# Grok-2 (Dec 2024) vs Llama 3.1-70B

> Grok-2 (Dec 2024) is the stronger model overall, scoring 33.7 to 29.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-2-vs-llama-3-1-70b
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. Grok-2 (Dec 2024) scores higher in 7 categories and Llama 3.1-70B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Grok-2 (Dec 2024) leads 48.6 to 35.4.
- The biggest single-benchmark swing is MATH Level 5: 63.5% for Grok-2 (Dec 2024) and 36.7% for Llama 3.1-70B.
- Llama 3.1-70B has downloadable open weights; the other is API-only.

## Snapshot

| | Grok-2 (Dec 2024) | Llama 3.1-70B |
|---|---|---|
| Provider | xAI | Meta |
| Noometry Index | 33.7 | 29.6 |
| Rank | 239 | 308 |
| Context | — | 128K |
| Input $/M | — | $0.40 |
| Output $/M | — | $0.40 |
| Weights | Proprietary | Open |

## Coding

- Grok-2 (Dec 2024): 33.3 (#258)
- Llama 3.1-70B: 30.3 (#296)

| Benchmark | Grok-2 (Dec 2024) | Llama 3.1-70B |
|---|---|---|
| WeirdML | 22.2% | 9% |
| LMArena Coding | 1287 | 1260 |
| BigCodeBench Instruct | — | 46.1% |
| LiveBench Coding | 46.4% | — |
| BigCodeBench Complete | — | 54.8% |

## Agentic & Tool Use

- Grok-2 (Dec 2024): —
- Llama 3.1-70B: 25.1 (#112)

| Benchmark | Grok-2 (Dec 2024) | Llama 3.1-70B |
|---|---|---|
| TheAgentCompany | — | 6.9% |
| BALROG | — | 27.9% |

## Reasoning

- Grok-2 (Dec 2024): 16.9 (#299)
- Llama 3.1-70B: 21.6 (#220)

| Benchmark | Grok-2 (Dec 2024) | Llama 3.1-70B |
|---|---|---|
| LMArena Hard Prompts | 1272 | 1241 |
| DTBench | 65.2% | 60% |
| Epoch Capabilities Index | 130.48 | 125.92 |
| SimpleBench | 22.7% | — |
| LiveBench Reasoning | 54.8% | — |
| LiveBench Data Analysis | 54.5% | — |
| LMCA | — | 14.8% |
| LiveBench | 54.3% | — |

## Math

- Grok-2 (Dec 2024): 20.8 (#284)
- Llama 3.1-70B: 13.5 (#304)

| Benchmark | Grok-2 (Dec 2024) | Llama 3.1-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 11.5% | 3.6% |
| LMArena Math | 1283 | 1252 |
| MATH Level 5 | 63.5% | 36.7% |
| Omni-MATH | — | 21% |
| LiveBench Math | 54.9% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |

## Knowledge

- Grok-2 (Dec 2024): 29.8 (#233)
- Llama 3.1-70B: 24.2 (#269)

| Benchmark | Grok-2 (Dec 2024) | Llama 3.1-70B |
|---|---|---|
| GPQA Diamond | 53.8% | 44.2% |
| LMArena Expert | 1254 | 1209 |
| MMLU-Pro | — | 65.3% |
| Confabulations | 20.1% | — |
| GPQA (HELM) | — | 42.6% |
| MMLU | — | 80.1% |

## Multilingual

- Grok-2 (Dec 2024): 43.1 (#188)
- Llama 3.1-70B: 38.8 (#225)

| Benchmark | Grok-2 (Dec 2024) | Llama 3.1-70B |
|---|---|---|
| LMArena Non-English | 1282 | 1219 |
| LMArena Chinese | 1289 | 1215 |
| LMArena French | 1318 | 1261 |
| LMArena German | 1287 | 1222 |
| LMArena Japanese | 1244 | 1132 |
| LMArena Korean | 1237 | 1140 |
| LMArena Russian | 1286 | 1234 |
| LMArena Spanish | 1281 | 1253 |

## Instruction Following

- Grok-2 (Dec 2024): 66.9 (#202)
- Llama 3.1-70B: 65.3 (#223)

| Benchmark | Grok-2 (Dec 2024) | Llama 3.1-70B |
|---|---|---|
| LMArena Instruction Following | 1270 | 1231 |
| LiveBench Instruction Following | 69.6% | — |
| IFEval | — | 82.1% |

## Long Context

- Grok-2 (Dec 2024): 38.8 (#190)
- Llama 3.1-70B: 37.6 (#214)

| Benchmark | Grok-2 (Dec 2024) | Llama 3.1-70B |
|---|---|---|
| LMArena Longer Query | 1276 | 1241 |

## Writing & Preference

- Grok-2 (Dec 2024): 48.6 (#198)
- Llama 3.1-70B: 35.4 (#267)

| Benchmark | Grok-2 (Dec 2024) | Llama 3.1-70B |
|---|---|---|
| LMArena Text | 1305 | 1261 |
| LMArena Creative Writing | 1284 | 1232 |
| LMArena Multi-Turn | 1290 | 1256 |
| Short-Story Creative Writing | 63.6% | — |
| EQ-Bench Creative Writing | — | 784 |
| WildBench | — | 75.8% |
| LiveBench Language | 45.6% | — |

## FAQ

### Is Grok-2 (Dec 2024) better than Llama 3.1-70B?

Grok-2 (Dec 2024) is the stronger model overall, scoring 33.7 to 29.6 on the Noometry Index.

### Is Grok-2 (Dec 2024) or Llama 3.1-70B better for coding?

Grok-2 (Dec 2024) scores higher on coding benchmarks: 33.3 versus 30.3 in the Noometry coding category.

### How many benchmarks do Grok-2 (Dec 2024) and Llama 3.1-70B share?

23 benchmarks have published results for both models. Grok-2 (Dec 2024) has 34 scored results on Noometry and Llama 3.1-70B has 35.