Claude Haiku 5.5 vs Claude Sonnet 4.6
Claude Haiku 5.5 leads Claude Sonnet 4.6 on coding, 70.7 to 57.3 on the Vibe Coding Index, and Claude Haiku 5.5 costs 30.0x less per output token ($0.50 vs $15.00 per million). Updated October 22, 2026; 43 benchmarks have a result for both.
vs boards won of 43 shared
Claude Haiku 5.5 or Claude Sonnet 4.6: which to pick
Pick Claude Haiku 5.5 for coding: it scores 70.7 on the Vibe Coding Index against 57.3 for Claude Sonnet 4.6.
Pick Claude Sonnet 4.6 for work measured by Excel Modeling Benchmark (Vals AI) or Vending-Bench 2, where it leads.
Model AClaude Haiku 5.5Anthropic70.7/1005 independent benchmarksNo cost per solved taskModel BClaude Sonnet 4.6Anthropic57.3/1005 independent benchmarks$1.68 per solved task
- Shared results
- 43
- Claude Haiku 5.5 only
- 20
- Claude Sonnet 4.6 only
- 75
- 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 Haiku 5.5
Claude Haiku 5.5 scores 70.7 against 57.3.
5 benchmarks behind each score
Intelligence
AA Intelligence Index
Claude Haiku 5.5
Claude Haiku 5.5 leads, 43.4 against 30.1.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Claude Sonnet 4.6 has a AA Agentic Index result, 31.8.
Directional only
Reasoning
GPQA Diamond
No call
Only Claude Sonnet 4.6 has a GPQA Diamond result, 85.6%.
Directional only
Math
AIME
No call
Only Claude Sonnet 4.6 has a AIME result, 92.3%.
Directional only
Knowledge
MMLU Pro
No call
Only Claude Sonnet 4.6 has a MMLU Pro result, 87.3%.
Directional only
Web design
Design Arena: websites
Claude Sonnet 4.6
Claude Sonnet 4.6 leads, 1284 against 1262.
Same board, both measured
General
Vals Index
No call
Only Claude Haiku 5.5 has a Vals Index result, 58.7%.
Directional only
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 Sonnet 4.6 $0.011 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Claude Sonnet 4.6 has the number, $1.68.
Directional only
At a glance
- Coding index
- Claude Haiku 5.5 70.7 from 5 benchmarks, Claude Sonnet 4.6 57.3 from 5 benchmarks.
- Widest category
- Math, 28.5 points apart on average over 2 shared benchmarks.
- Widest single gap
- ProofBench: Claude Haiku 5.5 89.0% against Claude Sonnet 4.6 45.0%, 44.0 points apart on a 0 to 100 scale.
- Output price
- Claude Haiku 5.5 $0.50 against Claude Sonnet 4.6 $15.00 per million output tokens, 30.0x 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 Sonnet 4.6 | Ties |
|---|---|---|---|---|
| Coding | 9 | 0 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 7 | 2 | ||
| Reasoning | 6 | 1 | ||
| Math | 2 | 0 | ||
| Knowledge | 0 | 1 | ||
| Web design | 6 | 1 | ||
| General | 4 | 2 | ||
| All shared boards | 36 | 7 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index90
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| LiveBench coding | 91.4% | 66.6% | Claude Haiku 5.5 |
| LMArena WebDev | 1586 | 1523 | Claude Haiku 5.5 |
| AA Coding Index | Not measured | 63.0 | Only Claude Sonnet 4.6 measured |
| LiveCodeBench (Vals AI) | Not measured | 82.1% | Only Claude Sonnet 4.6 measured |
| SWE-bench Verified (Vals AI) | Not measured | 77.4% | Only Claude Sonnet 4.6 measured |
| Terminal-Bench 4 (Vals AI) | 36.4% | Not measured | Only Claude Haiku 5.5 measured |
| Vibe Code Bench (Vals AI) | 90.4% | 55.8% | Claude Haiku 5.5 |
| Code migration (Vals AI) | 44.6% | 39.9% | Claude Haiku 5.5 |
| ProgramBench (Vals AI) | 2.5% | 0.5% | Claude Haiku 5.5 |
| IOI (Vals AI) | 47.3% | Not measured | Only Claude Haiku 5.5 measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 57.3% | Only Claude Sonnet 4.6 measured |
| SWE-bench Multimodal (Vals AI) | 44.0% | Not measured | Only Claude Haiku 5.5 measured |
| Vibe Code Bench 1-100 (Vals AI) | 20.4% | Not measured | Only Claude Haiku 5.5 measured |
| SkillsBench (Vals AI) | Not measured | 49.0% | Only Claude Sonnet 4.6 measured |
| CyberBench (Vals AI) | 75.5% | Not measured | Only Claude Haiku 5.5 measured |
| SciCode | 55.0% | 50.1% | Claude Haiku 5.5 |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 53.0% | Only Claude Sonnet 4.6 measured |
