Qwen3.6vsGLM-4.7

Qwen3.6
GLM-4.7
Specifications
Parameters
35B
Context window
256k
128k
Benchmarks
Agentic coding
SWE-Bench Pro
49.5%
Coding
SWE-Bench Verified
73.4%
73.8%Best
Multilingual coding
SWE-Bench Multilingual
67.2%
Agentic terminal coding
Terminal-Bench 2.0
51.5%Best
41%
Web browsing
BrowseComp
52%
Multidisciplinary reasoning
Humanity's Last Exam · no tools
21.4%
24.8%Best
Multidisciplinary reasoning
Humanity's Last Exam · with tools
42.8%
Science
GPQA Diamond
86%Best
85.7%
Chart reasoning
CharXiv Reasoning
78%
Multimodal reasoning
MMMU-Pro
75.3%
Multimodal
MMMU
81.7%
Community preference (code)
Arena Elo (Code)
1433
Overview
CompanyQwenZ.ai
Release dateApr 16 2026Dec 22 2025
AccessOpen WeightOpen Weight

Which is better: Qwen3.6 or GLM-4.7?

Qwen3.6 and GLM-4.7 are evenly matched across the 4 benchmarks they both report (SWE-Bench Verified, Terminal-Bench 2.0, Humanity's Last Exam, GPQA Diamond). GLM-4.7 shipped 115 days before Qwen3.6, so benchmark comparisons should account for the intervening progress.

Context windows are 256k (Qwen3.6) vs 128k (GLM-4.7).

On SWE-Bench Verified, GLM-4.7 leads at 73.8% vs Qwen3.6 at 73.4%. On Terminal-Bench 2.0, Qwen3.6 leads at 51.5% vs GLM-4.7 at 41%. On Humanity's Last Exam · no tools, GLM-4.7 leads at 24.8% vs Qwen3.6 at 21.4%. On GPQA Diamond, Qwen3.6 leads at 86% vs GLM-4.7 at 85.7%.

Frequently asked questions

Qwen3.6 was released by Qwen on Apr 16 2026.

GLM-4.7 was released by Z.ai on Dec 22 2025.

GLM-4.7 leads on SWE-Bench Verified — Qwen3.6 73.4% vs GLM-4.7 73.8%.

GLM-4.7 leads on Humanity's Last Exam · no tools — Qwen3.6 21.4% vs GLM-4.7 24.8%.

Qwen3.6 has a 256k context window; GLM-4.7 has 128k.

Other comparisons