Claude Haiku 4.5 vs GPT-5.6 Sol
GPT-5.6 Sol leads Claude Haiku 4.5 on coding, 73.5 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; 84 benchmarks have a result for both.
Claude Haiku 4.5 or GPT-5.6 Sol: which to pick
Pick GPT-5.6 Sol for coding: it scores 73.5 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) and a lower cost per solved task ($0.56 against $1.20).
- Shared results
- 84
- Claude Haiku 4.5 only
- 11
- GPT-5.6 Sol only
- 55
- 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
GPT-5.6 Sol
GPT-5.6 Sol scores 73.5 against 35.6.
4 and 6 benchmarks behind the scores
Intelligence
AA Intelligence Index
GPT-5.6 Sol
GPT-5.6 Sol leads, 47.0 against 16.9.
Same board, both measured
Agentic work
AA Agentic Index
GPT-5.6 Sol
GPT-5.6 Sol leads, 50.2 against 8.0.
Same board, both measured
Reasoning
GPQA Diamond
GPT-5.6 Sol
GPT-5.6 Sol leads, 95.2% against 72.2%.
Same board, both measured
Math
AIME
No call
Only Claude Haiku 4.5 has a AIME result, 82.7%.
Directional only
Knowledge
MMLU Pro
GPT-5.6 Sol
GPT-5.6 Sol leads, 89.1% against 78.7%.
Same board, both measured
Web design
Design Arena: websites
GPT-5.6 Sol
GPT-5.6 Sol leads, 1321 against 1122.
Same board, both measured
Long documents
Listed context window
GPT-5.6 Sol
GPT-5.6 Sol takes 1.05M tokens against 200K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Claude Haiku 4.5
Claude Haiku 4.5 $0.0035, GPT-5.6 Sol $0.014 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Claude Haiku 4.5
Claude Haiku 4.5 $0.56, GPT-5.6 Sol $1.20.
Published attempt cost and pass rate
At a glance
- Coding index
- Claude Haiku 4.5 35.6 from 4 benchmarks, GPT-5.6 Sol 73.5 from 6 benchmarks.
- Widest category
- Agentic work, 36.1 points apart on average over 10 shared benchmarks.
- Widest single gap
- ARC-AGI-2: Claude Haiku 4.5 4.0% against GPT-5.6 Sol 92.5%, 88.5 points apart on a 0 to 100 scale.
- Output price
- Claude Haiku 4.5 $5.00 against GPT-5.6 Sol $20.00 per million output tokens, 4.0x apart.
- Context window
- Claude Haiku 4.5 takes 200K tokens, GPT-5.6 Sol 1.05M.
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 | GPT-5.6 Sol | Ties |
|---|---|---|---|---|
| Coding | 0 | 14 | ||
| Intelligence | 0 | 2 | ||
| Agentic work | 0 | 10 | ||
| Reasoning | 0 | 10 | ||
| Math | 0 | 3 | ||
| Knowledge | 0 | 11 | ||
| Web design | 0 | 9 | ||
| General | 0 | 24 | 1 tie | |
| All shared boards | 0 | 83 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index014
| Benchmark | Claude Haiku 4.5 | GPT-5.6 Sol | 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% | 89.7% | GPT-5.6 Sol |
| LMArena WebDev | 1330 | 1616 | GPT-5.6 Sol |
| AA Coding Index | 43.9 | 77.4 | GPT-5.6 Sol |
| LiveCodeBench (Vals AI) | 41.2% | 82.6% | GPT-5.6 Sol |
| SWE-bench Verified (Vals AI) | 66.6% | 96.2% | GPT-5.6 Sol |
| Terminal-Bench 4 (Vals AI) | Not measured | 37.9% | Only GPT-5.6 Sol measured |
| Vibe Code Bench (Vals AI) | 11.4% | 80.5% | GPT-5.6 Sol |
| Code migration (Vals AI) | 7.5% | 52.9% | GPT-5.6 Sol |
| ProgramBench (Vals AI) | 0.0% | 1.5% | GPT-5.6 Sol |
| IOI (Vals AI) | Not measured | 91.2% | Only GPT-5.6 Sol measured |
| Terminal-Bench 2.1 (Vals AI) | 43.8% | 85.8% | GPT-5.6 Sol |
| Vibe Code Bench 1-100 (Vals AI) | Not measured | 20.0% | Only GPT-5.6 Sol measured |
| SkillsBench (Vals AI) | Not measured | 54.1% | Only GPT-5.6 Sol measured |
| CyberBench (Vals AI) | Not measured | 76.3% | Only GPT-5.6 Sol measured |
| SciCode | Not measured | 57.8% | Only GPT-5.6 Sol measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 65.9% | Only GPT-5.6 Sol measured |
| Terminal-Bench | 35.5% | Not measured | Only Claude Haiku 4.5 measured |
