# Grok-2 (Dec 2024) vs Llama-3.3-70B-Instruct

> Grok-2 (Dec 2024) is the stronger model overall, scoring 33.4 to 30.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-2-vs-llama-3-3-70b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 32

## Summary

- They share 32 benchmarks with published results for both. Grok-2 (Dec 2024) scores higher in 6 categories and Llama-3.3-70B-Instruct in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where Grok-2 (Dec 2024) leads 38.7 to 26.4.
- The biggest single-benchmark swing is MATH Level 5: 63.5% for Grok-2 (Dec 2024) and 41.6% for Llama-3.3-70B-Instruct.
- Llama-3.3-70B-Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Grok-2 (Dec 2024) | Llama-3.3-70B-Instruct |
|---|---|---|
| Provider | xAI | Meta |
| Noometry Index | 33.4 | 30.3 |
| Rank | 233 | 283 |
| Context | — | 128K |
| Input $/M | — | $0.10 |
| Output $/M | — | $0.32 |
| Weights | Proprietary | Open |

## Coding

- Grok-2 (Dec 2024): 32.9 (#254)
- Llama-3.3-70B-Instruct: 30.6 (#284)

| Benchmark | Grok-2 (Dec 2024) | Llama-3.3-70B-Instruct |
|---|---|---|
| WeirdML | 22.2% | 14.4% |
| LiveBench Coding | 46.4% | 36.6% |
| LMArena Coding | 1288 | 1268 |
| SciCode | — | 26% |
| BigCodeBench Instruct | — | 46.9% |
| BigCodeBench Complete | — | 57.5% |

## Agentic & Tool Use

- Grok-2 (Dec 2024): —
- Llama-3.3-70B-Instruct: 25.6 (#100)

| Benchmark | Grok-2 (Dec 2024) | Llama-3.3-70B-Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 31.9% |
| BALROG | — | 23% |

## Reasoning

- Grok-2 (Dec 2024): 16.4 (#294)
- Llama-3.3-70B-Instruct: 13.7 (#325)

| Benchmark | Grok-2 (Dec 2024) | Llama-3.3-70B-Instruct |
|---|---|---|
| SimpleBench | 22.7% | 19.9% |
| LiveBench Reasoning | 54.8% | 50.8% |
| LMArena Hard Prompts | 1272 | 1257 |
| DTBench | 65.2% | 59.5% |
| LiveBench Data Analysis | 54.5% | 49.5% |
| Epoch Capabilities Index | 130.48 | 127.32 |
| LiveBench | 54.3% | 50.2% |
| CritPt | — | 0% |
| LMCA | — | 17.5% |
| ForecastBench | — | 58.6 |

## Math

- Grok-2 (Dec 2024): 19.6 (#278)
- Llama-3.3-70B-Instruct: 14.3 (#292)

| Benchmark | Grok-2 (Dec 2024) | Llama-3.3-70B-Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 11.5% | 5.1% |
| LiveBench Math | 54.9% | 42.2% |
| LMArena Math | 1283 | 1268 |
| MATH Level 5 | 63.5% | 41.6% |
| FrontierMath (Feb 2025 set) | 0.7% | — |

## Knowledge

- Grok-2 (Dec 2024): 29.8 (#225)
- Llama-3.3-70B-Instruct: 30.5 (#220)

| Benchmark | Grok-2 (Dec 2024) | Llama-3.3-70B-Instruct |
|---|---|---|
| GPQA Diamond | 53.8% | 47.4% |
| Confabulations | 20.1% | 22.8% |
| LMArena Expert | 1254 | 1225 |
| Vectara Hallucination Rate | — | 4.1% |
| MMLU | — | 86.3% |

## Multilingual

- Grok-2 (Dec 2024): 43.3 (#183)
- Llama-3.3-70B-Instruct: 40.1 (#215)

| Benchmark | Grok-2 (Dec 2024) | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1282 | 1236 |
| LMArena Chinese | 1289 | 1217 |
| LMArena French | 1317 | 1280 |
| LMArena German | 1287 | 1251 |
| LMArena Japanese | 1244 | 1149 |
| LMArena Korean | 1237 | 1142 |
| LMArena Russian | 1286 | 1252 |
| LMArena Spanish | 1281 | 1269 |

## Instruction Following

- Grok-2 (Dec 2024): 67.1 (#197)
- Llama-3.3-70B-Instruct: 71.3 (#152)

| Benchmark | Grok-2 (Dec 2024) | Llama-3.3-70B-Instruct |
|---|---|---|
| LiveBench Instruction Following | 69.6% | 82.7% |
| LMArena Instruction Following | 1270 | 1242 |

## Long Context

- Grok-2 (Dec 2024): 38.7 (#186)
- Llama-3.3-70B-Instruct: 26.4 (#290)

| Benchmark | Grok-2 (Dec 2024) | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Longer Query | 1276 | 1256 |
| Fiction.LiveBench | — | 33.3% |

## Writing & Preference

- Grok-2 (Dec 2024): 48.8 (#192)
- Llama-3.3-70B-Instruct: 47.8 (#203)

| Benchmark | Grok-2 (Dec 2024) | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1305 | 1274 |
| LMArena Creative Writing | 1284 | 1250 |
| LMArena Multi-Turn | 1290 | 1279 |
| LiveBench Language | 45.6% | 39.2% |
| Short-Story Creative Writing | 63.6% | — |

## FAQ

### Is Grok-2 (Dec 2024) better than Llama-3.3-70B-Instruct?

Grok-2 (Dec 2024) is the stronger model overall, scoring 33.4 to 30.3 on the Noometry Index.

### Is Grok-2 (Dec 2024) or Llama-3.3-70B-Instruct better for coding?

Grok-2 (Dec 2024) scores higher on coding benchmarks: 32.9 versus 30.6 in the Noometry coding category.

### How many benchmarks do Grok-2 (Dec 2024) and Llama-3.3-70B-Instruct share?

32 benchmarks have published results for both models. Grok-2 (Dec 2024) has 34 scored results on Noometry and Llama-3.3-70B-Instruct has 43.