# Llama 2-70B vs Qwen3.5-Flash

> Qwen3.5-Flash is the stronger model overall, scoring 42.3 to 24.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/llama-2-70b-vs-qwen3-5-flash
- 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 Qwen3.5-Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5-Flash leads 43.2 to 7.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0% for Llama 2-70B and 84.4% for Qwen3.5-Flash.
- Llama 2-70B has downloadable open weights; the other is API-only.

## Snapshot

| | Llama 2-70B | Qwen3.5-Flash |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 24.1 | 42.3 |
| Rank | 343 | 105 |
| Context | — | 1M |
| Input $/M | — | $0.10 |
| Output $/M | — | $0.40 |
| Weights | Open | Proprietary |

## Coding

- Llama 2-70B: 30.8 (#281)
- Qwen3.5-Flash: 33.9 (#234)

| Benchmark | Llama 2-70B | Qwen3.5-Flash |
|---|---|---|
| LMArena Coding | 1079 | 1412 |
| LMArena WebDev | — | 1243 |
| ALE-Bench | — | 221.8 |

## Agentic & Tool Use

- Llama 2-70B: —
- Qwen3.5-Flash: —

| Benchmark | Llama 2-70B | Qwen3.5-Flash |
|---|---|---|
| Vending-Bench 2 | — | 462.69 |

## Reasoning

- Llama 2-70B: 13.7 (#324)
- Qwen3.5-Flash: 32.9 (#63)

| Benchmark | Llama 2-70B | Qwen3.5-Flash |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1404 |
| DTBench | 41.6% | 82.9% |
| Epoch Capabilities Index | 113.59 | 143.98 |
| Chess Puzzles | — | 21% |
| Mystery Game Puzzles | — | 20% |
| LMCA | — | 29.1% |
| 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)
- Qwen3.5-Flash: 36.3 (#117)

| Benchmark | Llama 2-70B | Qwen3.5-Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0% | 84.4% |
| LMArena Math | 1091 | 1407 |
| FrontierMath (Tiers 1-3) | — | 18.2% |
| MATH Level 5 | 3.3% | — |
| FrontierMath (Feb 2025 set) | — | 6.2% |
| FrontierMath Tier 4 (v1) | — | 0% |
| GSM8K | 69.6% | — |

## Knowledge

- Llama 2-70B: 7.4 (#305)
- Qwen3.5-Flash: 43.2 (#84)

| Benchmark | Llama 2-70B | Qwen3.5-Flash |
|---|---|---|
| GPQA Diamond | 26.3% | 82.3% |
| LMArena Expert | 1039 | 1408 |
| SimpleQA Verified | — | 20.3% |
| Vectara Hallucination Rate | — | 10.5% |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |

## Multilingual

- Llama 2-70B: 27.9 (#269)
- Qwen3.5-Flash: 50.7 (#116)

| Benchmark | Llama 2-70B | Qwen3.5-Flash |
|---|---|---|
| LMArena Non-English | 1045 | 1385 |
| LMArena Chinese | 994 | 1446 |
| LMArena French | 1089 | 1412 |
| LMArena German | 1041 | 1389 |
| LMArena Japanese | 927 | 1369 |
| LMArena Korean | 964 | 1344 |
| LMArena Russian | 1083 | 1379 |
| LMArena Spanish | 1142 | 1399 |

## Instruction Following

- Llama 2-70B: 55.1 (#273)
- Qwen3.5-Flash: 72.8 (#133)

| Benchmark | Llama 2-70B | Qwen3.5-Flash |
|---|---|---|
| LMArena Instruction Following | 1071 | 1374 |

## Long Context

- Llama 2-70B: 32.2 (#265)
- Qwen3.5-Flash: 42.4 (#119)

| Benchmark | Llama 2-70B | Qwen3.5-Flash |
|---|---|---|
| LMArena Longer Query | 1062 | 1392 |

## Writing & Preference

- Llama 2-70B: 32.5 (#274)
- Qwen3.5-Flash: 58.2 (#117)

| Benchmark | Llama 2-70B | Qwen3.5-Flash |
|---|---|---|
| LMArena Text | 1115 | 1397 |
| LMArena Creative Writing | 1075 | 1343 |
| LMArena Multi-Turn | 1088 | 1393 |

## FAQ

### Is Llama 2-70B better than Qwen3.5-Flash?

Qwen3.5-Flash is the stronger model overall, scoring 42.3 to 24.1 on the Noometry Index.

### Is Llama 2-70B or Qwen3.5-Flash better for coding?

Qwen3.5-Flash scores higher on coding benchmarks: 33.9 versus 30.8 in the Noometry coding category.

### How many benchmarks do Llama 2-70B and Qwen3.5-Flash share?

21 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Qwen3.5-Flash has 32.