# Llama 2-70B vs Qwen3.8 Max

> Qwen3.8 Max is the stronger model overall, scoring 56.0 to 24.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/llama-2-70b-vs-qwen3-8-max
- Last updated: 2026-10-10
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Qwen3.8 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.1 to 7.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0% for Llama 2-70B and 100% for Qwen3.8 Max.
- Llama 2-70B has downloadable open weights; the other is API-only.

## Snapshot

| | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 24.1 | 56.0 |
| Rank | 343 | 21 |
| Context | — | 1M |
| Input $/M | — | $2 |
| Output $/M | — | $6 |
| Weights | Open | Proprietary |

## Coding

- Llama 2-70B: 30.8 (#281)
- Qwen3.8 Max: 52.3 (#30)

| Benchmark | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| LMArena Coding | 1079 | 1499 |
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1671 |
| FrontierSWE | — | 15.8% |
| SciCode | — | 52.9% |

## Agentic & Tool Use

- Llama 2-70B: —
- Qwen3.8 Max: 42.6 (#17)

| Benchmark | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |

## Reasoning

- Llama 2-70B: 13.7 (#324)
- Qwen3.8 Max: 53.4 (#26)

| Benchmark | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1493 |
| DTBench | 41.6% | 92% |
| Epoch Capabilities Index | 113.59 | 156.69 |
| NYT Connections (extended) | — | 88.3% |
| CritPt | — | 20% |
| Chess Puzzles | — | 40% |
| Mystery Game Puzzles | — | 38% |
| LMCA | — | 46.2% |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |

## Math

- Llama 2-70B: 7.5 (#320)
- Qwen3.8 Max: 73.1 (#15)

| Benchmark | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0% | 100% |
| LMArena Math | 1091 | 1498 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| MATH Level 5 | 3.3% | — |
| GSM8K | 69.6% | — |

## Knowledge

- Llama 2-70B: 7.4 (#305)
- Qwen3.8 Max: 61.7 (#25)

| Benchmark | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 26.3% | 92.7% |
| LMArena Expert | 1039 | 1504 |
| SimpleQA Verified | — | 47.3% |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |

## Multimodal

- Llama 2-70B: —
- Qwen3.8 Max: 36.9 (#70)

| Benchmark | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |

## Multilingual

- Llama 2-70B: 27.9 (#269)
- Qwen3.8 Max: 56.6 (#18)

| Benchmark | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1045 | 1468 |
| LMArena Chinese | 994 | 1537 |
| LMArena French | 1089 | 1502 |
| LMArena German | 1041 | 1489 |
| LMArena Japanese | 927 | 1466 |
| LMArena Korean | 964 | 1467 |
| LMArena Russian | 1083 | 1478 |
| LMArena Spanish | 1142 | 1495 |

## Instruction Following

- Llama 2-70B: 55.1 (#273)
- Qwen3.8 Max: 77.7 (#19)

| Benchmark | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1071 | 1478 |

## Long Context

- Llama 2-70B: 32.2 (#265)
- Qwen3.8 Max: 45.5 (#30)

| Benchmark | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1062 | 1486 |

## Writing & Preference

- Llama 2-70B: 32.5 (#274)
- Qwen3.8 Max: 67.0 (#31)

| Benchmark | Llama 2-70B | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1115 | 1480 |
| LMArena Creative Writing | 1075 | 1478 |
| LMArena Multi-Turn | 1088 | 1487 |

## FAQ

### Is Llama 2-70B better than Qwen3.8 Max?

Qwen3.8 Max is the stronger model overall, scoring 56.0 to 24.1 on the Noometry Index.

### Is Llama 2-70B or Qwen3.8 Max better for coding?

Qwen3.8 Max scores higher on coding benchmarks: 52.3 versus 30.8 in the Noometry coding category.

### How many benchmarks do Llama 2-70B and Qwen3.8 Max share?

21 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Qwen3.8 Max has 38.