Claude Sonnet 4.5 vs Grok 4.3
Claude Sonnet 4.5 leads Grok 4.3 on coding, 39.5 to 37.4 on the Vibe Coding Index, and Grok 4.3 costs 6.0x less per output token ($2.50 vs $15.00 per million). Updated October 22, 2026; 70 benchmarks have a result for both.
Claude Sonnet 4.5 or Grok 4.3: which to pick
Pick Claude Sonnet 4.5 for coding: it scores 39.5 on the Vibe Coding Index against 37.4 for Grok 4.3.
Pick Grok 4.3 for output tokens 6.0x cheaper ($2.50 vs $15.00 per million) and a lower cost per solved task ($0.31 against $1.16).
- Shared results
- 70
- Claude Sonnet 4.5 only
- 25
- Grok 4.3 only
- 35
- 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 Sonnet 4.5
Claude Sonnet 4.5 scores 39.5 against 37.4.
3 benchmarks behind each score
Intelligence
AA Intelligence Index
Grok 4.3
Grok 4.3 leads, 24.9 against 20.7.
Same board, both measured
Agentic work
AA Agentic Index
Claude Sonnet 4.5
Claude Sonnet 4.5 leads, 15.8 against 15.5.
Same board, both measured
Reasoning
GPQA Diamond
Grok 4.3
Grok 4.3 leads, 91.4% against 81.6%.
Same board, both measured
Math
AIME
No call
Only Claude Sonnet 4.5 has a AIME result, 88.2%.
Directional only
Knowledge
MMLU Pro
Claude Sonnet 4.5
Claude Sonnet 4.5 leads, 87.4% against 85.8%.
Same board, both measured
Web design
Design Arena: websites
Grok 4.3
Grok 4.3 leads, 1196 against 1189.
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
Grok 4.3
Claude Sonnet 4.5 $0.011, Grok 4.3 $0.0025 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Grok 4.3
Claude Sonnet 4.5 $1.16, Grok 4.3 $0.31.
Published attempt cost and pass rate
At a glance
- Coding index
- Claude Sonnet 4.5 39.5 from 3 benchmarks, Grok 4.3 37.4 from 3 benchmarks.
- Widest category
- Math, 12.7 points apart on average over 6 shared benchmarks.
- Widest single gap
- Vending-Bench 2: Claude Sonnet 4.5 $3,839 against Grok 4.3 $35, 23.8 points apart on a 0 to 100 scale.
- Output price
- Claude Sonnet 4.5 $15.00 against Grok 4.3 $2.50 per million output tokens, 6.0x 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 Sonnet 4.5 | Bars | Grok 4.3 | Ties |
|---|---|---|---|---|
| Coding | 7 | 3 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 4 | 1 | ||
| Reasoning | 3 | 4 | ||
| Math | 3 | 3 | ||
| Knowledge | 8 | 4 | ||
| Web design | 3 | 5 | ||
| General | 20 | 1 | ||
| All shared boards | 48 | 22 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index73
| Benchmark | Claude Sonnet 4.5 | Grok 4.3 | Leader |
|---|---|---|---|
| SWE-bench Verified | 74.8% | Not measured | Only Claude Sonnet 4.5 measured |
| SWE-bench Multilingual | 67.0% | Not measured | Only Claude Sonnet 4.5 measured |
| LMArena WebDev | 1394 | 1358 | Claude Sonnet 4.5 |
| AA Coding Index | 52.1 | 42.2 | Claude Sonnet 4.5 |
| LiveCodeBench (Vals AI) | 73.0% | 84.5% | Grok 4.3 |
| SWE-bench Verified (Vals AI) | 70.0% | 71.4% | Grok 4.3 |
| Vibe Code Bench (Vals AI) | 22.6% | 19.4% | Claude Sonnet 4.5 |
| Code migration (Vals AI) | Not measured | 6.8% | Only Grok 4.3 measured |
| ProgramBench (Vals AI) | Not measured | 0.0% | Only Grok 4.3 measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 42.0% | Only Grok 4.3 measured |
| SkillsBench (Vals AI) | Not measured | 40.6% | Only Grok 4.3 measured |
| SciCode | Not measured | 48.3% | Only Grok 4.3 measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 37.9% | Only Grok 4.3 measured |
| Terminal-Bench | 46.5% | Not measured | Only Claude Sonnet 4.5 measured |
| GSO-Bench | 14.7% | Not measured | Only Claude Sonnet 4.5 measured |
