Claude Sonnet 5.5 vs Gemini 4 Carbon
Gemini 4 Carbon leads Claude Sonnet 5.5 on coding, 86.6 to 84.3 on the Vibe Coding Index, at about the same output price ($10.00 vs $10.00 per million). Updated November 14, 2026; 70 benchmarks have a result for both.
vs boards won of 70 shared, 1 tied
Claude Sonnet 5.5 or Gemini 4 Carbon: which to pick
Pick Gemini 4 Carbon for coding: it scores 86.6 on the Vibe Coding Index against 84.3 for Claude Sonnet 5.5, though on 4 benchmarks against 5.
Pick Claude Sonnet 5.5 for work measured by LMArena WebDev or Terminal-Bench 4 (Vals AI), where it leads.
Model AClaude Sonnet 5.5Anthropic84.3/1005 independent benchmarksNo cost per solved taskModel BGemini 4 CarbonGoogle86.6/1004 independent benchmarksNo cost per solved task
- Shared results
- 70
- Claude Sonnet 5.5 only
- 51
- Gemini 4 Carbon only
- 3
- 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
Gemini 4 Carbon
Gemini 4 Carbon scores 86.6 against 84.3.
5 and 4 benchmarks behind the scores
Intelligence
AA Intelligence Index
Gemini 4 Carbon
Gemini 4 Carbon leads, 60.0 against 54.5.
Same board, both measured
Web design
Design Arena: websites
Gemini 4 Carbon
Gemini 4 Carbon leads, 1319 against 1309.
Same board, both measured
General
Vals Index
Claude Sonnet 5.5
Claude Sonnet 5.5 leads, 70.8% against 70.5%.
Same board, both measured
Long documents
Listed context window
Gemini 4 Carbon
Gemini 4 Carbon takes 1.05M tokens against 1M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
No call
Claude Sonnet 5.5 $0.0070, Gemini 4 Carbon $0.0070 per turn.
List prices
At a glance
- Coding index
- Claude Sonnet 5.5 84.3 from 5 benchmarks, Gemini 4 Carbon 86.6 from 4 benchmarks.
- Widest category
- Knowledge, 14.4 points apart on average over 7 shared benchmarks.
- Widest single gap
- SimpleQA Verified: Claude Sonnet 5.5 46.5% against Gemini 4 Carbon 77.7%, 31.2 points apart on a 0 to 100 scale.
- Output price
- Both cost $10.00 per million output tokens.
- Context window
- Claude Sonnet 5.5 takes 1M tokens, Gemini 4 Carbon 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 Sonnet 5.5 | Bars | Gemini 4 Carbon | Ties |
|---|---|---|---|---|
| Coding | 6 | 7 | ||
| Intelligence | 0 | 2 | ||
| Agentic work | 2 | 8 | ||
| Reasoning | 1 | 11 | ||
| Math | 0 | 3 | 1 tie | |
| Knowledge | 0 | 7 | ||
| Web design | 3 | 3 | ||
| General | 3 | 13 | ||
| All shared boards | 15 | 54 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index67
| Benchmark | Claude Sonnet 5.5 | Gemini 4 Carbon | Leader |
|---|---|---|---|
| LiveBench coding | 77.2% | 91.0% | Gemini 4 Carbon |
| LMArena WebDev | 1772 | 1768 | Claude Sonnet 5.5 |
| Terminal-Bench 4 (Vals AI) | 64.1% | 63.1% | Claude Sonnet 5.5 |
| Vibe Code Bench (Vals AI) | 92.4% | 89.1% | Claude Sonnet 5.5 |
| Code migration (Vals AI) | 69.8% | Not measured | Only Claude Sonnet 5.5 measured |
| ProgramBench (Vals AI) | 6.5% | Not measured | Only Claude Sonnet 5.5 measured |
| IOI (Vals AI) | 83.1% | Not measured | Only Claude Sonnet 5.5 measured |
| Terminal-Bench 2.1 (Vals AI) | 83.2% | Not measured | Only Claude Sonnet 5.5 measured |
| SWE-bench Multimodal (Vals AI) | 58.3% | Not measured | Only Claude Sonnet 5.5 measured |
| CyberBench (Vals AI) | 59.6% | Not measured | Only Claude Sonnet 5.5 measured |
| SciCode | 61.0% | Not measured | Only Claude Sonnet 5.5 measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | 63.6% | 65.2% | Gemini 4 Carbon |
| FrontierCode 2.0 | 41.7% | 40.6% | Claude Sonnet 5.5 |
| FrontierCode | 52.1% | Not measured | Only Claude Sonnet 5.5 measured |
