Qwen3.5-2B
Qwen3.5-2B is an open-weights language model from Qwen, released February 24, 2026. It has 2.3B parameters (dense). The context window is 256K tokens.
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Qwen3.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to …
Check it against your own PC →
Can you run Qwen3.5-2B locally?
The smallest common GPU it runs on is 6 GB of VRAM (with 32 GB of RAM): Q6_K, fully on GPU, roughly 58 tokens per second. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.7 GB | – | ~58 |
| 8 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.7 GB | – | ~81 |
| 12 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.7 GB | – | ~100 |
| 16 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.7 GB | – | ~140 |
| 24 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.7 GB | – | ~230 |
| 32 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.7 GB | – | ~370 |
Ready-made quantizations on Hugging Face: mradermacher/Qwen3.5-2B-Base-i1-GGUF · titan0115/Qwen3.5-2B-Base · mradermacher/Qwen3.5-2B-Base-GGUF · Mungert/Qwen3.5-2B-GGUF · ZuzeTt/Qwen3.5-2B-imatrix-GGUF · aashish1904/Qwen3.5-2B-Base-GGUF
Benchmark results
Epoch AI has not published results for Qwen3.5-2B yet. New models are usually evaluated within days to weeks.
Links
Announcement · Hugging Face: Qwen/Qwen3.5-2B-Base
More from Qwen
Qwen3.8 Omni Flash · Qwen3.8 Max (0902) · Qwen3.8 Flash · Qwen3.8 Flash Next · Qwen 3.8 27B · Qwen3.8-2.4T-A95B · Qwen 3.8 Max · Qwen3.7 Flash