| WeirdML | 47.7% | 49.9% | Grok 4.3 |
| Cybench | 60.0% | Not measured | Only Claude Sonnet 4.5 measured |
| ALE-Bench | Not measured | 944.2 | Only Grok 4.3 measured |
| LMArena Text: coding | 1488 | 1415 | Claude Sonnet 4.5 |
| LMArena Text: software and IT | 1473 | 1416 | Claude Sonnet 4.5 |
| LMArena WebDev: React | 1386 | 1353 | Claude Sonnet 4.5 |
| LMArena WebDev: HTML | 1399 | 1367 | Claude Sonnet 4.5 |
| LMArena WebDev: from an image | Not measured | 1375 | Only Grok 4.3 measured |
| Neither model has a result on the other 16 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Claude Sonnet 4.5 | Grok 4.3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 20.7 | 24.9 | Grok 4.3 |
| Epoch Capabilities Index | 146.7 | Not measured | Only Claude Sonnet 4.5 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index41
| Benchmark | Claude Sonnet 4.5 | Grok 4.3 | Leader |
|---|---|---|---|
| AA Agentic Index | 15.8 | 15.5 | Claude Sonnet 4.5 |
| Design Arena: agents building web apps | 1060 | 1126 | Grok 4.3 |
| Design Arena: agents building full-stack apps | 1050 | 1006 | Claude Sonnet 4.5 |
| Design Arena: agents building mobile apps | 1137 | 1058 | Claude Sonnet 4.5 |
| Design Arena: agents building Android apps | Not measured | 962 | Only Grok 4.3 measured |
| Design Arena: agents building Godot games | Not measured | 914 | Only Grok 4.3 measured |
| Design Arena: agents building games | Not measured | 1007 | Only Grok 4.3 measured |
| Design Arena: HTML slides | Not measured | 959 | Only Grok 4.3 measured |
| Design Arena: PowerPoint slides | Not measured | 988 | Only Grok 4.3 measured |
| Design Arena: spreadsheets | Not measured | 879 | Only Grok 4.3 measured |
| Finance Agent (Vals AI) | Not measured | 37.7% | Only Grok 4.3 measured |
| Tax Agent Bench (Vals AI) | Not measured | 6.2% | Only Grok 4.3 measured |
| Legal Research Bench (Vals AI) | Not measured | 15.4% | Only Grok 4.3 measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 18.4% | Only Grok 4.3 measured |
| τ²-Bench Telecom | Not measured | 97.7% | Only Grok 4.3 measured |
| METR time horizons | 67.4% | Not measured | Only Claude Sonnet 4.5 measured |
| Remote Labor Index | 2.1% | Not measured | Only Claude Sonnet 4.5 measured |
| GDPval | 42.5% | Not measured | Only Claude Sonnet 4.5 measured |
| DeepResearch Bench | 52.6% | Not measured | Only Claude Sonnet 4.5 measured |
| OSWorld | 62.9% | Not measured | Only Claude Sonnet 4.5 measured |
| Vending-Bench 2 | $3,839 | $35 | Claude Sonnet 4.5 |
| LMArena Search | Not measured | 1166 | Only Grok 4.3 measured |
| Berkeley Function Calling Leaderboard (BFCL) | 73.2% | Not measured | Only Claude Sonnet 4.5 measured |
| Neither model has a result on the other 18 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index34
| Benchmark | Claude Sonnet 4.5 | Grok 4.3 | Leader |
|---|---|---|---|
| GPQA Diamond | 81.6% | 91.4% | Grok 4.3 |
| MMMU Pro (Vals AI) | 79.3% | 83.1% | Grok 4.3 |
| GPQA Diamond (Artificial Analysis run) | Not measured | 90.1% | Only Grok 4.3 measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 37.2% | Only Grok 4.3 measured |
| GPQA Diamond (Epoch AI run) | 82.3% | 88.8% | Grok 4.3 |
| ARC-AGI | 63.7% | Not measured | Only Claude Sonnet 4.5 measured |
| ARC-AGI-2 | 13.6% | Not measured | Only Claude Sonnet 4.5 measured |
| SimpleBench | 54.3% | Not measured | Only Claude Sonnet 4.5 measured |
| Chess puzzles | 10.0% | 18.0% | Grok 4.3 |
| Mystery game puzzles | 17.0% | Not measured | Only Claude Sonnet 4.5 measured |