| Terminal-Bench | Not measured | 53.4% | Only Claude Sonnet 4.6 measured |
| FrontierCode | 46.4% | 24.3% | Claude Haiku 5.5 |
| FrontierSWE | 43.8% | Not measured | Only Claude Haiku 5.5 measured |
| WeirdML | Not measured | 66.1% | Only Claude Sonnet 4.6 measured |
| LMArena Text: coding | Not measured | 1503 | Only Claude Sonnet 4.6 measured |
| LMArena Text: software and IT | Not measured | 1491 | Only Claude Sonnet 4.6 measured |
| LMArena WebDev: React | 1605 | 1512 | Claude Haiku 5.5 |
| LMArena WebDev: HTML | 1569 | 1527 | Claude Haiku 5.5 |
| LMArena WebDev: from an image | Not measured | 1529 | Only Claude Sonnet 4.6 measured |
| SciCode | 9.2% | Not measured | Only Claude Haiku 5.5 measured |
| Neither model has a result on the other 12 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| AA Intelligence Index | 43.4 | 30.1 | Claude Haiku 5.5 |
| Epoch Capabilities Index | 159.3 | 152.8 | Claude Haiku 5.5 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index72
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| AA Agentic Index | Not measured | 31.8 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building web apps | Not measured | 1190 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building full-stack apps | Not measured | 1213 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building mobile apps | Not measured | 1212 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building Android apps | Not measured | 1197 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building Godot games | Not measured | 1226 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building games | Not measured | 1177 | Only Claude Sonnet 4.6 measured |
| LiveBench research | 81.9% | 63.6% | Claude Haiku 5.5 |
| LiveBench browser | 80.0% | 68.0% | Claude Haiku 5.5 |
| LiveBench long agentic | 57.6% | 23.1% | Claude Haiku 5.5 |
| SREBench (Vals AI) | 5.7% | Not measured | Only Claude Haiku 5.5 measured |
| Finance Agent (Vals AI) | 54.1% | 51.0% | Claude Haiku 5.5 |
| Tax Agent Bench (Vals AI) | 31.9% | 25.8% | Claude Haiku 5.5 |
| Legal Research Bench (Vals AI) | 43.3% | 38.5% | Claude Haiku 5.5 |
| Harvey's Legal Agent Benchmark (Vals AI) | 4.2% | Not measured | Only Claude Haiku 5.5 measured |
| Excel Modeling Benchmark (Vals AI) | 59.0% | 60.2% | Claude Sonnet 4.6 |
| BioMysteryBench (Vals AI) | 70.4% | Not measured | Only Claude Haiku 5.5 measured |
| Terminal-Bench-Science (Vals AI) | 10.0% | Not measured | Only Claude Haiku 5.5 measured |
| Vals Science Index | 38.5% | Not measured | Only Claude Haiku 5.5 measured |
| τ²-Bench Telecom | Not measured | 79.5% | Only Claude Sonnet 4.6 measured |
| APEX-Agents | 56.1% | 43.0% | Claude Haiku 5.5 |
| OSWorld 2.0 | Not measured | 9.3% | Only Claude Sonnet 4.6 measured |
| DeepResearch Bench | Not measured | 54.9% | Only Claude Sonnet 4.6 measured |
| OSWorld | Not measured | 72.1% | Only Claude Sonnet 4.6 measured |
| Vending-Bench 2 | $3,602 | $7,204 | Claude Sonnet 4.6 |
| Neither model has a result on the other 16 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index61
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| GPQA Diamond | Not measured | 85.6% | Only Claude Sonnet 4.6 measured |
| LiveBench reasoning | 82.4% | 86.4% | Claude Sonnet 4.6 |
| MysteryMechanism (Vals AI) | 35.1% | Not measured | Only Claude Haiku 5.5 measured |
| MMMU Pro (Vals AI) | Not measured | 83.6% | Only Claude Sonnet 4.6 measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 87.5% | Only Claude Sonnet 4.6 measured |
| Humanity's Last Exam (Artificial Analysis run) | 44.4% | 33.6% | Claude Haiku 5.5 |
| GPQA Diamond (Epoch AI run) | 89.6% | 87.4% | Claude Haiku 5.5 |
| ARC-AGI | 89.0% | 86.5% | Claude Haiku 5.5 |
| ARC-AGI-2 | 61.3% | 60.4% | Claude Haiku 5.5 |
| Chess puzzles | 55.0% | 23.0% | Claude Haiku 5.5 |
| Mystery game puzzles | 30.0% | 16.0% | Claude Haiku 5.5 |
| EBR-bench | 43.8% | Not measured | Only Claude Haiku 5.5 measured |
