# Amazon Nova Pro vs Llama 3-70B

> Amazon Nova Pro is the stronger model overall, scoring 31.0 to 28.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/amazon-nova-pro-vs-llama-3-70b
- Last updated: 2026-10-10
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Amazon Nova Pro scores higher in 7 categories and Llama 3-70B in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Amazon Nova Pro leads 28.5 to 12.8.
- Llama 3-70B has downloadable open weights; the other is API-only.

## Snapshot

| | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| Provider | Amazon | Meta |
| Noometry Index | 31.0 | 28.8 |
| Rank | 281 | 323 |
| Context | 300K | — |
| Input $/M | $0.80 | — |
| Output $/M | $3.20 | — |
| Weights | Proprietary | Open |

## Coding

- Amazon Nova Pro: 35.1 (#229)
- Llama 3-70B: 35.8 (#218)

| Benchmark | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1270 | 1206 |
| BigCodeBench Instruct | — | 43.6% |
| LiveBench Coding | 38.1% | — |
| BigCodeBench Complete | — | 54.5% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |

## Agentic & Tool Use

- Amazon Nova Pro: 16.7 (#147)
- Llama 3-70B: 21.1 (#139)

| Benchmark | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | — |
| TheAgentCompany | 1.7% | — |
| Cybench | — | 5% |

## Reasoning

- Amazon Nova Pro: 20.0 (#243)
- Llama 3-70B: 18.0 (#288)

| Benchmark | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1195 |
| Epoch Capabilities Index | 123.8 | 122.93 |
| Kagi LLM Benchmark | — | 35.1% |
| LiveBench Reasoning | 32.6% | — |
| DTBench | — | 54.2% |
| LiveBench Data Analysis | 48.3% | — |
| ForecastBench | — | 57.1 |
| LiveBench | 43.5% | — |
| WinoGrande | — | 83.5% |

## Math

- Amazon Nova Pro: 28.5 (#243)
- Llama 3-70B: 12.8 (#305)

| Benchmark | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| LMArena Math | 1252 | 1218 |
| OTIS Mock AIME 2024-2025 | — | 4.3% |
| Omni-MATH | 24.2% | — |
| LiveBench Math | 38% | — |
| MATH Level 5 | — | 22.6% |

## Knowledge

- Amazon Nova Pro: 27.4 (#250)
- Llama 3-70B: 20.8 (#277)

| Benchmark | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| LMArena Expert | 1211 | 1149 |
| MMLU | 82% | 79.3% |
| GPQA Diamond | — | 40.6% |
| Humanity's Last Exam | 4.4% | — |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| Vectara Hallucination Rate | 5.1% | — |
| GPQA (HELM) | 44.6% | — |

## Multimodal

- Amazon Nova Pro: 25.0 (#126)
- Llama 3-70B: —

| Benchmark | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| LMArena Vision | 980 | — |

## Multilingual

- Amazon Nova Pro: 39.7 (#223)
- Llama 3-70B: 33.6 (#251)

| Benchmark | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1234 | 1142 |
| LMArena Chinese | 1244 | 1114 |
| LMArena French | 1271 | 1232 |
| LMArena German | 1243 | 1169 |
| LMArena Japanese | 1200 | 1017 |
| LMArena Korean | 1203 | 1017 |
| LMArena Russian | 1240 | 1159 |
| LMArena Spanish | 1182 | 1241 |

## Instruction Following

- Amazon Nova Pro: 64.9 (#226)
- Llama 3-70B: 62.5 (#238)

| Benchmark | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1235 | 1194 |
| LiveBench Instruction Following | 67.1% | — |
| IFEval | 81.5% | — |

## Long Context

- Amazon Nova Pro: 38.1 (#205)
- Llama 3-70B: 35.6 (#240)

| Benchmark | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1255 | 1174 |

## Writing & Preference

- Amazon Nova Pro: 43.9 (#226)
- Llama 3-70B: 42.8 (#231)

| Benchmark | Amazon Nova Pro | Llama 3-70B |
|---|---|---|
| LMArena Text | 1259 | 1221 |
| LMArena Creative Writing | 1212 | 1210 |
| LMArena Multi-Turn | 1246 | 1223 |
| Short-Story Creative Writing | 60.5% | — |
| WildBench | 77.7% | — |
| LiveBench Language | 37% | — |

## FAQ

### Is Amazon Nova Pro better than Llama 3-70B?

Amazon Nova Pro is the stronger model overall, scoring 31.0 to 28.8 on the Noometry Index.

### Is Amazon Nova Pro or Llama 3-70B better for coding?

They score almost the same on coding (35.1 vs 35.8); test both on your own repository before choosing.

### How many benchmarks do Amazon Nova Pro and Llama 3-70B share?

19 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Llama 3-70B has 31.