Claude Sonnet 5 vs GPT-5.2
Claude Sonnet 5 leads GPT-5.2 on coding, 57.7 to 55.9 on the Vibe Coding Index, and Claude Sonnet 5 costs 1.4x less per output token ($10.00 vs $14.00 per million). Updated October 22, 2026; 85 benchmarks have a result for both.
Claude Sonnet 5 or GPT-5.2: which to pick
Claude Sonnet 5 and GPT-5.2 are level for coding, 57.7 against 55.9 on the Vibe Coding Index, so price and context decide.
Pick GPT-5.2 for work measured by LiveCodeBench (Vals AI) or WeirdML, where it leads. Pick Claude Sonnet 5 for output tokens 1.4x cheaper ($10.00 vs $14.00 per million) and a lower cost per solved task ($1.87 against $1.91).
- Shared results
- 85
- Claude Sonnet 5 only
- 36
- GPT-5.2 only
- 29
- 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
No call
Level: 57.7 against 55.9, inside 2 points.
6 and 3 benchmarks behind the scores
Intelligence
AA Intelligence Index
Claude Sonnet 5
Claude Sonnet 5 leads, 38.2 against 30.4.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Claude Sonnet 5 has a AA Agentic Index result, 43.6.
Directional only
Reasoning
GPQA Diamond
GPT-5.2
GPT-5.2 leads, 91.7% against 88.9%.
Same board, both measured
Math
AIME
No call
Only GPT-5.2 has a AIME result, 96.9%.
Directional only
Knowledge
MMLU Pro
Claude Sonnet 5
Claude Sonnet 5 leads, 87.5% against 86.2%.
Same board, both measured
Web design
Design Arena: websites
Claude Sonnet 5
Claude Sonnet 5 leads, 1274 against 1195.
Same board, both measured
Long documents
Listed context window
Claude Sonnet 5
Claude Sonnet 5 takes 1M tokens against 400K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Claude Sonnet 5
Claude Sonnet 5 $0.0070, GPT-5.2 $0.0088 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Claude Sonnet 5
Claude Sonnet 5 $1.87, GPT-5.2 $1.91.
Published attempt cost and pass rate
At a glance
- Coding index
- Claude Sonnet 5 57.7 from 6 benchmarks, GPT-5.2 55.9 from 3 benchmarks.
- Widest category
- Agentic work, 18.3 points apart on average over 6 shared benchmarks.
- Widest single gap
- ProofBench: Claude Sonnet 5 77.0% against GPT-5.2 15.0%, 62.0 points apart on a 0 to 100 scale.
- Output price
- Claude Sonnet 5 $10.00 against GPT-5.2 $14.00 per million output tokens, 1.4x apart.
- Context window
- Claude Sonnet 5 takes 1M tokens, GPT-5.2 400K.
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 Sonnet 5 | Bars | GPT-5.2 | Ties |
|---|---|---|---|---|
| Coding | 8 | 2 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 6 | 0 | ||
| Reasoning | 9 | 4 | ||
| Math | 3 | 3 | ||
| Knowledge | 8 | 3 | ||
| Web design | 8 | 0 | ||
| General | 24 | 5 | ||
| All shared boards | 68 | 17 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index82
| Benchmark | Claude Sonnet 5 | GPT-5.2 | Leader |
|---|---|---|---|
| SWE-bench Verified | Not measured | 72.8% | Only GPT-5.2 measured |
| SWE-bench Multilingual | Not measured | 66.7% | Only GPT-5.2 measured |
| LiveBench coding | 74.7% | Not measured | Only Claude Sonnet 5 measured |
| LMArena WebDev | 1540 | 1417 | Claude Sonnet 5 |
| AA Coding Index | 71.5 | Not measured | Only Claude Sonnet 5 measured |
