Qwen3.5vsQwen3.6
Qwen3.5 | Qwen3.6 | |
|---|---|---|
| Specifications | ||
Parameters | 397B | 35B |
Context window | 1M | 256k |
| Benchmarks | ||
Agentic coding SWE-Bench ProCan the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better. | — | 49.5% |
Coding SWE-Bench VerifiedReal coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 76.4%Best | 73.4% |
Multilingual coding SWE-Bench MultilingualLike SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | 69.3%Best | 67.2% |
Agentic terminal coding Terminal-Bench 2.0Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? (Version 2.0 of the test.) Higher is better. | 52.5%Best | 51.5% |
Web browsing BrowseCompCan the AI browse the web and track down hard-to-find answers? Higher is better. | 69% | — |
Multidisciplinary reasoning Humanity's Last Exam · no toolsHumanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better. | 28.7%Best | 21.4% |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 88.4%Best | 86% |
Agentic computer use OSWorld-VerifiedCan the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better. | 62.2% | — |
Chart reasoning CharXiv ReasoningCan the AI read and reason about complex charts and figures, not just text? Higher is better. | 80.8%Best | 78% |
Multimodal reasoning MMMU-ProA tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | 79%Best | 75.3% |
Multimodal MMMUTests the AI on understanding images and text together across many college subjects. Higher is better. | 85%Best | 81.7% |
| Overview | ||
| Company | Qwen | Qwen |
| Release date | Feb 16 2026 | Apr 16 2026 |
| Access | Open Weight | Open Weight |
Which is better: Qwen3.5 or Qwen3.6?
Qwen3.5 leads Qwen3.6 on 8 of the 8 benchmarks they both report. Qwen3.5 shipped 59 days before Qwen3.6, so benchmark comparisons should account for the intervening progress.
Qwen3.5 has 397B parameters, while Qwen3.6 has 35B. Context windows are 1M (Qwen3.5) vs 256k (Qwen3.6).
On SWE-Bench Verified, Qwen3.5 leads at 76.4% vs Qwen3.6 at 73.4%. On SWE-Bench Multilingual, Qwen3.5 leads at 69.3% vs Qwen3.6 at 67.2%. On Terminal-Bench 2.0, Qwen3.5 leads at 52.5% vs Qwen3.6 at 51.5%. On Humanity's Last Exam · no tools, Qwen3.5 leads at 28.7% vs Qwen3.6 at 21.4%. On GPQA Diamond, Qwen3.5 leads at 88.4% vs Qwen3.6 at 86%. On CharXiv Reasoning, Qwen3.5 leads at 80.8% vs Qwen3.6 at 78%. On MMMU-Pro, Qwen3.5 leads at 79% vs Qwen3.6 at 75.3%. On MMMU, Qwen3.5 leads at 85% vs Qwen3.6 at 81.7%.
Frequently asked questions
Qwen3.5 was released by Qwen on Feb 16 2026.
Qwen3.6 was released by Qwen on Apr 16 2026.
Qwen3.5 leads on SWE-Bench Verified — Qwen3.5 76.4% vs Qwen3.6 73.4%.
Qwen3.5 leads on Humanity's Last Exam · no tools — Qwen3.5 28.7% vs Qwen3.6 21.4%.
Qwen3.5 has a 1M context window; Qwen3.6 has 256k.