← Back to directory
Alibaba · Released 2026-04
Qwen3.6 35B
Alibaba's April 2026 release; the 35B variant punches above its weight on coding and vision benchmarks, with a permissive license and strong Chinese support.
QwenCommercial use OK👁 Visiongeneralcodevision
Params (max)
35B
Variants
72B / 35B / 14B / 7B
Context window
128K tokens
MMLU
84.2
HumanEval
87.2
GSM8K
91.4
Min VRAM (fp16, smallest variant)
24GB
Smallest Q4 GGUF
~4GB
Languages supported
119
Pros
- ✓Best CN/EN bilingual model
- ✓Vision included
- ✓Strong code performance
Cons
- ×Smaller English fine-tune community than Llama
- ×Some vision features Chinese-tuned
Highlights
- ●119 languages supported
- ●Strong vision-language results
- ●Permissive Qwen license
Where to download
Hugging Face: Qwen/Qwen3.6-35B
Or via Ollama (
ollama pull qwen-3.6-35b) or LM Studio's in-app browser.Homepage: https://qwenlm.ai
Related reading
Best Open Source LLMs 2026: Honest Picks by Use Case
Which open-source LLM should you actually run in 2026? Honest picks by use case — frontier reasoning, coding, RAG, edge devices, multilingual.
Open Source LLMs vs Claude / GPT in 2026: When Does Open Win?
Open-source LLMs caught up to GPT-4 in 2024 and Claude Opus in 2026 — but should you actually switch? Cost, quality, latency, privacy compared.
Where to Download Open LLM Weights Safely in 2026
Hugging Face is the default but not the only option. Mirrors, torrents, official sources, and how to verify checksums.