Model comparison
Claude 3 Opus vs Grok 3
Grok 3 is the stronger model overall, scoring 39.6 to 29.3 on the Noometry Index.
Last verified . 24 shared benchmarks.
Summary
- They share 24 benchmarks with published results for both. Claude 3 Opus scores higher in 1 category and Grok 3 in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 3 leads 36.2 to 13.9.
- The biggest single-benchmark swing is MATH Level 5: 37.5% for Claude 3 Opus and 88.7% for Grok 3.
Side by side
| Claude 3 Opus | Grok 3 | |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 29.3 | 39.6 |
| Released | 2024-02-29 | 2025-04-09 |
| Weights | Proprietary | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 46 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
Claude 3 Opus: 32.4 (#261), Grok 3: 41.5 (#113)
| Benchmark | Claude 3 Opus | Grok 3 |
|---|---|---|
| WeirdML | 19.2% | 37.2% |
| LMArena Coding | 1264 | 1432 |
| Aider Polyglot | — | 53.3% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Grok 3 leads
Claude 3 Opus: 24.3 (#110), Grok 3: 30.2 (#69)
| Benchmark | Claude 3 Opus | Grok 3 |
|---|---|---|
| Cybench | 10% | — |
| BALROG | — | 29.5% |
| METR Time Horizons | 29.5% | — |
Reasoning Too close to call
Claude 3 Opus: 14.2 (#318), Grok 3: 13.8 (#322)
| Benchmark | Claude 3 Opus | Grok 3 |
|---|---|---|
| SimpleBench | 23.5% | 36.1% |
| LMArena Hard Prompts | 1246 | 1433 |
| Epoch Capabilities Index | 126.91 | 138.34 |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 61.3% |
| ARC-AGI-1 | — | 5.5% |
| Chess Puzzles | 5% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 40.6% | — |
| DTBench | 61.6% | — |
| LiveBench Data Analysis | 57.9% | — |
| LMCA | 17% | — |
| ForecastBench | 58.4 | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math Grok 3 leads
Claude 3 Opus: 13.9 (#293), Grok 3: 36.2 (#122)
| Benchmark | Claude 3 Opus | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 55.6% |
| LMArena Math | 1273 | 1390 |
| MATH Level 5 | 37.5% | 88.7% |
| Omni-MATH | — | 46.4% |
| LiveBench Math | 43.6% | — |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Grok 3 leads
Claude 3 Opus: 24.5 (#262), Grok 3: 46.0 (#76)
| Benchmark | Claude 3 Opus | Grok 3 |
|---|---|---|
| GPQA Diamond | 47.2% | 75.8% |
| Confabulations | 22.7% | 14.2% |
| LMArena Expert | 1223 | 1420 |
| SimpleQA Verified | 12.6% | — |
| MMLU-Pro | — | 78.8% |
| Vectara Hallucination Rate | — | 5.8% |
| GPQA (HELM) | — | 65% |
| MMLU | 84.6% | — |
Multimodal Not comparable
Claude 3 Opus: 26.6 (#112), Grok 3: —
| Benchmark | Claude 3 Opus | Grok 3 |
|---|---|---|
| LMArena Vision | 1023 | — |
Multilingual Grok 3 leads
Claude 3 Opus: 41.6 (#202), Grok 3: 52.5 (#81)
| Benchmark | Claude 3 Opus | Grok 3 |
|---|---|---|
| LMArena Non-English | 1258 | 1411 |
| LMArena Chinese | 1248 | 1448 |
| LMArena French | 1274 | 1460 |
| LMArena German | 1258 | 1431 |
| LMArena Japanese | 1204 | 1389 |
| LMArena Korean | 1187 | 1374 |
| LMArena Russian | 1280 | 1416 |
| LMArena Spanish | 1246 | 1417 |
Instruction Following Grok 3 leads
Claude 3 Opus: 64.3 (#223), Grok 3: 75.2 (#69)
| Benchmark | Claude 3 Opus | Grok 3 |
|---|---|---|
| LMArena Instruction Following | 1248 | 1409 |
| LiveBench Instruction Following | 63.9% | — |
| IFEval | — | 88.4% |
Long Context Too close to call
Claude 3 Opus: 38.2 (#197), Grok 3: 38.7 (#187)
| Benchmark | Claude 3 Opus | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1259 | 1439 |
| Fiction.LiveBench | — | 58.3% |
Writing & Preference Grok 3 leads
Claude 3 Opus: 47.4 (#208), Grok 3: 56.0 (#136)
| Benchmark | Claude 3 Opus | Grok 3 |
|---|---|---|
| LMArena Text | 1262 | 1426 |
| LMArena Creative Writing | 1235 | 1414 |
| LMArena Multi-Turn | 1275 | 1424 |
| Short-Story Creative Writing | — | 76.4% |
| EQ-Bench Creative Writing | — | 1186 |
| WildBench | — | 84.9% |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than Grok 3?
Grok 3 is the stronger model overall, scoring 39.6 to 29.3 on the Noometry Index.
Is Claude 3 Opus or Grok 3 better for coding?
Grok 3 scores higher on coding benchmarks: 41.5 versus 32.4 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and Grok 3 share?
24 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and Grok 3 has 40.