Neura Intelligence Index

AI Model Benchmarks

Back to Leaderboard

Qwen3.5-35B-A3B vs Qwen3-Next-80B-A3B-Thinking

Q

Qwen3.5-35B-A3B

Alibaba Cloud / Qwen Team 🇨🇳

65.9

#91—

Q

Qwen3-Next-80B-A3B-Thinking

Alibaba Cloud / Qwen Team 🇨🇳

66.8

#88—

Category Comparison

Benchmark-by-Benchmark

reasoning

Benchmark	Qwen3.5-35B-A3B	Qwen3-Next-80B-A3B-Thinking
GPQA Diamond	84.2	77.2
Humanity's Last Exam	47.4	—

coding

Benchmark	Qwen3.5-35B-A3B	Qwen3-Next-80B-A3B-Thinking
SWE-bench Verified	69.2	—

multimodal

Benchmark	Qwen3.5-35B-A3B	Qwen3-Next-80B-A3B-Thinking
MMMU	81.4	—
MMMU-Pro	75.1	—
CharXiv Reasoning	77.5	—
ScreenSpot Pro	68.6	—

general

Benchmark	Qwen3.5-35B-A3B	Qwen3-Next-80B-A3B-Thinking
MMMLU	85.2	—

agent

Benchmark	Qwen3.5-35B-A3B	Qwen3-Next-80B-A3B-Thinking
BrowseComp	61	—
TAU-Bench Retail	—	69.6

math

Benchmark	Qwen3.5-35B-A3B	Qwen3-Next-80B-A3B-Thinking
AIME 2025	—	87.8

Category Winners

coding

Q

Qwen3.5-35B-A3B

59.3

math

Q

Qwen3-Next-80B-A3B-Thinking

66.1

reasoning

Q

Qwen3.5-35B-A3B

82.0

general

Q

Qwen3.5-35B-A3B

57.8

multimodal

Q

Qwen3.5-35B-A3B

70.8

agent

Q

Qwen3-Next-80B-A3B-Thinking

55.5

Qwen3.5-35B-A3B vs Qwen3-Next-80B-A3B-Thinking — Benchmark Comparison | Neura Market