| FrontierSWE | 61.9% | 61.2% | Claude Sonnet 5.5 |
| MirrorCode | 97.8% | Not measured | Only Claude Sonnet 5.5 measured |
| ALE-Bench | 1819.1 | 2374.7 | Gemini 4 Carbon |
| LMArena Text: coding | 1501 | 1564 | Gemini 4 Carbon |
| LMArena Text: software and IT | 1497 | 1552 | Gemini 4 Carbon |
| LMArena WebDev: React | 1822 | 1792 | Claude Sonnet 5.5 |
| LMArena WebDev: HTML | 1753 | 1770 | Gemini 4 Carbon |
| LMArena WebDev: from an image | 1744 | Not measured | Only Claude Sonnet 5.5 measured |
| SciCode | 13.8% | Not measured | Only Claude Sonnet 5.5 measured |
| SWE-Bench Pro V2 | 95.8% | Not measured | Only Claude Sonnet 5.5 measured |
| GameBench 2 | 91.4 | Not measured | Only Claude Sonnet 5.5 measured |
| Terminal-Bench 4.0 | 61.8% | Not measured | Only Claude Sonnet 5.5 measured |
| CursorBench 4.0 | 55.5% | 60.4% | Gemini 4 Carbon |
| Neither model has a result on the other 25 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Claude Sonnet 5.5 | Gemini 4 Carbon | Leader |
|---|---|---|---|
| AA Intelligence Index | 54.5 | 60.0 | Gemini 4 Carbon |
| Epoch Capabilities Index | 166.6 | 169.8 | Gemini 4 Carbon |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index28
| Benchmark | Claude Sonnet 5.5 | Gemini 4 Carbon | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | 1296 | Not measured | Only Claude Sonnet 5.5 measured |
| Design Arena: agents building mobile apps | 1259 | 1284 | Gemini 4 Carbon |
| Design Arena: agents building Android apps | 1250 | Not measured | Only Claude Sonnet 5.5 measured |
| Design Arena: spreadsheets | 1238 | 1224 | Claude Sonnet 5.5 |
| LiveBench research | 88.0% | 92.5% | Gemini 4 Carbon |
| LiveBench browser | 89.3% | 80.0% | Claude Sonnet 5.5 |
| LiveBench long agentic | 63.4% | 72.7% | Gemini 4 Carbon |
| SREBench (Vals AI) | 30.1% | Not measured | Only Claude Sonnet 5.5 measured |
| Finance Agent (Vals AI) | 58.1% | Not measured | Only Claude Sonnet 5.5 measured |
| Tax Agent Bench (Vals AI) | 42.0% | 51.8% | Gemini 4 Carbon |
| Legal Research Bench (Vals AI) | 48.1% | 55.8% | Gemini 4 Carbon |
| Harvey's Legal Agent Benchmark (Vals AI) | 10.4% | 17.5% | Gemini 4 Carbon |
| Excel Modeling Benchmark (Vals AI) | 75.7% | Not measured | Only Claude Sonnet 5.5 measured |
| BioMysteryBench (Vals AI) | 81.1% | Not measured | Only Claude Sonnet 5.5 measured |
| Terminal-Bench-Science (Vals AI) | 47.1% | Not measured | Only Claude Sonnet 5.5 measured |
| Vals Science Index | 59.1% | Not measured | Only Claude Sonnet 5.5 measured |
| APEX-Agents | 75.5% | 84.2% | Gemini 4 Carbon |
| SpireClimb | 1.3% | Not measured | Only Claude Sonnet 5.5 measured |
| Vending-Bench 2 | $11,165 | $16,484 | Gemini 4 Carbon |
| Neither model has a result on the other 25 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index111
| Benchmark | Claude Sonnet 5.5 | Gemini 4 Carbon | Leader |
|---|---|---|---|
| LiveBench reasoning | 91.6% | 90.6% | Claude Sonnet 5.5 |
| MysteryMechanism (Vals AI) | 49.1% | Not measured | Only Claude Sonnet 5.5 measured |
| Humanity's Last Exam (Artificial Analysis run) | 55.0% | Not measured | Only Claude Sonnet 5.5 measured |
| HLE-Diamond (Artificial Analysis run) | 53.3% | 69.1% | Gemini 4 Carbon |
| CritPt v2, hard subset (Artificial Analysis run) | 59.0% | 69.0% | Gemini 4 Carbon |
| GPQA Diamond (Epoch AI run) | 95.6% | 96.5% | Gemini 4 Carbon |
| ARC-AGI | 96.5% | 98.5% | Gemini 4 Carbon |
| ARC-AGI-2 | 83.9% | 94.2% | Gemini 4 Carbon |
| SimpleBench | 75.9% | 92.6% | Gemini 4 Carbon |
| Chess puzzles | 59.0% | 78.0% | Gemini 4 Carbon |
