Model comparison
Grok 3 vs Llama 3-70B
Grok 3 is the stronger model overall, scoring 39.9 to 28.8 on the Noometry Index.
Last verified . 22 shared benchmarks.
MetaSummary
- They share 22 benchmarks with published results for both. Grok 3 scores higher in 8 categories and Llama 3-70B in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok 3 leads 46.2 to 20.8.
- The biggest single-benchmark swing is MATH Level 5: 88.7% for Grok 3 and 22.6% for Llama 3-70B.
- Llama 3-70B has downloadable open weights; the other is API-only.
Side by side
| Grok 3 | Llama 3-70B | |
|---|---|---|
| Provider | xAI | Meta |
| Noometry Index | 39.9 | 28.8 |
| Released | 2025-04-09 | 2024-04-18 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 40 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
Grok 3: 41.9 (#115), Llama 3-70B: 35.8 (#218)
| Benchmark | Grok 3 | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1432 | 1206 |
| Aider Polyglot | 53.3% | — |
| WeirdML | 37.2% | — |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |
Agentic & Tool Use Grok 3 leads
Grok 3: 30.5 (#76), Llama 3-70B: 21.1 (#139)
Reasoning Llama 3-70B leads
Grok 3: 13.7 (#333), Llama 3-70B: 18.0 (#288)
| Benchmark | Grok 3 | Llama 3-70B |
|---|---|---|
| Kagi LLM Benchmark | 61.3% | 35.1% |
| LMArena Hard Prompts | 1434 | 1195 |
| Epoch Capabilities Index | 138.33 | 122.93 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 36.1% | — |
| ARC-AGI-1 | 5.5% | — |
| DTBench | — | 54.2% |
| ForecastBench | — | 57.1 |
| WinoGrande | — | 83.5% |
Math Grok 3 leads
Grok 3: 38.0 (#145), Llama 3-70B: 12.8 (#305)
| Benchmark | Grok 3 | Llama 3-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 55.6% | 4.3% |
| LMArena Math | 1391 | 1218 |
| MATH Level 5 | 88.7% | 22.6% |
| Omni-MATH | 46.4% | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Grok 3 leads
Grok 3: 46.2 (#82), Llama 3-70B: 20.8 (#277)
| Benchmark | Grok 3 | Llama 3-70B |
|---|---|---|
| GPQA Diamond | 75.8% | 40.6% |
| LMArena Expert | 1421 | 1149 |
| MMLU-Pro | 78.8% | — |
| Confabulations | 14.2% | — |
| Vectara Hallucination Rate | 5.8% | — |
| GPQA (HELM) | 65% | — |
| MMLU | — | 79.3% |
Multilingual Grok 3 leads
Grok 3: 52.3 (#87), Llama 3-70B: 33.6 (#251)
| Benchmark | Grok 3 | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1410 | 1142 |
| LMArena Chinese | 1448 | 1114 |
| LMArena French | 1460 | 1232 |
| LMArena German | 1431 | 1169 |
| LMArena Japanese | 1387 | 1017 |
| LMArena Korean | 1373 | 1017 |
| LMArena Russian | 1416 | 1159 |
| LMArena Spanish | 1417 | 1241 |
Instruction Following Grok 3 leads
Grok 3: 75.0 (#73), Llama 3-70B: 62.5 (#238)
| Benchmark | Grok 3 | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1409 | 1194 |
| IFEval | 88.4% | — |
Long Context Grok 3 leads
Grok 3: 38.7 (#192), Llama 3-70B: 35.6 (#240)
| Benchmark | Grok 3 | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1439 | 1174 |
| Fiction.LiveBench | 58.3% | — |
Writing & Preference Grok 3 leads
Grok 3: 55.8 (#141), Llama 3-70B: 42.8 (#231)
| Benchmark | Grok 3 | Llama 3-70B |
|---|---|---|
| LMArena Text | 1426 | 1221 |
| LMArena Creative Writing | 1414 | 1210 |
| LMArena Multi-Turn | 1425 | 1223 |
| Short-Story Creative Writing | 76.4% | — |
| EQ-Bench Creative Writing | 1186 | — |
| WildBench | 84.9% | — |
Frequently asked questions
Is Grok 3 better than Llama 3-70B?
Grok 3 is the stronger model overall, scoring 39.9 to 28.8 on the Noometry Index.
Is Grok 3 or Llama 3-70B better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 35.8 in the Noometry coding category.
How many benchmarks do Grok 3 and Llama 3-70B share?
22 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Llama 3-70B has 31.