| LiveCodeBench (Vals AI) | 82.4% | 85.4% | GPT-5.2 |
| SWE-bench Verified (Vals AI) | 79.6% | 75.8% | Claude Sonnet 5 |
| Terminal-Bench 4 (Vals AI) | 10.6% | Not measured | Only Claude Sonnet 5 measured |
| Vibe Code Bench (Vals AI) | 81.3% | 53.5% | Claude Sonnet 5 |
| Code migration (Vals AI) | 44.4% | Not measured | Only Claude Sonnet 5 measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only Claude Sonnet 5 measured |
| IOI (Vals AI) | 45.0% | Not measured | Only Claude Sonnet 5 measured |
| Terminal-Bench 2.1 (Vals AI) | 74.5% | Not measured | Only Claude Sonnet 5 measured |
| Vibe Code Bench 1-100 (Vals AI) | 13.8% | Not measured | Only Claude Sonnet 5 measured |
| SkillsBench (Vals AI) | 46.5% | Not measured | Only Claude Sonnet 5 measured |
| CyberBench (Vals AI) | 70.3% | Not measured | Only Claude Sonnet 5 measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 89.4% | Only GPT-5.2 measured |
| SciCode | 54.3% | Not measured | Only Claude Sonnet 5 measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 47.0% | Only GPT-5.2 measured |
| Terminal-Bench | Not measured | 64.9% | Only GPT-5.2 measured |
| FrontierCode | 42.7% | Not measured | Only Claude Sonnet 5 measured |
| GSO-Bench | 37.3% | 27.5% | Claude Sonnet 5 |
| WeirdML | 68.8% | 72.2% | GPT-5.2 |
| ALE-Bench | 1463.1 | 1293.5 | Claude Sonnet 5 |
| LMArena Text: coding | 1483 | 1442 | Claude Sonnet 5 |
| LMArena Text: software and IT | 1474 | 1434 | Claude Sonnet 5 |
| LMArena WebDev: React | 1541 | Not measured | Only Claude Sonnet 5 measured |
| LMArena WebDev: HTML | 1527 | 1427 | Claude Sonnet 5 |
| LMArena WebDev: from an image | 1540 | Not measured | Only Claude Sonnet 5 measured |
| SciCode | 10.7% | Not measured | Only Claude Sonnet 5 measured |
| Neither model has a result on the other 9 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Claude Sonnet 5 | GPT-5.2 | Leader |
|---|---|---|---|
| AA Intelligence Index | 38.2 | 30.4 | Claude Sonnet 5 |
| Epoch Capabilities Index | 156.9 | 153.7 | Claude Sonnet 5 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | Claude Sonnet 5 | GPT-5.2 | Leader |
|---|---|---|---|
| AA Agentic Index | 43.6 | Not measured | Only Claude Sonnet 5 measured |
| Design Arena: agents building web apps | 1245 | 1086 | Claude Sonnet 5 |
| Design Arena: agents building full-stack apps | 1249 | 1043 | Claude Sonnet 5 |
| Design Arena: agents building mobile apps | 1215 | 1093 | Claude Sonnet 5 |
| Design Arena: agents building Android apps | 1227 | 1059 | Claude Sonnet 5 |
| Design Arena: agents building Godot games | 1225 | 1142 | Claude Sonnet 5 |
| Design Arena: agents building games | 1227 | Not measured | Only Claude Sonnet 5 measured |
| Design Arena: HTML slides | 1168 | Not measured | Only Claude Sonnet 5 measured |
| Design Arena: PowerPoint slides | 1191 | Not measured | Only Claude Sonnet 5 measured |
| Design Arena: spreadsheets | 1035 | Not measured | Only Claude Sonnet 5 measured |
| LiveBench research | 82.8% | Not measured | Only Claude Sonnet 5 measured |
| LiveBench browser | 72.0% | Not measured | Only Claude Sonnet 5 measured |