| EBR-bench | 2.4% | Not measured | Only Claude Sonnet 4.5 measured |
| VPCT | 39.8% | Not measured | Only Claude Sonnet 4.5 measured |
| LMArena Text: hard prompts | 1462 | 1396 | Claude Sonnet 4.5 |
| LMArena Text: expert | 1483 | 1384 | Claude Sonnet 4.5 |
| LMArena Text: hard prompts in English | 1471 | 1402 | Claude Sonnet 4.5 |
| LMArena Vision: diagrams | Not measured | 1241 | Only Grok 4.3 measured |
| LMArena Vision: homework | Not measured | 1241 | Only Grok 4.3 measured |
| Kagi LLM Benchmark | 57.9% | Not measured | Only Claude Sonnet 4.5 measured |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index33
| Benchmark | Claude Sonnet 4.5 | Grok 4.3 | Leader |
|---|---|---|---|
| AIME | 88.2% | Not measured | Only Claude Sonnet 4.5 measured |
| MGSM (Vals AI) | 94.3% | Not measured | Only Claude Sonnet 4.5 measured |
| OTIS Mock AIME | 77.8% | 93.3% | Grok 4.3 |
| FrontierMath, tiers 1 to 3 | 23.9% | 42.8% | Grok 4.3 |
| FrontierMath, tier 4 | 2.4% | 14.6% | Grok 4.3 |
| ProofBench | 19.0% | 11.0% | Claude Sonnet 4.5 |
| MATH level 5 | 97.7% | Not measured | Only Claude Sonnet 4.5 measured |
| LMArena Text: math | 1449 | 1387 | Claude Sonnet 4.5 |
| LMArena Text: mathematical work | 1462 | 1396 | Claude Sonnet 4.5 |
| Neither model has a result on the other 4 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index84
| Benchmark | Claude Sonnet 4.5 | Grok 4.3 | Leader |
|---|---|---|---|
| MMLU Pro | 87.4% | 85.8% | Claude Sonnet 4.5 |
| LegalBench (Vals AI) | 84.1% | 84.5% | Grok 4.3 |
| CaseLaw (Vals AI) | 62.2% | 79.3% | Grok 4.3 |
| CorpFin (Vals AI) | 62.0% | 68.5% | Grok 4.3 |
| TaxEval (Vals AI) | 73.3% | 70.8% | Claude Sonnet 4.5 |
| MortgageTax (Vals AI) | 64.0% | 48.3% | Claude Sonnet 4.5 |
| MedQA (Vals AI) | 94.7% | Not measured | Only Claude Sonnet 4.5 measured |
| MedCode (Vals AI) | 44.1% | 38.1% | Claude Sonnet 4.5 |
| SimpleQA Verified | 30.7% | 33.2% | Grok 4.3 |
| LMArena Text: science | 1448 | 1399 | Claude Sonnet 4.5 |
| LMArena Text: medicine and healthcare | 1444 | 1384 | Claude Sonnet 4.5 |
| LMArena Text: legal and government | 1451 | 1386 | Claude Sonnet 4.5 |
| LMArena Text: business and finance | 1440 | 1381 | Claude Sonnet 4.5 |
| FACTS Benchmark Suite | 40.6% | Not measured | Only Claude Sonnet 4.5 measured |
| Neither model has a result on the other 4 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index35
| Benchmark | Claude Sonnet 4.5 | Grok 4.3 | Leader |
|---|---|---|---|
| Design Arena: websites | 1189 | 1196 | Grok 4.3 |
| Design Arena: UI components | 1170 | 1190 | Grok 4.3 |
| Design Arena: game dev | 1166 | 1173 | Grok 4.3 |
| Design Arena: data viz | 1167 | 1179 | Grok 4.3 |
| Design Arena: 3D | 1157 | 1124 | Claude Sonnet 4.5 |
| Design Arena: all code categories | 1182 | 1186 | Grok 4.3 |
| Design Arena: SVG | 1115 | 1080 | Claude Sonnet 4.5 |
| Design Arena: ASCII art | 1205 | 1125 | Claude Sonnet 4.5 |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index201
| Benchmark | Claude Sonnet 4.5 | Grok 4.3 | Leader |
|---|---|---|---|
| Public Benefits Bench (Vals AI) | Not measured | 51.7% | Only Grok 4.3 measured |
| SAGE (Vals AI) | 36.1% | 19.7% | Claude Sonnet 4.5 |
| MedScribe (Vals AI) | 84.5% | 74.4% | Claude Sonnet 4.5 |
| Vals Multimodal Index | Not measured | 43.3% | Only Grok 4.3 measured |
| IFBench | Not measured | 83.3% | Only Grok 4.3 measured |
| Long context reasoning (AA-LCR) | Not measured | 75.0% | Only Grok 4.3 measured |
| DTBench | 83.2% | 90.7% | Grok 4.3 |
