Model comparison
Grok-3 mini vs Llama 3-70B
Grok-3 mini is the stronger model overall, scoring 41.0 to 28.5 on the Noometry Index.
Last verified . 22 shared benchmarks.
MetaSummary
- They share 22 benchmarks with published results for both. Grok-3 mini scores higher in 7 categories and Llama 3-70B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok-3 mini leads 40.3 to 12.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 77.8% for Grok-3 mini and 4.3% for Llama 3-70B.
- Llama 3-70B has downloadable open weights; the other is API-only.
Side by side
| Grok-3 mini | Llama 3-70B | |
|---|---|---|
| Provider | xAI | Meta |
| Noometry Index | 41.0 | 28.5 |
| Released | 2025-04-09 | 2024-04-18 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok-3 mini leads
Grok-3 mini: 40.4 (#126), Llama 3-70B: 35.3 (#215)
| Benchmark | Grok-3 mini | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1380 | 1206 |
| Aider Polyglot | 49.3% | — |
| WeirdML | 42.6% | — |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |
Agentic & Tool Use Not comparable
Grok-3 mini: —, Llama 3-70B: 20.9 (#132)
| Benchmark | Grok-3 mini | Llama 3-70B |
|---|---|---|
| Cybench | — | 5% |
Reasoning Llama 3-70B leads
Grok-3 mini: 13.8 (#323), Llama 3-70B: 17.4 (#284)
| Benchmark | Grok-3 mini | Llama 3-70B |
|---|---|---|
| Kagi LLM Benchmark | 61.3% | 35.1% |
| LMArena Hard Prompts | 1374 | 1195 |
| Epoch Capabilities Index | 140.35 | 122.91 |
| ARC-AGI-2 | 0.4% | — |
| ARC-AGI-1 | 16.5% | — |
| DTBench | — | 54.2% |
| ForecastBench | — | 57.1 |
| WinoGrande | — | 83.5% |
Math Grok-3 mini leads
Grok-3 mini: 40.3 (#78), Llama 3-70B: 12.0 (#299)
| Benchmark | Grok-3 mini | Llama 3-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 77.8% | 4.3% |
| LMArena Math | 1386 | 1218 |
| MATH Level 5 | 90.9% | 22.6% |
| Omni-MATH | 31.8% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
Knowledge Grok-3 mini leads
Grok-3 mini: 46.3 (#75), Llama 3-70B: 20.8 (#272)
| Benchmark | Grok-3 mini | Llama 3-70B |
|---|---|---|
| GPQA Diamond | 76.3% | 40.6% |
| LMArena Expert | 1395 | 1149 |
| MMLU-Pro | 79.9% | — |
| Confabulations | 10.8% | — |
| GPQA (HELM) | 67.5% | — |
| MMLU | — | 79.3% |
Multilingual Grok-3 mini leads
Grok-3 mini: 48.3 (#141), Llama 3-70B: 33.8 (#246)
| Benchmark | Grok-3 mini | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1352 | 1142 |
| LMArena Chinese | 1387 | 1114 |
| LMArena French | 1357 | 1231 |
| LMArena German | 1350 | 1169 |
| LMArena Japanese | 1343 | 1017 |
| LMArena Korean | 1336 | 1017 |
| LMArena Russian | 1352 | 1159 |
| LMArena Spanish | 1379 | 1241 |
Instruction Following Grok-3 mini leads
Grok-3 mini: 78.7 (#8), Llama 3-70B: 62.8 (#233)
| Benchmark | Grok-3 mini | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1356 | 1194 |
| IFEval | 95.1% | — |
Long Context Grok-3 mini leads
Grok-3 mini: 41.0 (#142), Llama 3-70B: 35.6 (#235)
| Benchmark | Grok-3 mini | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1372 | 1174 |
| Fiction.LiveBench | 66.7% | — |
Writing & Preference Grok-3 mini leads
Grok-3 mini: 52.7 (#163), Llama 3-70B: 43.0 (#225)
| Benchmark | Grok-3 mini | Llama 3-70B |
|---|---|---|
| LMArena Text | 1370 | 1221 |
| LMArena Creative Writing | 1342 | 1210 |
| LMArena Multi-Turn | 1355 | 1223 |
| Short-Story Creative Writing | 73.5% | — |
| WildBench | 65.1% | — |
Frequently asked questions
Is Grok-3 mini better than Llama 3-70B?
Grok-3 mini is the stronger model overall, scoring 41.0 to 28.5 on the Noometry Index.
Is Grok-3 mini or Llama 3-70B better for coding?
Grok-3 mini scores higher on coding benchmarks: 40.4 versus 35.3 in the Noometry coding category.
How many benchmarks do Grok-3 mini and Llama 3-70B share?
22 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Llama 3-70B has 31.