| LiveBench long agentic | 39.0% | Not measured | Only Claude Sonnet 5 measured |
| SREBench (Vals AI) | 0.4% | Not measured | Only Claude Sonnet 5 measured |
| Finance Agent (Vals AI) | 53.9% | Not measured | Only Claude Sonnet 5 measured |
| Tax Agent Bench (Vals AI) | 29.4% | Not measured | Only Claude Sonnet 5 measured |
| Legal Research Bench (Vals AI) | 41.8% | Not measured | Only Claude Sonnet 5 measured |
| Excel Modeling Benchmark (Vals AI) | 66.3% | Not measured | Only Claude Sonnet 5 measured |
| Terminal-Bench-Science (Vals AI) | 5.7% | Not measured | Only Claude Sonnet 5 measured |
| τ²-Bench Telecom | Not measured | 84.8% | Only GPT-5.2 measured |
| APEX-Agents | 54.5% | Not measured | Only Claude Sonnet 5 measured |
| METR time horizons | Not measured | 75.3% | Only GPT-5.2 measured |
| Remote Labor Index | Not measured | 2.5% | Only GPT-5.2 measured |
| GDPval | Not measured | 49.7% | Only GPT-5.2 measured |
| DeepResearch Bench | Not measured | 41.1% | Only GPT-5.2 measured |
| Vending-Bench 2 | $6,378 | $3,591 | Claude Sonnet 5 |
| Berkeley Function Calling Leaderboard (BFCL) | Not measured | 55.9% | Only GPT-5.2 measured |
| Neither model has a result on the other 14 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index94
| Benchmark | Claude Sonnet 5 | GPT-5.2 | Leader |
|---|---|---|---|
| GPQA Diamond | 88.9% | 91.7% | GPT-5.2 |
| LiveBench reasoning | 88.7% | Not measured | Only Claude Sonnet 5 measured |
| MMMU Pro (Vals AI) | 83.0% | 86.7% | GPT-5.2 |
| GPQA Diamond (Artificial Analysis run) | 91.1% | 90.3% | Claude Sonnet 5 |
| Humanity's Last Exam (Artificial Analysis run) | 41.3% | 37.7% | Claude Sonnet 5 |
| GPQA Diamond (Epoch AI run) | 90.5% | 91.4% | GPT-5.2 |
| ARC-AGI | Not measured | 86.2% | Only GPT-5.2 measured |
| ARC-AGI-2 | Not measured | 52.9% | Only GPT-5.2 measured |
| SimpleBench | 60.6% | 45.8% | Claude Sonnet 5 |
| Chess puzzles | 45.0% | 48.0% | GPT-5.2 |
| Mystery game puzzles | 35.0% | 23.0% | Claude Sonnet 5 |
| EBR-bench | Not measured | 23.0% | Only GPT-5.2 measured |
| VPCT | Not measured | 84.0% | Only GPT-5.2 measured |
| LMArena Text: hard prompts | 1461 | 1428 | Claude Sonnet 5 |
| LMArena Text: expert | 1489 | 1445 | Claude Sonnet 5 |
| LMArena Text: hard prompts in English | 1464 | 1431 | Claude Sonnet 5 |
| LMArena Vision: diagrams | 1293 | 1265 | Claude Sonnet 5 |
| LMArena Vision: homework | 1311 | 1276 | Claude Sonnet 5 |
| Kagi LLM Benchmark | Not measured | 73.3% | Only GPT-5.2 measured |
| Neither model has a result on the other 3 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index33
| Benchmark | Claude Sonnet 5 | GPT-5.2 | Leader |
|---|---|---|---|
| AIME | Not measured | 96.9% | Only GPT-5.2 measured |
| MGSM (Vals AI) | Not measured | 94.0% | Only GPT-5.2 measured |
| AA Math Index | Not measured | 99.0 | Only GPT-5.2 measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 99.0% | Only GPT-5.2 measured |
| OTIS Mock AIME | 94.7% | 96.1% | GPT-5.2 |
| FrontierMath, tiers 1 to 3 | 65.6% | 67.4% | GPT-5.2 |
| FrontierMath, tier 4 | 29.3% | 31.7% | GPT-5.2 |
| ProofBench | 77.0% | 15.0% | Claude Sonnet 5 |
| LMArena Text: math | 1467 | 1440 | Claude Sonnet 5 |
