# Codellama 70b Instruct vs Grok 4.3

> Grok 4.3 is the stronger model overall, scoring 43.8 to 33.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/codellama-70b-instruct-vs-grok-4-3
- Last updated: 2026-10-11
- Shared benchmarks: 4

## Summary

- They share 4 benchmarks with published results for both. Codellama 70b Instruct scores higher in 0 categories and Grok 4.3 in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Grok 4.3 leads 50.5 to 24.8.
- Codellama 70b Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| Provider | Meta | xAI |
| Noometry Index | 33.7 | 43.8 |
| Rank | 237 | 86 |
| Context | — | 1M |
| Input $/M | — | $1.25 |
| Output $/M | — | $2.50 |
| Weights | Open | Proprietary |

## Coding

- Codellama 70b Instruct: 37.6 (#193)
- Grok 4.3: 41.6 (#121)

| Benchmark | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| LMArena WebDev | — | 1357 |
| SciCode | — | 47.3% |
| WeirdML | — | 49.9% |
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1415 |
| BigCodeBench Complete | 49.6% | — |
| ALE-Bench | — | 944.17 |
| HumanEval+ | 65.9% | — |

## Agentic & Tool Use

- Codellama 70b Instruct: —
- Grok 4.3: 27.7 (#99)

| Benchmark | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| GDP.pdf | — | 8% |
| LMArena Search | — | 1165 |
| Vending-Bench 2 | — | 35.26 |

## Reasoning

- Codellama 70b Instruct: 20.1 (#242)
- Grok 4.3: 35.9 (#68)

| Benchmark | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1396 |
| NYT Connections (extended) | — | 55.2% |
| CritPt | — | 8% |
| Chess Puzzles | — | 25% |
| DTBench | — | 90.7% |
| LMCA | — | 38.3% |
| Epoch Capabilities Index | — | 149.16 |
| ForecastBench | — | 60.3 |

## Math

- Codellama 70b Instruct: —
- Grok 4.3: 46.0 (#74)

| Benchmark | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 42.8% |
| FrontierMath Tier 4 | — | 14.6% |
| OTIS Mock AIME 2024-2025 | — | 93.3% |
| ProofBench | — | 11% |
| LMArena Math | — | 1388 |

## Knowledge

- Codellama 70b Instruct: —
- Grok 4.3: 52.5 (#62)

| Benchmark | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| GPQA Diamond | — | 88.8% |
| SimpleQA Verified | — | 33.2% |
| LMArena Expert | — | 1385 |

## Multimodal

- Codellama 70b Instruct: —
- Grok 4.3: 31.6 (#104)

| Benchmark | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| LMArena Vision | — | 1229 |
| Blueprint-Bench 2 | — | 0% |

## Multilingual

- Codellama 70b Instruct: 24.8 (#288)
- Grok 4.3: 50.5 (#120)

| Benchmark | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| LMArena Non-English | 992 | 1385 |
| LMArena Chinese | — | 1422 |
| LMArena French | — | 1412 |
| LMArena German | — | 1395 |
| LMArena Japanese | — | 1379 |
| LMArena Korean | — | 1356 |
| LMArena Russian | — | 1399 |
| LMArena Spanish | — | 1398 |

## Instruction Following

- Codellama 70b Instruct: 51.9 (#293)
- Grok 4.3: 72.1 (#140)

| Benchmark | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| LMArena Instruction Following | 1024 | 1366 |

## Long Context

- Codellama 70b Instruct: —
- Grok 4.3: 42.5 (#123)

| Benchmark | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| LMArena Longer Query | — | 1393 |

## Writing & Preference

- Codellama 70b Instruct: 33.4 (#277)
- Grok 4.3: 58.5 (#118)

| Benchmark | Codellama 70b Instruct | Grok 4.3 |
|---|---|---|
| LMArena Text | 1057 | 1397 |
| LMArena Creative Writing | — | 1380 |
| EQ-Bench 4 | — | 1075 |
| LMArena Multi-Turn | — | 1406 |

## FAQ

### Is Codellama 70b Instruct better than Grok 4.3?

Grok 4.3 is the stronger model overall, scoring 43.8 to 33.7 on the Noometry Index.

### Is Codellama 70b Instruct or Grok 4.3 better for coding?

Grok 4.3 scores higher on coding benchmarks: 41.6 versus 37.6 in the Noometry coding category.

### How many benchmarks do Codellama 70b Instruct and Grok 4.3 share?

4 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and Grok 4.3 has 40.