| Mystery game puzzles | 65.0% | 73.0% | Gemini 4 Carbon |
| EBR-bench | Not measured | 87.6% | Only Gemini 4 Carbon measured |
| LMArena Text: hard prompts | 1488 | 1551 | Gemini 4 Carbon |
| LMArena Text: expert | 1508 | 1582 | Gemini 4 Carbon |
| LMArena Text: hard prompts in English | 1494 | 1545 | Gemini 4 Carbon |
| LMArena Vision: diagrams | 1316 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Vision: homework | 1322 | Not measured | Only Claude Sonnet 5.5 measured |
| Humanity's Last Exam (Scale run) | 37.5% | Not measured | Only Claude Sonnet 5.5 measured |
| GoBench | 2085 | Not measured | Only Claude Sonnet 5.5 measured |
| Neither model has a result on the other 7 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | Claude Sonnet 5.5 | Gemini 4 Carbon | Leader |
|---|---|---|---|
| OTIS Mock AIME | 100.0% | 100.0% | Tie |
| FrontierMath, tiers 1 to 3 | 88.8% | 90.5% | Gemini 4 Carbon |
| FrontierMath, tier 4 | 80.5% | 95.1% | Gemini 4 Carbon |
| ProofBench | 100.0% | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Text: math | 1483 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Text: mathematical work | 1501 | 1538 | Gemini 4 Carbon |
| MathArena | Not measured | 85.4% | Only Gemini 4 Carbon measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index07
| Benchmark | Claude Sonnet 5.5 | Gemini 4 Carbon | Leader |
|---|---|---|---|
| MedCode (Vals AI) | 52.9% | Not measured | Only Claude Sonnet 5.5 measured |
| AA-Omniscience | 30.5 | 53.5 | Gemini 4 Carbon |
| SimpleQA Verified | 46.5% | 77.7% | Gemini 4 Carbon |
| LMArena Text: science | 1482 | 1535 | Gemini 4 Carbon |
| LMArena Text: medicine and healthcare | 1466 | 1525 | Gemini 4 Carbon |
| LMArena Text: legal and government | 1480 | 1542 | Gemini 4 Carbon |
| LMArena Text: business and finance | 1465 | 1520 | Gemini 4 Carbon |
| FACTS Benchmark Suite | 58.7% | 78.4% | Gemini 4 Carbon |
| PRBench Legal | Not measured | 53.3% | Only Gemini 4 Carbon measured |
| Neither model has a result on the other 12 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index33
| Benchmark | Claude Sonnet 5.5 | Gemini 4 Carbon | Leader |
|---|---|---|---|
| Design Arena: websites | 1309 | 1319 | Gemini 4 Carbon |
| Design Arena: UI components | 1389 | 1357 | Claude Sonnet 5.5 |
| Design Arena: game dev | 1387 | 1326 | Claude Sonnet 5.5 |
| Design Arena: data viz | 1359 | Not measured | Only Claude Sonnet 5.5 measured |
| Design Arena: 3D | 1400 | 1434 | Gemini 4 Carbon |
| Design Arena: all code categories | 1345 | 1348 | Gemini 4 Carbon |
| Design Arena: SVG | 1299 | Not measured | Only Claude Sonnet 5.5 measured |
| LiveBench design | 81.3% | 80.2% | Claude Sonnet 5.5 |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index313
| Benchmark | Claude Sonnet 5.5 | Gemini 4 Carbon | Leader |
|---|---|---|---|
| Vals Index | 70.8% | 70.5% | Claude Sonnet 5.5 |
| LiveBench data analysis | 59.5% | 86.5% | Gemini 4 Carbon |
| Public Benefits Bench (Vals AI) | 67.2% | Not measured | Only Claude Sonnet 5.5 measured |
| SAGE (Vals AI) | 88.0% | Not measured | Only Claude Sonnet 5.5 measured |
| MedScribe (Vals AI) | 91.1% | Not measured | Only Claude Sonnet 5.5 measured |
| Long context reasoning (AA-LCR) | 82.7% | 79.3% | Claude Sonnet 5.5 |
| GDPval-AA | 1838 | 1733 | Claude Sonnet 5.5 |
| DTBench | 97.3% | Not measured | Only Claude Sonnet 5.5 measured |
| LMCA | 68.3% | Not measured | Only Claude Sonnet 5.5 measured |
| Furniture assembly | 75.0% | 77.5% | Gemini 4 Carbon |
| LMArena Text | 1469 | 1538 | Gemini 4 Carbon |