| LMArena Text: mathematical work | 1481 | 1442 | Claude Sonnet 5 |
| Neither model has a result on the other 3 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index83
| Benchmark | Claude Sonnet 5 | GPT-5.2 | Leader |
|---|---|---|---|
| MMLU Pro | 87.5% | 86.2% | Claude Sonnet 5 |
| LegalBench (Vals AI) | 83.9% | 82.8% | Claude Sonnet 5 |
| CaseLaw (Vals AI) | Not measured | 66.0% | Only GPT-5.2 measured |
| CorpFin (Vals AI) | 67.0% | 65.9% | Claude Sonnet 5 |
| TaxEval (Vals AI) | 75.6% | 75.8% | GPT-5.2 |
| MortgageTax (Vals AI) | 70.0% | 67.1% | Claude Sonnet 5 |
| MedQA (Vals AI) | Not measured | 94.1% | Only GPT-5.2 measured |
| MedCode (Vals AI) | 47.5% | 49.8% | GPT-5.2 |
| MMLU Pro (Artificial Analysis run) | Not measured | 87.4% | Only GPT-5.2 measured |
| SimpleQA Verified | 33.7% | 37.1% | GPT-5.2 |
| LMArena Text: science | 1460 | 1425 | Claude Sonnet 5 |
| LMArena Text: medicine and healthcare | 1442 | 1425 | Claude Sonnet 5 |
| LMArena Text: legal and government | 1454 | 1427 | Claude Sonnet 5 |
| LMArena Text: business and finance | 1436 | 1414 | Claude Sonnet 5 |
| FACTS Benchmark Suite | Not measured | 48.5% | Only GPT-5.2 measured |
| Neither model has a result on the other 3 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index80
| Benchmark | Claude Sonnet 5 | GPT-5.2 | Leader |
|---|---|---|---|
| Design Arena: websites | 1274 | 1195 | Claude Sonnet 5 |
| Design Arena: UI components | 1277 | 1190 | Claude Sonnet 5 |
| Design Arena: game dev | 1277 | 1195 | Claude Sonnet 5 |
| Design Arena: data viz | 1251 | 1202 | Claude Sonnet 5 |
| Design Arena: 3D | 1261 | 1079 | Claude Sonnet 5 |
| Design Arena: all code categories | 1275 | 1170 | Claude Sonnet 5 |
| Design Arena: SVG | 1192 | 1139 | Claude Sonnet 5 |
| Design Arena: ASCII art | 1207 | 1152 | Claude Sonnet 5 |
| LiveBench design | 82.0% | Not measured | Only Claude Sonnet 5 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index245
| Benchmark | Claude Sonnet 5 | GPT-5.2 | Leader |
|---|---|---|---|
| LiveBench data analysis | 71.7% | Not measured | Only Claude Sonnet 5 measured |
| Public Benefits Bench (Vals AI) | 66.0% | Not measured | Only Claude Sonnet 5 measured |
| SAGE (Vals AI) | 48.9% | 49.3% | GPT-5.2 |
| MedScribe (Vals AI) | 76.0% | 84.4% | GPT-5.2 |
| Vals Multimodal Index | 68.8% | Not measured | Only Claude Sonnet 5 measured |
| IFBench | Not measured | 75.4% | Only GPT-5.2 measured |
| Long context reasoning (AA-LCR) | 82.0% | 82.7% | GPT-5.2 |
| DTBench | 92.5% | 90.9% | Claude Sonnet 5 |
| LMCA | 50.0% | 43.9% | Claude Sonnet 5 |
| CL-bench | Not measured | 18.2% | Only GPT-5.2 measured |
| Surface Evolver Bench | 60.0% | Not measured | Only Claude Sonnet 5 measured |
| Furniture assembly | Not measured | 38.3% | Only GPT-5.2 measured |
| LMArena Text | 1442 | 1416 | Claude Sonnet 5 |
| LMArena Text: instruction following | 1452 | 1409 | Claude Sonnet 5 |
| LMArena Text: creative writing | 1416 | 1376 | Claude Sonnet 5 |
| LMArena Vision | 1273 | 1246 | Claude Sonnet 5 |
| LMArena Text: English | 1450 | 1422 | Claude Sonnet 5 |
