# Llama-3.3-70B-Instruct vs Qwen2.5-Max

> Qwen2.5-Max is the stronger model overall, scoring 40.7 to 30.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/llama-3-3-70b-instruct-vs-qwen2-5-max
- Last updated: 2026-10-12
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. Llama-3.3-70B-Instruct scores higher in 0 categories and Qwen2.5-Max in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Max leads 36.9 to 15.3.
- The biggest single-benchmark swing is LiveBench Coding: 36.6% for Llama-3.3-70B-Instruct and 64.4% for Qwen2.5-Max.
- Llama-3.3-70B-Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Llama-3.3-70B-Instruct | Qwen2.5-Max |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.6 | 40.7 |
| Rank | 291 | 146 |
| Context | 128K | — |
| Input $/M | $0.10 | — |
| Output $/M | $0.32 | — |
| Weights | Open | Proprietary |

## Coding

- Llama-3.3-70B-Instruct: 31.0 (#290)
- Qwen2.5-Max: 41.8 (#117)

| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Max |
|---|---|---|
| LiveBench Coding | 36.6% | 64.4% |
| LMArena Coding | 1268 | 1359 |
| SciCode | 26% | — |
| WeirdML | 14.4% | — |
| BigCodeBench Instruct | 46.9% | — |
| BigCodeBench Complete | 57.5% | — |

## Agentic & Tool Use

- Llama-3.3-70B-Instruct: 25.8 (#105)
- Qwen2.5-Max: —

| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 31.9% | — |
| BALROG | 23% | — |

## Reasoning

- Llama-3.3-70B-Instruct: 14.1 (#327)
- Qwen2.5-Max: 25.6 (#147)

| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Max |
|---|---|---|
| LiveBench Reasoning | 50.8% | 51.4% |
| LMArena Hard Prompts | 1257 | 1360 |
| LiveBench Data Analysis | 49.5% | 67.9% |
| Epoch Capabilities Index | 127.33 | 132.53 |
| LiveBench | 50.2% | 62.3% |
| SimpleBench | 19.9% | — |
| CritPt | 0% | — |
| DTBench | 59.5% | — |
| LMCA | 17.5% | — |
| ForecastBench | 58.6 | — |

## Math

- Llama-3.3-70B-Instruct: 15.3 (#298)
- Qwen2.5-Max: 36.9 (#162)

| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Max |
|---|---|---|
| LiveBench Math | 42.2% | 58.4% |
| LMArena Math | 1267 | 1369 |
| OTIS Mock AIME 2024-2025 | 5.1% | — |
| MATH Level 5 | 41.6% | — |

## Knowledge

- Llama-3.3-70B-Instruct: 30.6 (#226)
- Qwen2.5-Max: 35.3 (#186)

| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Max |
|---|---|---|
| Confabulations | 22.8% | 21.8% |
| LMArena Expert | 1225 | 1337 |
| GPQA Diamond | 47.4% | — |
| Vectara Hallucination Rate | 4.1% | — |
| MMLU | 86.3% | — |

## Multilingual

- Llama-3.3-70B-Instruct: 39.9 (#220)
- Qwen2.5-Max: 48.1 (#146)

| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1236 | 1352 |
| LMArena Chinese | 1217 | 1382 |
| LMArena French | 1281 | 1396 |
| LMArena German | 1251 | 1350 |
| LMArena Japanese | 1150 | 1300 |
| LMArena Korean | 1143 | 1304 |
| LMArena Russian | 1252 | 1353 |
| LMArena Spanish | 1270 | 1377 |

## Instruction Following

- Llama-3.3-70B-Instruct: 71.1 (#157)
- Qwen2.5-Max: 71.3 (#152)

| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Max |
|---|---|---|
| LiveBench Instruction Following | 82.7% | 75.3% |
| LMArena Instruction Following | 1242 | 1335 |

## Long Context

- Llama-3.3-70B-Instruct: 26.4 (#295)
- Qwen2.5-Max: 41.4 (#142)

| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1256 | 1358 |
| Fiction.LiveBench | 33.3% | — |

## Writing & Preference

- Llama-3.3-70B-Instruct: 47.6 (#207)
- Qwen2.5-Max: 55.4 (#146)

| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1274 | 1367 |
| LMArena Creative Writing | 1250 | 1339 |
| LMArena Multi-Turn | 1280 | 1364 |
| LiveBench Language | 39.2% | 56.3% |
| Short-Story Creative Writing | — | 72.9% |

## FAQ

### Is Llama-3.3-70B-Instruct better than Qwen2.5-Max?

Qwen2.5-Max is the stronger model overall, scoring 40.7 to 30.6 on the Noometry Index.

### Is Llama-3.3-70B-Instruct or Qwen2.5-Max better for coding?

Qwen2.5-Max scores higher on coding benchmarks: 41.8 versus 31.0 in the Noometry coding category.

### How many benchmarks do Llama-3.3-70B-Instruct and Qwen2.5-Max share?

26 benchmarks have published results for both models. Llama-3.3-70B-Instruct has 43 scored results on Noometry and Qwen2.5-Max has 27.