Model comparison
Claude Sonnet 4 vs Grok 4.3
Grok 4.3 is the stronger model overall, scoring 43.6 to 40.6 on the Noometry Index.
Last verified . 28 shared benchmarks.
Summary
- They share 28 benchmarks with published results for both. Claude Sonnet 4 scores higher in 2 categories and Grok 4.3 in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.3 leads 34.9 to 23.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for Claude Sonnet 4 and 93.3% for Grok 4.3.
- Grok 4.3 is cheaper at $1.25 / $2.50 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Grok 4.3 accepts more context: 1M tokens versus 200K.
Side by side
| Claude Sonnet 4 | Grok 4.3 | |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 40.6 | 43.6 |
| Released | 2025-05-22 | 2026-04-17 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1M |
| Max output | 64K | 30K |
| Input $ / M tokens | $3 | $1.25 |
| Output $ / M tokens | $15 | $2.50 |
| Results tracked | 58 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude Sonnet 4: 44.0 (#70), Grok 4.3: 41.3 (#117)
| Benchmark | Claude Sonnet 4 | Grok 4.3 |
|---|---|---|
| SciCode | 40% | 47.3% |
| WeirdML | 46.1% | 49.9% |
| LMArena Coding | 1414 | 1415 |
| ALE-Bench | 655.35 | 944.17 |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1357 |
| GSO | 4.9% | — |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.3 (#29), Grok 4.3: 27.6 (#93)
| Benchmark | Claude Sonnet 4 | Grok 4.3 |
|---|---|---|
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| GDP.pdf | — | 8% |
| LMArena Search | — | 1165 |
| METR Time Horizons | 62% | — |
| Vending-Bench 2 | — | 35.26 |
Reasoning Grok 4.3 leads
Claude Sonnet 4: 23.0 (#169), Grok 4.3: 34.9 (#60)
| Benchmark | Claude Sonnet 4 | Grok 4.3 |
|---|---|---|
| CritPt | 0.3% | 8% |
| LMArena Hard Prompts | 1372 | 1397 |
| DTBench | 77.1% | 90.7% |
| LMCA | 29% | 38.3% |
| Epoch Capabilities Index | 141.69 | 149.13 |
| ForecastBench | 60.2 | 60.3 |
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| Kagi LLM Benchmark | 73% | — |
| NYT Connections (extended) | — | 55.2% |
| ARC-AGI-1 | 40% | — |
| Chess Puzzles | — | 25% |
| EnigmaEval | 3.1% | — |
Math Grok 4.3 leads
Claude Sonnet 4: 41.5 (#71), Grok 4.3: 45.8 (#60)
| Benchmark | Claude Sonnet 4 | Grok 4.3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 93.3% |
| LMArena Math | 1375 | 1388 |
| FrontierMath (Tiers 1-3) | — | 42.8% |
| FrontierMath Tier 4 | — | 14.6% |
| ProofBench | — | 11% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Grok 4.3 leads
Claude Sonnet 4: 41.6 (#99), Grok 4.3: 52.5 (#57)
| Benchmark | Claude Sonnet 4 | Grok 4.3 |
|---|---|---|
| GPQA Diamond | 79.2% | 88.8% |
| LMArena Expert | 1371 | 1385 |
| Humanity's Last Exam | 7.8% | — |
| SimpleQA Verified | — | 33.2% |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| Vectara Hallucination Rate | 10.3% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal Grok 4.3 leads
Claude Sonnet 4: 25.6 (#116), Grok 4.3: 31.6 (#96)
| Benchmark | Claude Sonnet 4 | Grok 4.3 |
|---|---|---|
| LMArena Vision | 1191 | 1231 |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| Blueprint-Bench 2 | — | 0% |
| MindCube | 44.8% | — |
Multilingual Grok 4.3 leads
Claude Sonnet 4: 46.9 (#151), Grok 4.3: 50.7 (#114)
| Benchmark | Claude Sonnet 4 | Grok 4.3 |
|---|---|---|
| LMArena Non-English | 1333 | 1385 |
| LMArena Chinese | 1351 | 1423 |
| LMArena French | 1364 | 1412 |
| LMArena German | 1331 | 1395 |
| LMArena Japanese | 1301 | 1379 |
| LMArena Korean | 1293 | 1357 |
| LMArena Russian | 1355 | 1399 |
| LMArena Spanish | 1356 | 1402 |
Instruction Following Too close to call
Claude Sonnet 4: 71.8 (#141), Grok 4.3: 72.3 (#135)
| Benchmark | Claude Sonnet 4 | Grok 4.3 |
|---|---|---|
| LMArena Instruction Following | 1375 | 1366 |
| IFEval | 84% | — |
Long Context Grok 4.3 leads
Claude Sonnet 4: 33.6 (#254), Grok 4.3: 42.5 (#118)
| Benchmark | Claude Sonnet 4 | Grok 4.3 |
|---|---|---|
| LMArena Longer Query | 1398 | 1393 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference Grok 4.3 leads
Claude Sonnet 4: 57.3 (#127), Grok 4.3: 58.6 (#112)
| Benchmark | Claude Sonnet 4 | Grok 4.3 |
|---|---|---|
| LMArena Text | 1351 | 1397 |
| LMArena Creative Writing | 1345 | 1380 |
| LMArena Multi-Turn | 1376 | 1406 |
| Short-Story Creative Writing | 81.4% | — |
| EQ-Bench Creative Writing | 1483 | — |
| WildBench | 83.8% | — |
| EQ-Bench 4 | — | 1075 |
Frequently asked questions
Is Claude Sonnet 4 better than Grok 4.3?
Grok 4.3 is the stronger model overall, scoring 43.6 to 40.6 on the Noometry Index.
Which is cheaper, Claude Sonnet 4 or Grok 4.3?
Grok 4.3 is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Grok 4.3 better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 44.0 versus 41.3 in the Noometry coding category.
Which has the bigger context window?
Grok 4.3 does, with 1M tokens against 200K.
How many benchmarks do Claude Sonnet 4 and Grok 4.3 share?
28 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Grok 4.3 has 40.