| LMArena Text: hard prompts | Not measured | 1485 | Only Claude Sonnet 4.6 measured |
| LMArena Text: expert | Not measured | 1500 | Only Claude Sonnet 4.6 measured |
| LMArena Text: hard prompts in English | Not measured | 1493 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: diagrams | Not measured | 1306 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: homework | Not measured | 1311 | Only Claude Sonnet 4.6 measured |
| Neither model has a result on the other 5 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| AIME | Not measured | 92.3% | Only Claude Sonnet 4.6 measured |
| OTIS Mock AIME | 98.9% | 85.8% | Claude Haiku 5.5 |
| FrontierMath, tiers 1 to 3 | 75.1% | Not measured | Only Claude Haiku 5.5 measured |
| FrontierMath, tier 4 | 46.3% | Not measured | Only Claude Haiku 5.5 measured |
| ProofBench | 89.0% | 45.0% | Claude Haiku 5.5 |
| LMArena Text: math | Not measured | 1462 | Only Claude Sonnet 4.6 measured |
| LMArena Text: mathematical work | Not measured | 1483 | Only Claude Sonnet 4.6 measured |
| Neither model has a result on the other 6 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 87.3% | Only Claude Sonnet 4.6 measured |
| LegalBench (Vals AI) | Not measured | 82.1% | Only Claude Sonnet 4.6 measured |
| CaseLaw (Vals AI) | Not measured | 64.0% | Only Claude Sonnet 4.6 measured |
| CorpFin (Vals AI) | Not measured | 65.3% | Only Claude Sonnet 4.6 measured |
| TaxEval (Vals AI) | Not measured | 77.1% | Only Claude Sonnet 4.6 measured |
| MortgageTax (Vals AI) | Not measured | 67.7% | Only Claude Sonnet 4.6 measured |
| MedQA (Vals AI) | Not measured | 92.1% | Only Claude Sonnet 4.6 measured |
| MedCode (Vals AI) | 51.5% | Not measured | Only Claude Haiku 5.5 measured |
| SimpleQA Verified | 23.8% | 35.5% | Claude Sonnet 4.6 |
| LMArena Text: science | Not measured | 1474 | Only Claude Sonnet 4.6 measured |
| LMArena Text: medicine and healthcare | Not measured | 1451 | Only Claude Sonnet 4.6 measured |
| LMArena Text: legal and government | Not measured | 1470 | Only Claude Sonnet 4.6 measured |
| LMArena Text: business and finance | Not measured | 1464 | Only Claude Sonnet 4.6 measured |
| FACTS Benchmark Suite | Not measured | 37.0% | Only Claude Sonnet 4.6 measured |
| Neither model has a result on the other 4 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index61
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| Design Arena: websites | 1262 | 1284 | Claude Sonnet 4.6 |
| Design Arena: UI components | 1291 | 1269 | Claude Haiku 5.5 |
| Design Arena: game dev | 1313 | 1258 | Claude Haiku 5.5 |
| Design Arena: data viz | 1294 | 1278 | Claude Haiku 5.5 |
| Design Arena: 3D | 1342 | 1234 | Claude Haiku 5.5 |
| Design Arena: all code categories | 1278 | 1275 | Claude Haiku 5.5 |
| Design Arena: SVG | Not measured | 1196 | Only Claude Sonnet 4.6 measured |
| Design Arena: ASCII art | Not measured | 1235 | Only Claude Sonnet 4.6 measured |
| LiveBench design | 83.1% | 71.8% | Claude Haiku 5.5 |
GeneralLike-for-likePublic boards, shown as published, kept out of the index42
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| Vals Index | 58.7% | Not measured | Only Claude Haiku 5.5 measured |
| LiveBench data analysis | 53.1% | 76.1% | Claude Sonnet 4.6 |
| Public Benefits Bench (Vals AI) | 63.2% | 62.5% | Claude Haiku 5.5 |
| SAGE (Vals AI) | 45.8% | 46.6% | Claude Sonnet 4.6 |
| MedScribe (Vals AI) | 89.5% | Not measured | Only Claude Haiku 5.5 measured |
| Vals Multimodal Index | Not measured | 60.6% | Only Claude Sonnet 4.6 measured |
| IFBench | Not measured | 56.6% | Only Claude Sonnet 4.6 measured |
| Long context reasoning (AA-LCR) | 82.7% | 80.0% | Claude Haiku 5.5 |
| DTBench | 95.0% | 89.9% | Claude Haiku 5.5 |
| LMCA | 51.1% | 46.5% | Claude Haiku 5.5 |
| Furniture assembly | 47.5% | Not measured | Only Claude Haiku 5.5 measured |
| ExploitBench | Not measured | 23.6% | Only Claude Sonnet 4.6 measured |
| LMArena Text | Not measured | 1458 | Only Claude Sonnet 4.6 measured |
