# Grok 3 vs Llama 3.1-70B

> Grok 3 is the stronger model overall, scoring 39.6 to 29.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-3-vs-llama-3-1-70b
- Last updated: 2026-10-10
- Shared benchmarks: 29

## Summary

- They share 29 benchmarks with published results for both. Grok 3 scores higher in 8 categories and Llama 3.1-70B in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 3 leads 36.2 to 12.7.
- The biggest single-benchmark swing is MATH Level 5: 88.7% for Grok 3 and 36.7% for Llama 3.1-70B.
- Llama 3.1-70B has downloadable open weights; the other is API-only.

## Snapshot

| | Grok 3 | Llama 3.1-70B |
|---|---|---|
| Provider | xAI | Meta |
| Noometry Index | 39.6 | 29.4 |
| Rank | 150 | 299 |
| Context | — | 128K |
| Input $/M | — | $0.40 |
| Output $/M | — | $0.40 |
| Weights | Proprietary | Open |

## Coding

- Grok 3: 41.5 (#113)
- Llama 3.1-70B: 29.8 (#292)

| Benchmark | Grok 3 | Llama 3.1-70B |
|---|---|---|
| WeirdML | 37.2% | 9% |
| LMArena Coding | 1432 | 1260 |
| Aider Polyglot | 53.3% | — |
| BigCodeBench Instruct | — | 46.1% |
| BigCodeBench Complete | — | 54.8% |

## Agentic & Tool Use

- Grok 3: 30.2 (#69)
- Llama 3.1-70B: 24.9 (#104)

| Benchmark | Grok 3 | Llama 3.1-70B |
|---|---|---|
| BALROG | 29.5% | 27.9% |
| TheAgentCompany | — | 6.9% |

## Reasoning

- Grok 3: 13.8 (#322)
- Llama 3.1-70B: 20.8 (#215)

| Benchmark | Grok 3 | Llama 3.1-70B |
|---|---|---|
| LMArena Hard Prompts | 1433 | 1241 |
| Epoch Capabilities Index | 138.34 | 125.91 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 36.1% | — |
| Kagi LLM Benchmark | 61.3% | — |
| ARC-AGI-1 | 5.5% | — |
| DTBench | — | 60% |
| LMCA | — | 14.8% |

## Math

- Grok 3: 36.2 (#122)
- Llama 3.1-70B: 12.7 (#298)

| Benchmark | Grok 3 | Llama 3.1-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 55.6% | 3.6% |
| Omni-MATH | 46.4% | 21% |
| LMArena Math | 1390 | 1252 |
| MATH Level 5 | 88.7% | 36.7% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |

## Knowledge

- Grok 3: 46.0 (#76)
- Llama 3.1-70B: 24.1 (#264)

| Benchmark | Grok 3 | Llama 3.1-70B |
|---|---|---|
| GPQA Diamond | 75.8% | 44.2% |
| MMLU-Pro | 78.8% | 65.3% |
| GPQA (HELM) | 65% | 42.6% |
| LMArena Expert | 1420 | 1209 |
| Confabulations | 14.2% | — |
| Vectara Hallucination Rate | 5.8% | — |
| MMLU | — | 80.1% |

## Multilingual

- Grok 3: 52.5 (#81)
- Llama 3.1-70B: 38.9 (#220)

| Benchmark | Grok 3 | Llama 3.1-70B |
|---|---|---|
| LMArena Non-English | 1411 | 1219 |
| LMArena Chinese | 1448 | 1215 |
| LMArena French | 1460 | 1260 |
| LMArena German | 1431 | 1222 |
| LMArena Japanese | 1389 | 1132 |
| LMArena Korean | 1374 | 1140 |
| LMArena Russian | 1416 | 1234 |
| LMArena Spanish | 1417 | 1253 |

## Instruction Following

- Grok 3: 75.2 (#69)
- Llama 3.1-70B: 65.5 (#218)

| Benchmark | Grok 3 | Llama 3.1-70B |
|---|---|---|
| IFEval | 88.4% | 82.1% |
| LMArena Instruction Following | 1409 | 1231 |

## Long Context

- Grok 3: 38.7 (#187)
- Llama 3.1-70B: 37.6 (#208)

| Benchmark | Grok 3 | Llama 3.1-70B |
|---|---|---|
| LMArena Longer Query | 1439 | 1241 |
| Fiction.LiveBench | 58.3% | — |

## Writing & Preference

- Grok 3: 56.0 (#136)
- Llama 3.1-70B: 35.6 (#262)

| Benchmark | Grok 3 | Llama 3.1-70B |
|---|---|---|
| LMArena Text | 1426 | 1261 |
| LMArena Creative Writing | 1414 | 1232 |
| EQ-Bench Creative Writing | 1186 | 784 |
| WildBench | 84.9% | 75.8% |
| LMArena Multi-Turn | 1424 | 1256 |
| Short-Story Creative Writing | 76.4% | — |

## FAQ

### Is Grok 3 better than Llama 3.1-70B?

Grok 3 is the stronger model overall, scoring 39.6 to 29.4 on the Noometry Index.

### Is Grok 3 or Llama 3.1-70B better for coding?

Grok 3 scores higher on coding benchmarks: 41.5 versus 29.8 in the Noometry coding category.

### How many benchmarks do Grok 3 and Llama 3.1-70B share?

29 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Llama 3.1-70B has 35.