Neura Intelligence Index

AI Model Benchmarks

Back to Leaderboard

Llama 3.3 70B Instruct vs Qwen2.5 VL 72B Instruct

M

Llama 3.3 70B Instruct

Meta 🇺🇸

18.8

#307—

Q

Qwen2.5 VL 72B Instruct

Alibaba Cloud / Qwen Team 🇨🇳

18.0

#310—

Category Comparison

Benchmark-by-Benchmark

reasoning

Benchmark	Llama 3.3 70B Instruct	Qwen2.5 VL 72B Instruct
GPQA Diamond	50.5	—

multimodal

Benchmark	Llama 3.3 70B Instruct	Qwen2.5 VL 72B Instruct
MMMU	—	70.2
MMMU-Pro	—	51.1
ScreenSpot Pro	—	43.6

agent

Benchmark	Llama 3.3 70B Instruct	Qwen2.5 VL 72B Instruct
OSWorld	—	8.8

Category Winners

reasoning

M

Llama 3.3 70B Instruct

18.8

multimodal

Q

Qwen2.5 VL 72B Instruct

25.0

agent

Q

Qwen2.5 VL 72B Instruct

3.8

Llama 3.3 70B Instruct vs Qwen2.5 VL 72B Instruct — Benchmark Comparison | Neura Market