| LMArena Text: instruction following | Not measured | 1476 | Only Claude Sonnet 4.6 measured |
| LMArena Text: creative writing | Not measured | 1435 | Only Claude Sonnet 4.6 measured |
| LMArena Vision | Not measured | 1283 | Only Claude Sonnet 4.6 measured |
| LMArena Text: English | Not measured | 1472 | Only Claude Sonnet 4.6 measured |
| LMArena Text: non-English | Not measured | 1440 | Only Claude Sonnet 4.6 measured |
| LMArena Text: Chinese | Not measured | 1491 | Only Claude Sonnet 4.6 measured |
| LMArena Text: German | Not measured | 1428 | Only Claude Sonnet 4.6 measured |
| LMArena Text: Japanese | Not measured | 1420 | Only Claude Sonnet 4.6 measured |
| LMArena Text: Korean | Not measured | 1410 | Only Claude Sonnet 4.6 measured |
| LMArena Text: French | Not measured | 1465 | Only Claude Sonnet 4.6 measured |
| LMArena Text: Spanish | Not measured | 1464 | Only Claude Sonnet 4.6 measured |
| LMArena Text: Russian | Not measured | 1440 | Only Claude Sonnet 4.6 measured |
| LMArena Text: Polish | Not measured | 1443 | Only Claude Sonnet 4.6 measured |
| LMArena Text: multi-turn | Not measured | 1464 | Only Claude Sonnet 4.6 measured |
| LMArena Text: longer queries | Not measured | 1479 | Only Claude Sonnet 4.6 measured |
| LMArena Text: writing and language | Not measured | 1444 | Only Claude Sonnet 4.6 measured |
| LMArena Text: entertainment and media | Not measured | 1431 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: English | Not measured | 1298 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: Chinese | Not measured | 1307 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: reading text | Not measured | 1296 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: humor | Not measured | 1248 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: recognizing things | Not measured | 1195 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: creative writing | Not measured | 1272 | Only Claude Sonnet 4.6 measured |
| LMArena Document | Not measured | 1482 | Only Claude Sonnet 4.6 measured |
| Neither model has a result on the other 9 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 4.6 | 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$0.0003Fits in one request
Claude Sonnet 4.6$0.011Fits in one request
Claude Haiku 5.5 costs less
Repository review
50K input + 3K output tokens
Claude Haiku 5.5$0.0065Fits in one request
Claude Sonnet 4.6$0.195Fits in one request
Claude Haiku 5.5 costs less
Agent loop
200K input + 10K output tokens
Claude Haiku 5.5$0.025Fits in one request
Claude Sonnet 4.6$0.750Fits in one request
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 Sonnet 4.6 |
|---|---|---|
| Maker | Anthropic | Anthropic |
| Weights | No public weights listed | No public weights listed |
| Released | October 7, 2026 | February 17, 2026 |
| Context window | 1M | 1M |
| Input $/Mtok | $0.10 | $3.00 |
| Output $/Mtok | $0.50 | $15.00 |
| Coding index | 70.7 | 57.3 |
| Independent measurements behind it | 5 | 5 |
| Cost per solved task | Not computable | $1.68 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Haiku 5.5 better than Claude Sonnet 4.6 for coding?
Pick Claude Haiku 5.5 for coding: it scores 70.7 on the Vibe Coding Index against 57.3 for Claude Sonnet 4.6. Pick Claude Sonnet 4.6 for work measured by Excel Modeling Benchmark (Vals AI) or Vending-Bench 2, where it leads.
Which is cheaper, Claude Haiku 5.5 or Claude Sonnet 4.6?
Claude Haiku 5.5 costs $0.10 per million input tokens and $0.50 per million output tokens; Claude Sonnet 4.6 costs $3.00 and $15.00. Cost per solved task is known for Claude Sonnet 4.6 only.
Which has the bigger context window, Claude Haiku 5.5 or Claude Sonnet 4.6?
Neither: both take 1M tokens.
How many benchmarks is the Claude Haiku 5.5 vs Claude Sonnet 4.6 comparison based on?
43 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 Sonnet 4.6's on 5. Where only one model has a result, the page says so instead of naming a winner.