| LMArena Text: instruction following | 1487 | 1539 | Gemini 4 Carbon |
| LMArena Text: creative writing | 1466 | 1533 | Gemini 4 Carbon |
| LMArena Vision | 1284 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Text: English | 1479 | 1530 | Gemini 4 Carbon |
| LMArena Text: non-English | 1455 | 1534 | Gemini 4 Carbon |
| LMArena Text: Chinese | 1533 | 1613 | Gemini 4 Carbon |
| LMArena Text: German | 1448 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Text: Japanese | 1456 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Text: Korean | 1413 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Text: French | 1494 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Text: Spanish | 1463 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Text: Russian | 1460 | 1560 | Gemini 4 Carbon |
| LMArena Text: Polish | 1492 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Text: multi-turn | 1462 | 1569 | Gemini 4 Carbon |
| LMArena Text: longer queries | 1487 | 1549 | Gemini 4 Carbon |
| LMArena Text: writing and language | 1470 | 1542 | Gemini 4 Carbon |
| LMArena Text: entertainment and media | 1454 | 1527 | Gemini 4 Carbon |
| LMArena Vision: English | 1282 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Vision: Chinese | 1387 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Vision: reading text | 1296 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Vision: humor | 1266 | Not measured | Only Claude Sonnet 5.5 measured |
| LMArena Vision: creative writing | 1264 | Not measured | Only Claude Sonnet 5.5 measured |
| Neither model has a result on the other 16 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Sonnet 5.5 | Gemini 4 Carbon | 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 5.5$0.0070Fits in one request
Gemini 4 Carbon$0.0070Fits in one request
Same cost
Repository review
50K input + 3K output tokens
Claude Sonnet 5.5$0.130Fits in one request
Gemini 4 Carbon$0.130Fits in one request
Same cost
Agent loop
200K input + 10K output tokens
Claude Sonnet 5.5$0.500Fits in one request
Gemini 4 Carbon$0.500Fits in one request
Same cost
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 5.5 | Google Gemini 4 Carbon |
|---|---|---|
| Maker | Anthropic | |
| Weights | No public weights listed | No public weights listed |
| Released | September 28, 2026 | November 10, 2026 |
| Context window | 1M | 1.05M |
| Input $/Mtok | $2.00 | $2.00 |
| Output $/Mtok | $10.00 | $10.00 |
| Coding index | 84.3 | 86.6 |
| Independent measurements behind it | 5 | 4 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Sonnet 5.5 better than Gemini 4 Carbon for coding?
Pick Gemini 4 Carbon for coding: it scores 86.6 on the Vibe Coding Index against 84.3 for Claude Sonnet 5.5, though on 4 benchmarks against 5. Pick Claude Sonnet 5.5 for work measured by LMArena WebDev or Terminal-Bench 4 (Vals AI), where it leads.
Which is cheaper, Claude Sonnet 5.5 or Gemini 4 Carbon?
Claude Sonnet 5.5 costs $2.00 per million input tokens and $10.00 per million output tokens; Gemini 4 Carbon costs $2.00 and $10.00. Cost per solved task is known for neither of them.
Which has the bigger context window, Claude Sonnet 5.5 or Gemini 4 Carbon?
Gemini 4 Carbon, with 1.05M tokens against 1M.
How many benchmarks is the Claude Sonnet 5.5 vs Gemini 4 Carbon comparison based on?
70 benchmarks have a result for both models as of November 14, 2026. Claude Sonnet 5.5's coding index rests on 5 independent benchmarks and Gemini 4 Carbon's on 4. Where only one model has a result, the page says so instead of naming a winner.