| FrontierCode | Not measured | 47.5% | Only GPT-5.6 Sol measured |
| FrontierSWE | Not measured | 32.2% | Only GPT-5.6 Sol measured |
| GSO-Bench | Not measured | 76.5% | Only GPT-5.6 Sol measured |
| WeirdML | 45.4% | 88.8% | GPT-5.6 Sol |
| MirrorCode | Not measured | 20.0% | Only GPT-5.6 Sol measured |
| ALE-Bench | Not measured | 2176.9 | Only GPT-5.6 Sol measured |
| LMArena Text: coding | 1454 | 1496 | GPT-5.6 Sol |
| LMArena Text: software and IT | 1439 | 1489 | GPT-5.6 Sol |
| LMArena WebDev: React | 1326 | 1620 | GPT-5.6 Sol |
| LMArena WebDev: HTML | 1318 | 1618 | GPT-5.6 Sol |
| LMArena WebDev: from an image | Not measured | 1609 | Only GPT-5.6 Sol measured |
| SciCode | Not measured | 10.8% | Only GPT-5.6 Sol measured |
| Neither model has a result on the other 8 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Claude Haiku 4.5 | GPT-5.6 Sol | Leader |
|---|---|---|---|
| AA Intelligence Index | 16.9 | 47.0 | GPT-5.6 Sol |
| Epoch Capabilities Index | 142.2 | 162.8 | GPT-5.6 Sol |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index010
| Benchmark | Claude Haiku 4.5 | GPT-5.6 Sol | Leader |
|---|---|---|---|
| AA Agentic Index | 8.0 | 50.2 | GPT-5.6 Sol |
| Design Arena: agents building web apps | Not measured | 1245 | Only GPT-5.6 Sol measured |
| Design Arena: agents building full-stack apps | Not measured | 1261 | Only GPT-5.6 Sol measured |
| Design Arena: agents building mobile apps | Not measured | 1248 | Only GPT-5.6 Sol measured |
| Design Arena: agents building Android apps | Not measured | 1240 | Only GPT-5.6 Sol measured |
| Design Arena: agents building Godot games | Not measured | 1260 | Only GPT-5.6 Sol measured |
| Design Arena: agents building games | Not measured | 1266 | Only GPT-5.6 Sol measured |
| Design Arena: HTML slides | Not measured | 1236 | Only GPT-5.6 Sol measured |
| Design Arena: PowerPoint slides | Not measured | 1215 | Only GPT-5.6 Sol measured |
| LiveBench research | 42.2% | 54.5% | GPT-5.6 Sol |
| LiveBench browser | 40.0% | 88.0% | GPT-5.6 Sol |
| LiveBench long agentic | 11.1% | 48.5% | GPT-5.6 Sol |
| SREBench (Vals AI) | Not measured | 30.5% | Only GPT-5.6 Sol measured |
| Finance Agent (Vals AI) | 31.0% | 53.8% | GPT-5.6 Sol |
| Tax Agent Bench (Vals AI) | 5.9% | 31.1% | GPT-5.6 Sol |
| Legal Research Bench (Vals AI) | 10.6% | 48.1% | GPT-5.6 Sol |
| Excel Modeling Benchmark (Vals AI) | 22.7% | 72.3% | GPT-5.6 Sol |
| BioMysteryBench (Vals AI) | Not measured | 71.1% | Only GPT-5.6 Sol measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 20.0% | Only GPT-5.6 Sol measured |
| Vals Science Index | Not measured | 41.5% | Only GPT-5.6 Sol measured |
| Time Horizon Index (Vals AI) | Not measured | 23.8% | Only GPT-5.6 Sol measured |
| CUA-bench (Vals AI) | Not measured | 8.3% | Only GPT-5.6 Sol measured |
| τ²-Bench Telecom | Not measured | 85.1% | Only GPT-5.6 Sol measured |
| OSWorld 2.0 | Not measured | 27.3% | Only GPT-5.6 Sol measured |
| BALROG | 31.2% | 60.0% | GPT-5.6 Sol |
| DeepResearch Bench | 45.5% | Not measured | Only Claude Haiku 4.5 measured |
| PostTrainBench | Not measured | 36.2% | Only GPT-5.6 Sol measured |
| Vending-Bench 2 | $459 | $9,619 | GPT-5.6 Sol |
| LMArena Search | Not measured | 1261 | Only GPT-5.6 Sol measured |
| Berkeley Function Calling Leaderboard (BFCL) | 68.7% | Not measured | Only Claude Haiku 4.5 measured |
| Neither model has a result on the other 11 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index010
| Benchmark | Claude Haiku 4.5 | GPT-5.6 Sol | Leader |
|---|---|---|---|
| GPQA Diamond | 72.2% | 95.2% | GPT-5.6 Sol |
