 67.0% | Claude Opus 5.5 |
| Mystery game puzzles | 30.0% | 71.0% | Claude Opus 5.5 |
| EBR-bench | 43.8% | 71.4% | Claude Opus 5.5 |
| LMArena Text: hard prompts | Not measured | 1536 | Only Claude Opus 5.5 measured |
| LMArena Text: expert | Not measured | 1549 | Only Claude Opus 5.5 measured |
| LMArena Text: hard prompts in English | Not measured | 1535 | Only Claude Opus 5.5 measured |
| 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 8 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index04
| Benchmark | Claude Haiku 5.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 98.9% | 100.0% | Claude Opus 5.5 |
| FrontierMath, tiers 1 to 3 | 75.1% | 91.2% | Claude Opus 5.5 |
| FrontierMath, tier 4 | 46.3% | 95.0% | Claude Opus 5.5 |
| ProofBench | 89.0% | 100.0% | Claude Opus 5.5 |
| LMArena Text: math | Not measured | 1518 | Only Claude Opus 5.5 measured |
| LMArena Text: mathematical work | Not measured | 1516 | Only Claude Opus 5.5 measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | Claude Haiku 5.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| MedCode (Vals AI) | 51.5% | 49.8% | Claude Haiku 5.5 |
| SimpleQA Verified | 23.8% | 72.2% | Claude Opus 5.5 |
| LMArena Text: science | Not measured | 1524 | Only Claude Opus 5.5 measured |
| LMArena Text: medicine and healthcare | Not measured | 1509 | Only Claude Opus 5.5 measured |
| LMArena Text: legal and government | Not measured | 1518 | Only Claude Opus 5.5 measured |
| LMArena Text: business and finance | Not measured | 1494 | Only Claude Opus 5.5 measured |
| FACTS Benchmark Suite | Not measured | 65.3% | Only Claude Opus 5.5 measured |
| Neither model has a result on the other 11 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index07
| Benchmark | Claude Haiku 5.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| Design Arena: websites | 1262 | 1335 | Claude Opus 5.5 |
| Design Arena: UI components | 1291 | 1385 | Claude Opus 5.5 |
| Design Arena: game dev | 1313 | 1461 | Claude Opus 5.5 |
| Design Arena: data viz | 1294 | 1397 | Claude Opus 5.5 |
| Design Arena: 3D | 1342 | 1476 | Claude Opus 5.5 |
| Design Arena: all code categories | 1278 | 1390 | Claude Opus 5.5 |
| Design Arena: SVG | Not measured | 1329 | Only Claude Opus 5.5 measured |
| LiveBench design | 83.1% | 86.9% | Claude Opus 5.5 |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index09
| Benchmark | Claude Haiku 5.5 | Claude Opus 5.5 | Leader |
|---|---|---|---|
| Vals Index | 58.7% | 70.2% | Claude Opus 5.5 |
| LiveBench data analysis | 53.1% | 80.3% | Claude Opus 5.5 |
| Public Benefits Bench (Vals AI) | 63.2% | 70.6% | Claude Opus 5.5 |
| SAGE (Vals AI) | 45.8% | 45.8% | Claude Opus 5.5 |
| MedScribe (Vals AI) | 89.5% | 91.4% | Claude Opus 5.5 |
| Long context reasoning (AA-LCR) | 82.7% | 84.7% | Claude Opus 5.5 |
| DTBench | 95.0% | 98.9% | Claude Opus 5.5 |
| LMCA | 51.1% | 68.2% | Claude Opus 5.5 |
| Furniture assembly | 47.5% | 83.3% | Claude Opus 5.5 |
| LMArena Text | Not measured | 1516 | Only Claude Opus 5.5 measured |
| LMArena Text: instruction following | Not measured | 1537 | Only Claude Opus 5.5 measured |
| LMArena Text: creative writing | Not measured | 1535 | Only Claude Opus 5.5 measured |
| LMArena Vision | Not measured | 1321 | Only Claude Opus 5.5 measured |
| LMArena Text: English | Not measured | 1519 | Only Claude Opus 5.5 measured |
| LMArena Text: non-English | Not measured | 1507 | Only Claude Opus 5.5 measured |
| LMArena Text: Chinese | Not measured | 1582 | Only Claude Opus 5.5 measured |
| LMArena Text: Japanese | Not measured | 1534 | Only Claude Opus 5.5 measured |
| LMArena Text: French | Not measured | 1493 | Only Claude Opus 5.5 measured |
| LMArena Text: Spanish | Not measured | 1508 | Only Claude Opus 5.5 measured |
| LMArena Text: Russian | Not measured | 1517 | Only Claude Opus 5.5 measured |
| LMArena Text: multi-turn | Not measured | 1501 | Only Claude Opus 5.5 measured |
| LMArena Text: longer queries | Not measured | 1532 | Only Claude Opus 5.5 measured |
| LMArena Text: writing and language | Not measured | 1532 | Only Claude Opus 5.5 measured |
| LMArena Text: entertainment and media | Not measured | 1515 | Only Claude Opus 5.5 measured |
| 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 |
| Neither model has a result on the other 19 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Haiku 5.5 | 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 5.5 costs less
Repository review
50K input + 3K output tokens
Claude Haiku 5.5 costs less
Agent loop
200K input + 10K output tokens
Claude Haiku 5.5 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 5.5 | Anthropic Claude Opus 5.5 |
|---|---|---|
| Maker | Anthropic | Anthropic |
| Weights | No public weights listed | No public weights listed |
| Released | October 7, 2026 | September 22, 2026 |
| Context window | 1M | 1M |
| Input $/Mtok | $0.10 | $4.00 |
| Output $/Mtok | $0.50 | $20.00 |
| Coding index | 70.7 | 88.8 |
| Independent measurements behind it | 5 | 5 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Haiku 5.5 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 70.7 for Claude Haiku 5.5. Pick Claude Haiku 5.5 for output tokens 40.0x cheaper ($0.50 vs $20.00 per million) and work measured by Vibe Code Bench (Vals AI) or CyberBench (Vals AI), where it leads.
Which is cheaper, Claude Haiku 5.5 or Claude Opus 5.5?
Claude Haiku 5.5 costs $0.10 per million input tokens and $0.50 per million output tokens; Claude Opus 5.5 costs $4.00 and $20.00. Cost per solved task is known for neither of them.
Which has the bigger context window, Claude Haiku 5.5 or Claude Opus 5.5?
Neither: both take 1M tokens.
How many benchmarks is the Claude Haiku 5.5 vs Claude Opus 5.5 comparison based on?
63 benchmarks have a result for both models as of October 22, 2026. Claude Haiku 5.5's coding index rests on 5 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.