Which is better, Llama 3.1 Instruct 405B or Qwen3.5 2B?

Llama 3.1 Instruct 405B vs Qwen3.5 2B

25 SHARED BENCHMARKS

Across 25 shared benchmarks, Llama 3.1 Instruct 405B scores higher on 15 and Qwen3.5 2B on 8, with 2 level. The widest gap is AA Agentic Index, where Llama 3.1 Instruct 405B scores 6.3 against 0.8.

METAVSALIBABA25 SHARED15–8 HEAD-TO-HEAD

BenchmarkLlama 3.1 Instruct 405BMARGINQwen3.5 2B

AA Agentic Index6.30.8

AA Intelligence8.56.9

AA-LCR25.325.6

AA-Omniscience-17.1-60.2

AIME 202423.317

Artificial Analysis Coding Index14.52.9

BFCL v281.168.9

bfcl_v488.543.6

C-Eval72.573.2

critpt00

gdpval00

GPQA Diamond51.551.6

HLE4.25

IFBench3941.3

ifeval88.678.6

longbench_v236.138.7

math_500_em73.881.5

mmlu_redux86.279.6

MMLU-Pro73.466.5

mmmlu73.863.1

OmniScience Accuracy23.28.1

OmniScience Non-Hallucination4925.8

scicode29.97.2

Terminal-Bench Hard6.83.8

τ²-Bench Telecom (AA run)1981.6

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.

Compare Llama 3.1 Instruct 405B withALL PAIRINGS →

MetaMuse Spark 1.1

Across the fieldQwen3.6 27B Qwen3.7 Max Qwen3.8 Max Preview Qwen3.7 Plus Preview Claude Opus 4 Qwen3.5 4B Gemini 3.6 Flash Nemotron 3 Super 120B A12B

Compare Qwen3.5 2B withALL PAIRINGS →

AlibabaQwen3.5 35B A3B Qwen3 VL 4B (Reasoning)Qwen3.5 397B A17B Qwen2.5 Instruct 72B Qwen3.5 27B Qwen3.6 Plus

Across the fieldDeepSeek-V3.2 DeepSeek-V4-Flash Claude Opus 4.7 Gemini 2.5 Pro Gemini 1.5 Pro Claude Sonnet 4 Claude Sonnet 5 Muse Spark 1.1