| LiveBench reasoning | 61.7% | 91.7% | GPT-5.6 Sol |
| MysteryMechanism (Vals AI) | Not measured | 33.3% | Only GPT-5.6 Sol measured |
| MMMU Pro (Vals AI) | 46.1% | 88.8% | GPT-5.6 Sol |
| GPQA Diamond (Artificial Analysis run) | Not measured | 94.1% | Only GPT-5.6 Sol measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 49.5% | Only GPT-5.6 Sol measured |
| GPQA Diamond (Epoch AI run) | 71.2% | 93.5% | GPT-5.6 Sol |
| ARC-AGI | 47.7% | 97.5% | GPT-5.6 Sol |
| ARC-AGI-2 | 4.0% | 92.5% | GPT-5.6 Sol |
| SimpleBench | Not measured | 64.8% | Only GPT-5.6 Sol measured |
| Chess puzzles | 8.0% | 59.0% | GPT-5.6 Sol |
| Mystery game puzzles | Not measured | 58.0% | Only GPT-5.6 Sol measured |
| EBR-bench | Not measured | 44.8% | Only GPT-5.6 Sol measured |
| LMArena Text: hard prompts | 1421 | 1483 | GPT-5.6 Sol |
| LMArena Text: expert | 1443 | 1517 | GPT-5.6 Sol |
| LMArena Text: hard prompts in English | 1435 | 1477 | GPT-5.6 Sol |
| LMArena Vision: diagrams | Not measured | 1305 | Only GPT-5.6 Sol measured |
| LMArena Vision: homework | Not measured | 1305 | Only GPT-5.6 Sol measured |
| Kagi LLM Benchmark | Not measured | 67.0% | Only GPT-5.6 Sol measured |
| Neither model has a result on the other 3 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | Claude Haiku 4.5 | GPT-5.6 Sol | 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% | GPT-5.6 Sol |
| FrontierMath, tiers 1 to 3 | Not measured | 89.1% | Only GPT-5.6 Sol measured |
| FrontierMath, tier 4 | Not measured | 82.9% | Only GPT-5.6 Sol measured |
| ProofBench | Not measured | 83.0% | Only GPT-5.6 Sol measured |
| MATH level 5 | 96.4% | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text: math | 1398 | 1473 | GPT-5.6 Sol |
| LMArena Text: mathematical work | 1430 | 1496 | GPT-5.6 Sol |
| Neither model has a result on the other 4 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index011
| Benchmark | Claude Haiku 4.5 | GPT-5.6 Sol | Leader |
|---|---|---|---|
| MMLU Pro | 78.7% | 89.1% | GPT-5.6 Sol |
| LegalBench (Vals AI) | 81.2% | 87.0% | GPT-5.6 Sol |
| CaseLaw (Vals AI) | 56.5% | Not measured | Only Claude Haiku 4.5 measured |
| CorpFin (Vals AI) | 60.6% | 64.4% | GPT-5.6 Sol |
| TaxEval (Vals AI) | 67.5% | 74.8% | GPT-5.6 Sol |
| MortgageTax (Vals AI) | 62.2% | 67.3% | GPT-5.6 Sol |
| MedQA (Vals AI) | 79.6% | Not measured | Only Claude Haiku 4.5 measured |
| MedCode (Vals AI) | 32.7% | 44.0% | GPT-5.6 Sol |
| SimpleQA Verified | 13.2% | 69.7% | GPT-5.6 Sol |
| LMArena Text: science | 1406 | 1455 | GPT-5.6 Sol |
| LMArena Text: medicine and healthcare | 1394 | 1431 | GPT-5.6 Sol |
| LMArena Text: legal and government | 1400 | 1476 | GPT-5.6 Sol |
| LMArena Text: business and finance | 1400 | 1457 | GPT-5.6 Sol |
| FACTS Benchmark Suite | Not measured | 52.4% | Only GPT-5.6 Sol measured |
| Neither model has a result on the other 4 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index09
| Benchmark | Claude Haiku 4.5 | GPT-5.6 Sol | Leader |
|---|---|---|---|
| Design Arena: websites | 1122 | 1321 | GPT-5.6 Sol |
| Design Arena: UI components | 1097 | 1335 | GPT-5.6 Sol |
| Design Arena: game dev | 1097 | 1367 | GPT-5.6 Sol |
| Design Arena: data viz | 1124 | 1322 | GPT-5.6 Sol |
| Design Arena: 3D | 1071 | 1384 | GPT-5.6 Sol |
| Design Arena: all code categories | 1114 | 1341 | GPT-5.6 Sol |
| Design Arena: SVG | 1025 | 1314 | GPT-5.6 Sol |
| Design Arena: ASCII art | 1140 | 1304 | GPT-5.6 Sol |
| LiveBench design | 59.9% | 81.6% | GPT-5.6 Sol |
GeneralLike-for-likePublic boards, shown as published, kept out of the index024
| Benchmark | Claude Haiku 4.5 | GPT-5.6 Sol | Leader |