| LMArena Text: non-English | 1431 | 1405 | Claude Sonnet 5 |
| LMArena Text: Chinese | 1478 | 1447 | Claude Sonnet 5 |
| LMArena Text: German | 1442 | 1444 | GPT-5.2 |
| LMArena Text: Japanese | 1422 | 1420 | Claude Sonnet 5 |
| LMArena Text: Korean | 1412 | 1374 | Claude Sonnet 5 |
| LMArena Text: French | 1463 | 1454 | Claude Sonnet 5 |
| LMArena Text: Spanish | 1438 | 1413 | Claude Sonnet 5 |
| LMArena Text: Russian | 1450 | 1415 | Claude Sonnet 5 |
| LMArena Text: Polish | 1425 | 1426 | GPT-5.2 |
| LMArena Text: multi-turn | 1453 | 1419 | Claude Sonnet 5 |
| LMArena Text: longer queries | 1462 | 1413 | Claude Sonnet 5 |
| LMArena Text: writing and language | 1436 | 1392 | Claude Sonnet 5 |
| LMArena Text: entertainment and media | 1411 | 1378 | Claude Sonnet 5 |
| LMArena Vision: English | 1275 | 1243 | Claude Sonnet 5 |
| LMArena Vision: Chinese | 1331 | 1305 | Claude Sonnet 5 |
| LMArena Vision: reading text | 1288 | 1257 | Claude Sonnet 5 |
| LMArena Vision: humor | 1261 | 1227 | Claude Sonnet 5 |
| LMArena Vision: captioning | Not measured | 1267 | Only GPT-5.2 measured |
| LMArena Vision: recognizing things | Not measured | 1232 | Only GPT-5.2 measured |
| LMArena Vision: creative writing | 1246 | 1225 | Claude Sonnet 5 |
| LMArena Document | 1466 | 1405 | Claude Sonnet 5 |
| Neither model has a result on the other 8 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Sonnet 5 | GPT-5.2 | 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 Sonnet 5 costs less
Repository review
50K input + 3K output tokens
GPT-5.2 costs less
Agent loop
200K input + 10K output tokens
GPT-5.2 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 Sonnet 5 | OpenAI GPT-5.2 |
|---|---|---|
| Maker | Anthropic | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | June 30, 2026 | December 10, 2025 |
| Context window | 1M | 400K |
| Input $/Mtok | $2.00 | $1.75 |
| Output $/Mtok | $10.00 | $14.00 |
| Coding index | 57.7 | 55.9 |
| Independent measurements behind it | 6 | 3 |
| Cost per solved task | $1.87 | $1.91 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Sonnet 5 better than GPT-5.2 for coding?
Claude Sonnet 5 and GPT-5.2 are level for coding, 57.7 against 55.9 on the Vibe Coding Index, so price and context decide. Pick GPT-5.2 for work measured by LiveCodeBench (Vals AI) or WeirdML, where it leads. Pick Claude Sonnet 5 for output tokens 1.4x cheaper ($10.00 vs $14.00 per million) and a lower cost per solved task ($1.87 against $1.91).
Which is cheaper, Claude Sonnet 5 or GPT-5.2?
Claude Sonnet 5 costs $2.00 per million input tokens and $10.00 per million output tokens; GPT-5.2 costs $1.75 and $14.00. Per solved SWE-bench Verified task, Claude Sonnet 5 costs $1.87 and GPT-5.2 $1.91.
Which has the bigger context window, Claude Sonnet 5 or GPT-5.2?
Claude Sonnet 5, with 1M tokens against 400K.
How many benchmarks is the Claude Sonnet 5 vs GPT-5.2 comparison based on?
85 benchmarks have a result for both models as of October 22, 2026. Claude Sonnet 5's coding index rests on 6 independent benchmarks and GPT-5.2's on 3. Where only one model has a result, the page says so instead of naming a winner.