Model comparison
Grok-3 mini vs Llama 4 Maverick
Grok-3 mini is the stronger model overall, scoring 41.2 to 30.9 on the Noometry Index.
Last verified . 35 shared benchmarks.
MetaSummary
- They share 35 benchmarks with published results for both. Grok-3 mini scores higher in 8 categories and Llama 4 Maverick in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok-3 mini leads 42.1 to 26.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 77.8% for Grok-3 mini and 20.6% for Llama 4 Maverick.
- Llama 4 Maverick has downloadable open weights; the other is API-only.
Side by side
| Grok-3 mini | Llama 4 Maverick | |
|---|---|---|
| Provider | xAI | Meta |
| Noometry Index | 41.2 | 30.9 |
| Released | 2025-04-09 | 2025-04-05 |
| Weights | Proprietary | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.19 |
| Output $ / M tokens | — | $0.65 |
| Results tracked | 35 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok-3 mini leads
Grok-3 mini: 40.8 (#131), Llama 4 Maverick: 26.6 (#324)
| Benchmark | Grok-3 mini | Llama 4 Maverick |
|---|---|---|
| Aider Polyglot | 49.3% | 15.6% |
| WeirdML | 42.6% | 24.5% |
| LMArena Coding | 1379 | 1302 |
| SWE-bench Verified (bash only) | — | 21% |
| SciCode | — | 33.1% |
| BigCodeBench Instruct | — | 49.7% |
| BigCodeBench Complete | — | 61.4% |
| ALE-Bench | — | 172.97 |
Agentic & Tool Use Not comparable
Grok-3 mini: —, Llama 4 Maverick: 28.2 (#91)
| Benchmark | Grok-3 mini | Llama 4 Maverick |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.3% |
Reasoning Grok-3 mini leads
Grok-3 mini: 13.6 (#334), Llama 4 Maverick: 10.1 (#342)
| Benchmark | Grok-3 mini | Llama 4 Maverick |
|---|---|---|
| ARC-AGI-2 | 0.4% | 0% |
| Kagi LLM Benchmark | 61.3% | 55.9% |
| ARC-AGI-1 | 16.5% | 4.4% |
| LMArena Hard Prompts | 1375 | 1281 |
| Epoch Capabilities Index | 140.35 | 132.2 |
| SimpleBench | — | 27.7% |
| NYT Connections (extended) | — | 8% |
| CritPt | — | 0% |
| EnigmaEval | — | 0.6% |
| DTBench | — | 61.9% |
| LMCA | — | 15.9% |
| ForecastBench | — | 57.5 |
Math Grok-3 mini leads
Grok-3 mini: 42.1 (#85), Llama 4 Maverick: 26.0 (#262)
| Benchmark | Grok-3 mini | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 77.8% | 20.6% |
| Omni-MATH | 31.8% | 42.2% |
| LMArena Math | 1386 | 1299 |
| MATH Level 5 | 90.9% | 73% |
| FrontierMath (Feb 2025 set) | 5.9% | 0.7% |
Knowledge Grok-3 mini leads
Grok-3 mini: 46.4 (#81), Llama 4 Maverick: 33.4 (#204)
| Benchmark | Grok-3 mini | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 76.3% | 67% |
| MMLU-Pro | 79.9% | 81% |
| Confabulations | 10.8% | 22.6% |
| GPQA (HELM) | 67.5% | 65% |
| LMArena Expert | 1395 | 1259 |
| Humanity's Last Exam | — | 5.7% |
| Vectara Hallucination Rate | — | 8.2% |
Multimodal Not comparable
Grok-3 mini: —, Llama 4 Maverick: 31.6 (#105)
| Benchmark | Grok-3 mini | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | — | 1142 |
| GeoBench | — | 52% |
| SpatialViz-Bench | — | 31.8% |
Multilingual Grok-3 mini leads
Grok-3 mini: 48.1 (#145), Llama 4 Maverick: 42.2 (#195)
| Benchmark | Grok-3 mini | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1352 | 1269 |
| LMArena Chinese | 1387 | 1277 |
| LMArena French | 1357 | 1259 |
| LMArena German | 1349 | 1291 |
| LMArena Japanese | 1342 | 1207 |
| LMArena Korean | 1335 | 1203 |
| LMArena Russian | 1353 | 1286 |
| LMArena Spanish | 1381 | 1293 |
Instruction Following Grok-3 mini leads
Grok-3 mini: 78.5 (#9), Llama 4 Maverick: 71.7 (#146)
| Benchmark | Grok-3 mini | Llama 4 Maverick |
|---|---|---|
| IFEval | 95.1% | 90.8% |
| LMArena Instruction Following | 1357 | 1267 |
Long Context Grok-3 mini leads
Grok-3 mini: 41.0 (#147), Llama 4 Maverick: 31.4 (#279)
| Benchmark | Grok-3 mini | Llama 4 Maverick |
|---|---|---|
| Fiction.LiveBench | 66.7% | 46.2% |
| LMArena Longer Query | 1372 | 1280 |
Writing & Preference Grok-3 mini leads
Grok-3 mini: 52.5 (#169), Llama 4 Maverick: 38.8 (#252)
| Benchmark | Grok-3 mini | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1370 | 1287 |
| LMArena Creative Writing | 1342 | 1267 |
| Short-Story Creative Writing | 73.5% | 62% |
| WildBench | 65.1% | 80% |
| LMArena Multi-Turn | 1355 | 1289 |
| EQ-Bench Creative Writing | — | 860 |
Frequently asked questions
Is Grok-3 mini better than Llama 4 Maverick?
Grok-3 mini is the stronger model overall, scoring 41.2 to 30.9 on the Noometry Index.
Is Grok-3 mini or Llama 4 Maverick better for coding?
Grok-3 mini scores higher on coding benchmarks: 40.8 versus 26.6 in the Noometry coding category.
How many benchmarks do Grok-3 mini and Llama 4 Maverick share?
35 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Llama 4 Maverick has 54.