|---|---|---|---|
| LiveBench data analysis | 59.3% | 79.8% | GPT-5.6 Sol |
| Public Benefits Bench (Vals AI) | 54.3% | 66.5% | GPT-5.6 Sol |
| SAGE (Vals AI) | 31.8% | 52.6% | GPT-5.6 Sol |
| MedScribe (Vals AI) | 85.2% | 85.2% | Tie |
| Vals Multimodal Index | 42.9% | 72.6% | GPT-5.6 Sol |
| IFBench | Not measured | 72.7% | Only GPT-5.6 Sol measured |
| Long context reasoning (AA-LCR) | Not measured | 84.0% | Only GPT-5.6 Sol measured |
| DTBench | 73.6% | 95.7% | GPT-5.6 Sol |
| LMCA | 30.9% | 58.5% | GPT-5.6 Sol |
| Surface Evolver Bench | Not measured | 93.1% | Only GPT-5.6 Sol measured |
| Furniture assembly | Not measured | 56.7% | Only GPT-5.6 Sol measured |
| ExploitBench | 13.7% | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text | 1397 | 1457 | GPT-5.6 Sol |
| LMArena Text: instruction following | 1414 | 1483 | GPT-5.6 Sol |
| LMArena Text: creative writing | 1371 | 1449 | GPT-5.6 Sol |
| LMArena Vision | Not measured | 1281 | Only GPT-5.6 Sol measured |
| LMArena Text: English | 1414 | 1457 | GPT-5.6 Sol |
| LMArena Text: non-English | 1377 | 1451 | GPT-5.6 Sol |
| LMArena Text: Chinese | 1418 | 1525 | GPT-5.6 Sol |
| LMArena Text: German | 1374 | 1476 | GPT-5.6 Sol |
| LMArena Text: Japanese | 1338 | 1472 | GPT-5.6 Sol |
| LMArena Text: Korean | 1346 | 1442 | GPT-5.6 Sol |
| LMArena Text: French | 1411 | 1472 | GPT-5.6 Sol |
| LMArena Text: Spanish | 1421 | 1443 | GPT-5.6 Sol |
| LMArena Text: Russian | 1381 | 1471 | GPT-5.6 Sol |
| LMArena Text: Polish | 1343 | 1459 | GPT-5.6 Sol |
| LMArena Text: multi-turn | 1409 | 1461 | GPT-5.6 Sol |
| LMArena Text: longer queries | 1427 | 1481 | GPT-5.6 Sol |
| LMArena Text: writing and language | 1386 | 1462 | GPT-5.6 Sol |
| LMArena Text: entertainment and media | 1369 | 1450 | GPT-5.6 Sol |
| LMArena Vision: English | Not measured | 1280 | Only GPT-5.6 Sol measured |
| LMArena Vision: Chinese | Not measured | 1313 | Only GPT-5.6 Sol measured |
| LMArena Vision: reading text | Not measured | 1290 | Only GPT-5.6 Sol measured |
| LMArena Vision: humor | Not measured | 1246 | Only GPT-5.6 Sol measured |
| LMArena Vision: creative writing | Not measured | 1283 | Only GPT-5.6 Sol measured |
| LMArena Document | 1420 | 1483 | GPT-5.6 Sol |
| Neither model has a result on the other 10 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Haiku 4.5 | GPT-5.6 Sol | 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 | OpenAI GPT-5.6 Sol |
|---|---|---|
| Maker | Anthropic | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | October 15, 2025 | July 9, 2026 |
| Context window | 200K | 1.05M |
| Input $/Mtok | $1.00 | $4.00 |
| Output $/Mtok | $5.00 | $20.00 |
| Coding index | 35.6 | 73.5 |
| Independent measurements behind it | 4 | 6 |
| Cost per solved task | $0.56 | $1.20 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Haiku 4.5 better than GPT-5.6 Sol for coding?
Pick GPT-5.6 Sol for coding: it scores 73.5 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) and a lower cost per solved task ($0.56 against $1.20).
Which is cheaper, Claude Haiku 4.5 or GPT-5.6 Sol?
Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens; GPT-5.6 Sol costs $4.00 and $20.00. Per solved SWE-bench Verified task, Claude Haiku 4.5 costs $0.56 and GPT-5.6 Sol $1.20.
Which has the bigger context window, Claude Haiku 4.5 or GPT-5.6 Sol?
GPT-5.6 Sol, with 1.05M tokens against 200K.
How many benchmarks is the Claude Haiku 4.5 vs GPT-5.6 Sol 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 GPT-5.6 Sol's on 6. Where only one model has a result, the page says so instead of naming a winner.