Model comparison
GPT-4o vs Grok 3
Grok 3 is the stronger model overall, scoring 39.9 to 28.6 on the Noometry Index.
Last verified . 37 shared benchmarks.
Summary
- They share 37 benchmarks with published results for both. GPT-4o scores higher in 1 category and Grok 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 3 leads 38.0 to 10.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.4% for GPT-4o and 55.6% for Grok 3.
Side by side
| GPT-4o | Grok 3 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 28.6 | 39.9 |
| Released | 2024-05-13 | 2025-04-09 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | — |
| Max output | 16K | — |
| Input $ / M tokens | $2.50 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 72 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
GPT-4o: 24.8 (#328), Grok 3: 41.9 (#115)
| Benchmark | GPT-4o | Grok 3 |
|---|---|---|
| Aider Polyglot | 45.3% | 53.3% |
| WeirdML | 25.1% | 37.2% |
| LMArena Coding | 1297 | 1432 |
| SWE-bench Verified | 31% | — |
| SWE-bench Verified (bash only) | 21.6% | — |
| GSO | 0% | — |
| BigCodeBench Instruct | 51.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 61.1% | — |
| CadEval | 26% | — |
| HumanEval+ | 87.2% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Grok 3 leads
GPT-4o: 21.0 (#141), Grok 3: 30.5 (#76)
| Benchmark | GPT-4o | Grok 3 |
|---|---|---|
| BALROG | 32.3% | 29.5% |
| GDPval | 9.9% | — |
| TheAgentCompany | 8.6% | — |
| Cybench | 12.5% | — |
| LMArena Search | 1006 | — |
| METR Time Horizons | 40.8% | — |
Reasoning Grok 3 leads
GPT-4o: 9.4 (#343), Grok 3: 13.7 (#333)
| Benchmark | GPT-4o | Grok 3 |
|---|---|---|
| ARC-AGI-2 | 0% | 0% |
| SimpleBench | 17.8% | 36.1% |
| ARC-AGI-1 | 4.5% | 5.5% |
| LMArena Hard Prompts | 1281 | 1434 |
| Epoch Capabilities Index | 128.97 | 138.33 |
| Kagi LLM Benchmark | — | 61.3% |
| CritPt | 0% | — |
| Chess Puzzles | 13% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 55.8% | — |
| DTBench | 64.5% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 16.6% | — |
| ForecastBench | 57.7 | — |
| LiveBench | 55.3% | — |
Math Grok 3 leads
GPT-4o: 10.6 (#312), Grok 3: 38.0 (#145)
| Benchmark | GPT-4o | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.4% | 55.6% |
| Omni-MATH | 29.3% | 46.4% |
| LMArena Math | 1285 | 1391 |
| MATH Level 5 | 53.3% | 88.7% |
| FrontierMath (Feb 2025 set) | 0.3% | 3.8% |
| FrontierMath (Tiers 1-3) | 0.4% | — |
| LiveBench Math | 49.5% | — |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Grok 3 leads
GPT-4o: 28.8 (#242), Grok 3: 46.2 (#82)
| Benchmark | GPT-4o | Grok 3 |
|---|---|---|
| GPQA Diamond | 49.2% | 75.8% |
| MMLU-Pro | 71.3% | 78.8% |
| Confabulations | 15.3% | 14.2% |
| Vectara Hallucination Rate | 9.6% | 5.8% |
| GPQA (HELM) | 52% | 65% |
| LMArena Expert | 1250 | 1421 |
| Humanity's Last Exam | 2.7% | — |
| SimpleQA Verified | 26% | — |
| MMLU | 88.1% | — |
Multimodal Not comparable
GPT-4o: 34.5 (#91), Grok 3: —
| Benchmark | GPT-4o | Grok 3 |
|---|---|---|
| LMArena Vision | 1137 | — |
| Video-MME | 71.9% | — |
| GeoBench | 71% | — |
| VPCT | 40% | — |
| ScienceQA | 88.5% | — |
Multilingual Grok 3 leads
GPT-4o: 43.2 (#186), Grok 3: 52.3 (#87)
| Benchmark | GPT-4o | Grok 3 |
|---|---|---|
| LMArena Non-English | 1283 | 1410 |
| LMArena Chinese | 1277 | 1448 |
| LMArena French | 1304 | 1460 |
| LMArena German | 1282 | 1431 |
| LMArena Japanese | 1257 | 1387 |
| LMArena Korean | 1234 | 1373 |
| LMArena Russian | 1286 | 1416 |
| LMArena Spanish | 1292 | 1417 |
Instruction Following Grok 3 leads
GPT-4o: 66.6 (#207), Grok 3: 75.0 (#73)
| Benchmark | GPT-4o | Grok 3 |
|---|---|---|
| IFEval | 81.7% | 88.4% |
| LMArena Instruction Following | 1278 | 1409 |
| LiveBench Instruction Following | 68.6% | — |
Long Context Too close to call
GPT-4o: 39.4 (#179), Grok 3: 38.7 (#192)
| Benchmark | GPT-4o | Grok 3 |
|---|---|---|
| Fiction.LiveBench | 66.7% | 58.3% |
| LMArena Longer Query | 1289 | 1439 |
Writing & Preference Grok 3 leads
GPT-4o: 52.6 (#166), Grok 3: 55.8 (#141)
| Benchmark | GPT-4o | Grok 3 |
|---|---|---|
| LMArena Text | 1300 | 1426 |
| LMArena Creative Writing | 1292 | 1414 |
| Short-Story Creative Writing | 81.8% | 76.4% |
| WildBench | 82.8% | 84.9% |
| LMArena Multi-Turn | 1302 | 1425 |
| EQ-Bench Creative Writing | — | 1186 |
| LiveBench Language | 47.6% | — |
Frequently asked questions
Is GPT-4o better than Grok 3?
Grok 3 is the stronger model overall, scoring 39.9 to 28.6 on the Noometry Index.
Is GPT-4o or Grok 3 better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 24.8 in the Noometry coding category.
How many benchmarks do GPT-4o and Grok 3 share?
37 benchmarks have published results for both models. GPT-4o has 72 scored results on Noometry and Grok 3 has 40.