| LMCA | 38.8% | 38.3% | Claude Sonnet 4.5 |
| LMArena Text | 1438 | 1397 | Claude Sonnet 4.5 |
| LMArena Text: instruction following | 1459 | 1366 | Claude Sonnet 4.5 |
| LMArena Text: creative writing | 1442 | 1380 | Claude Sonnet 4.5 |
| LMArena Vision | Not measured | 1229 | Only Grok 4.3 measured |
| LMArena Text: English | 1450 | 1404 | Claude Sonnet 4.5 |
| LMArena Text: non-English | 1425 | 1385 | Claude Sonnet 4.5 |
| LMArena Text: Chinese | 1460 | 1422 | Claude Sonnet 4.5 |
| LMArena Text: German | 1427 | 1395 | Claude Sonnet 4.5 |
| LMArena Text: Japanese | 1390 | 1379 | Claude Sonnet 4.5 |
| LMArena Text: Korean | 1402 | 1354 | Claude Sonnet 4.5 |
| LMArena Text: French | 1459 | 1411 | Claude Sonnet 4.5 |
| LMArena Text: Spanish | 1458 | 1399 | Claude Sonnet 4.5 |
| LMArena Text: Russian | 1437 | 1399 | Claude Sonnet 4.5 |
| LMArena Text: Polish | 1423 | 1398 | Claude Sonnet 4.5 |
| LMArena Text: multi-turn | 1465 | 1406 | Claude Sonnet 4.5 |
| LMArena Text: longer queries | 1475 | 1393 | Claude Sonnet 4.5 |
| LMArena Text: writing and language | 1442 | 1385 | Claude Sonnet 4.5 |
| LMArena Text: entertainment and media | 1428 | 1370 | Claude Sonnet 4.5 |
| LMArena Vision: English | Not measured | 1235 | Only Grok 4.3 measured |
| LMArena Vision: Chinese | Not measured | 1257 | Only Grok 4.3 measured |
| LMArena Vision: reading text | Not measured | 1235 | Only Grok 4.3 measured |
| LMArena Vision: humor | Not measured | 1222 | Only Grok 4.3 measured |
| LMArena Vision: recognizing things | Not measured | 1217 | Only Grok 4.3 measured |
| LMArena Vision: creative writing | Not measured | 1237 | Only Grok 4.3 measured |
| LMArena Document | 1450 | Not measured | Only Claude Sonnet 4.5 measured |
| Neither model has a result on the other 13 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Sonnet 4.5 | Grok 4.3 | 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
Grok 4.3 costs less
Repository review
50K input + 3K output tokens
Grok 4.3 costs less
Agent loop
200K input + 10K output tokens
Grok 4.3 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 4.5 | SpaceXAI Grok 4.3 |
|---|---|---|
| Maker | Anthropic | xAI |
| Weights | No public weights listed | No public weights listed |
| Released | September 29, 2025 | April 30, 2026 |
| Context window | 1M | 1M |
| Input $/Mtok | $3.00 | $1.25 |
| Output $/Mtok | $15.00 | $2.50 |
| Coding index | 39.5 | 37.4 |
| Independent measurements behind it | 3 | 3 |
| Cost per solved task | $1.16 | $0.31 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Sonnet 4.5 better than Grok 4.3 for coding?
Pick Claude Sonnet 4.5 for coding: it scores 39.5 on the Vibe Coding Index against 37.4 for Grok 4.3. Pick Grok 4.3 for output tokens 6.0x cheaper ($2.50 vs $15.00 per million) and a lower cost per solved task ($0.31 against $1.16).
Which is cheaper, Claude Sonnet 4.5 or Grok 4.3?
Claude Sonnet 4.5 costs $3.00 per million input tokens and $15.00 per million output tokens; Grok 4.3 costs $1.25 and $2.50. Per solved SWE-bench Verified task, Claude Sonnet 4.5 costs $1.16 and Grok 4.3 $0.31.
Which has the bigger context window, Claude Sonnet 4.5 or Grok 4.3?
Neither: both take 1M tokens.
How many benchmarks is the Claude Sonnet 4.5 vs Grok 4.3 comparison based on?
70 benchmarks have a result for both models as of October 22, 2026. Claude Sonnet 4.5's coding index rests on 3 independent benchmarks and Grok 4.3's on 3. Where only one model has a result, the page says so instead of naming a winner.