Model comparison
GPT-4o mini vs Grok 3
Grok 3 is the stronger model overall, scoring 39.6 to 25.4 on the Noometry Index.
Last verified . 35 shared benchmarks.
Summary
- They share 35 benchmarks with published results for both. GPT-4o mini scores higher in 1 category and Grok 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok 3 leads 46.0 to 17.7.
- The biggest single-benchmark swing is Aider Polyglot: 3.6% for GPT-4o mini and 53.3% for Grok 3.
Side by side
| GPT-4o mini | Grok 3 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 25.4 | 39.6 |
| Released | 2024-07-18 | 2025-04-09 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | — |
| Max output | 16K | — |
| Input $ / M tokens | $0.15 | — |
| Output $ / M tokens | $0.60 | — |
| Results tracked | 60 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
GPT-4o mini: 21.6 (#330), Grok 3: 41.5 (#113)
| Benchmark | GPT-4o mini | Grok 3 |
|---|---|---|
| Aider Polyglot | 3.6% | 53.3% |
| WeirdML | 11.8% | 37.2% |
| LMArena Coding | 1290 | 1432 |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 43.1% | — |
| BigCodeBench Complete | 57.4% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Grok 3 leads
GPT-4o mini: 27.2 (#95), Grok 3: 30.2 (#69)
| Benchmark | GPT-4o mini | Grok 3 |
|---|---|---|
| BALROG | 17.4% | 29.5% |
Reasoning Grok 3 leads
GPT-4o mini: 8.7 (#343), Grok 3: 13.8 (#322)
| Benchmark | GPT-4o mini | Grok 3 |
|---|---|---|
| ARC-AGI-2 | 0% | 0% |
| SimpleBench | 10.7% | 36.1% |
| Kagi LLM Benchmark | 28.8% | 61.3% |
| LMArena Hard Prompts | 1267 | 1433 |
| Epoch Capabilities Index | 126.56 | 138.34 |
| ARC-AGI-1 | — | 5.5% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 32.8% | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 54.4% | — |
| LiveBench Data Analysis | 50% | — |
| LMCA | 10.4% | — |
| LiveBench | 41.3% | — |
| PIQA | 88.7% | — |
Math Grok 3 leads
GPT-4o mini: 9.9 (#307), Grok 3: 36.2 (#122)
| Benchmark | GPT-4o mini | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.9% | 55.6% |
| Omni-MATH | 28% | 46.4% |
| LMArena Math | 1267 | 1390 |
| MATH Level 5 | 52.6% | 88.7% |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| LiveBench Math | 36.3% | — |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 0% |
| GSM8K | 91.3% | — |
Knowledge Grok 3 leads
GPT-4o mini: 17.7 (#279), Grok 3: 46.0 (#76)
| Benchmark | GPT-4o mini | Grok 3 |
|---|---|---|
| GPQA Diamond | 37.7% | 75.8% |
| MMLU-Pro | 60.3% | 78.8% |
| Confabulations | 37.2% | 14.2% |
| GPQA (HELM) | 36.8% | 65% |
| LMArena Expert | 1235 | 1420 |
| SimpleQA Verified | 8.3% | — |
| Vectara Hallucination Rate | — | 5.8% |
| BoolQ | 88.7% | — |
| MMLU | 81.8% | — |
Multimodal Not comparable
GPT-4o mini: 25.3 (#117), Grok 3: —
| Benchmark | GPT-4o mini | Grok 3 |
|---|---|---|
| LMArena Vision | 1066 | — |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| VPCT | 34% | — |
Multilingual Grok 3 leads
GPT-4o mini: 42.2 (#194), Grok 3: 52.5 (#81)
| Benchmark | GPT-4o mini | Grok 3 |
|---|---|---|
| LMArena Non-English | 1266 | 1411 |
| LMArena Chinese | 1264 | 1448 |
| LMArena French | 1296 | 1460 |
| LMArena German | 1272 | 1431 |
| LMArena Japanese | 1216 | 1389 |
| LMArena Korean | 1195 | 1374 |
| LMArena Russian | 1275 | 1416 |
| LMArena Spanish | 1276 | 1417 |
Instruction Following Grok 3 leads
GPT-4o mini: 62.1 (#234), Grok 3: 75.2 (#69)
| Benchmark | GPT-4o mini | Grok 3 |
|---|---|---|
| IFEval | 78.2% | 88.4% |
| LMArena Instruction Following | 1259 | 1409 |
| LiveBench Instruction Following | 56.8% | — |
Long Context Too close to call
GPT-4o mini: 39.1 (#181), Grok 3: 38.7 (#187)
| Benchmark | GPT-4o mini | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1289 | 1439 |
| Fiction.LiveBench | — | 58.3% |
Writing & Preference Grok 3 leads
GPT-4o mini: 39.7 (#243), Grok 3: 56.0 (#136)
| Benchmark | GPT-4o mini | Grok 3 |
|---|---|---|
| LMArena Text | 1286 | 1426 |
| LMArena Creative Writing | 1268 | 1414 |
| Short-Story Creative Writing | 67.2% | 76.4% |
| EQ-Bench Creative Writing | 873 | 1186 |
| WildBench | 79.1% | 84.9% |
| LMArena Multi-Turn | 1285 | 1424 |
| LiveBench Language | 28.6% | — |
Frequently asked questions
Is GPT-4o mini better than Grok 3?
Grok 3 is the stronger model overall, scoring 39.6 to 25.4 on the Noometry Index.
Is GPT-4o mini or Grok 3 better for coding?
Grok 3 scores higher on coding benchmarks: 41.5 versus 21.6 in the Noometry coding category.
How many benchmarks do GPT-4o mini and Grok 3 share?
35 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and Grok 3 has 40.