LLMs · Alibaba

Qwen3.5 2B

Name: Qwen3.5 2B
Author: Alibaba

Parameters: 2.3B
Architecture: Dense
Max context: 256K tokens
License: apache-2.0
Released: 2026-02

How to run it

Quickest · Ollama

One command downloads a ready-to-run version and starts a chat.

Get Ollama at ollama.com/download (opens in a new tab) · model page (opens in a new tab)

ollama run qwen3.5:2b

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

huggingface.co/unsloth/Qwen3.5-2B-GGUF (opens in a new tab)

ollama run hf.co/unsloth/Qwen3.5-2B-GGUF

Original weights · Hugging Face

Full-precision files from Alibaba, for fine-tuning, vLLM, Transformers.

huggingface.co/Qwen/Qwen3.5-2B (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from Alibaba. The weights themselves download from Hugging Face.

github.com/QwenLM/Qwen3.8 (opens in a new tab)

Qwen’s current repo: Qwen3.5 and Qwen3.6 redirect here.

git clone https://github.com/QwenLM/Qwen3.8

Links checked 2 days ago against Hugging Face, Ollama and GitHub.

Apps that run Qwen3.5 2B

Our pick for each kind of computer, assuming the model fits in its memory. Pick your own GPU or device in the Model Finder for an exact answer.

Mac (Apple silicon)

LM Studio

Runs the MLX build, typically 35–90% faster than GGUF for a model with 2.3B active parameters on Apple silicon.

Also good:

Windows PC, NVIDIA GPU

Ollama

One command with Ollama’s own build of this model.

Also good:

Windows PC, AMD Radeon

Lemonade

AMD’s own app: llama.cpp tuned for Radeon on Vulkan and ROCm.

Also good:

Linux, NVIDIA GPU

Ollama

One command with Ollama’s own build of this model.

Also good:

Linux server, 80 GB+ GPUs

vLLM

The full model fits in GPU memory at 8-bit, so it can batch many requests: best for serving a team or an app.

Also good:

App	Type	Opens it as	Systems	Engine	Start
LM Studio	Desktop app	GGUF or MLX	macOSWindowsLinux	llama.cpp, MLX	Open in LM Studio
Ollama	CLI + API	Ollama library	macOSWindowsLinux	llama.cpp, MLX	`ollama run qwen3.5:2b`
llama.cpp	CLI + API	GGUF	macOSWindowsLinux	llama.cpp	`llama-server -hf unsloth/Qwen3.5-2B-GGUF:Q4_K_M`
Jan	Desktop app	GGUF or MLX	macOSWindowsLinux	llama.cpp, MLX	Open in Jan
MLX LM	CLI + API	MLX	macOS	MLX	`mlx_lm.chat --model lmstudio-community/Qwen3.5-2B-MLX-4bit`
vLLM	Server	Original weights	WSLWindowsLinux	vLLM	`vllm serve Qwen/Qwen3.5-2B`
SGLang	Server	Original weights	WSLWindowsLinux	SGLang	`python3 -m sglang.launch_server --model-path Qwen/Qwen3.5-2B`
Docker Model Runner	CLI + API	GGUF	macOSWindowsLinux	llama.cpp, vLLM	`docker model run hf.co/unsloth/Qwen3.5-2B-GGUF:Q4_K_M`
Lemonade	Desktop app	GGUF	macOSWindowsLinux	llama.cpp, ONNX Runtime GenAI	`lemonade pull unsloth/Qwen3.5-2B-GGUF:Q4_K_M`
Unsloth Desktop	Desktop app	GGUF or MLX	macOSWindowsLinux	llama.cpp, MLX, stable-diffusion.cpp	Open in Unsloth Desktop
Atomic Chat	Desktop app	GGUF or MLX	macOSWindowsLinux	llama.cpp, MLX-VLM	Open in Atomic Chat
KoboldCpp	Desktop app	GGUF	macOSWindowsLinux	llama.cpp	Get KoboldCpp
Text Generation WebUI	Desktop app	GGUF	macOSWindowsLinux	llama.cpp, Transformers, ExLlamaV3	Get Text Generation WebUI
llamafile	CLI + API	GGUF	macOSWindowsLinux	llama.cpp	Get llamafile
Msty Studio	Desktop app	GGUF or MLX	macOSWindowsLinux	Ollama, llama.cpp, MLX	Open in Msty Studio
Llama (macOS)	Desktop app	GGUF	macOS	llama.cpp	Open in Llama (macOS)
RecurseChat	Desktop app	GGUF	macOS	llama.cpp	Open in RecurseChat
Inferencer	Desktop app	MLX	macOS	MLX	Get Inferencer
node-llama-cpp	Library	GGUF	macOSWindowsLinux	llama.cpp	Get node-llama-cpp
exo	Server	MLX	macOSLinux	MLX	Get exo
PocketPal AI	Mobile app	GGUF	iOSAndroid	llama.cpp	App Store Google Play
ChatterUI	Mobile app	GGUF	Android	llama.cpp	Download the APK
LLM Farm	Mobile app	GGUF	iOS	llama.cpp	Not on the App Store or TestFlight for now (checked 2026-11-09).

Which apps can open a model depends on the files published for it: GGUF for llama.cpp-based apps, MLX for Apple’s engine, the original weights for vLLM and SGLang. Compare every app.

Memory needed

Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps. Try another context length.

KV cache: hybrid: only the full-attention layers (1 in 4) keep a KV cache; the rest use linear attention.

Quantization	Quality	Weights	Total at 8K context	Total at 32K context	Download
`Q2_K`	Low	0.9 GB	1.6 GB	1.9 GB	GGUF (UD-Q2_K_XL) (opens in a new tab)
`Q3_K_M`	Low	1.1 GB	1.8 GB	2.1 GB	GGUF (opens in a new tab)
`Q4_K_M`	Good	1.4 GB	2.1 GB	2.4 GB	GGUF (opens in a new tab)
`Q5_K_M`	High	1.6 GB	2.3 GB	2.6 GB	GGUF (opens in a new tab)
`Q6_K`	High	1.9 GB	2.6 GB	2.9 GB	GGUF (opens in a new tab)
`Q8_0`	Very high	2.4 GB	3.1 GB	3.4 GB	GGUF (opens in a new tab)
`F16`	Full	4.5 GB	5.2 GB	5.5 GB	GGUF (opens in a new tab) Original weights (opens in a new tab)

Best hardware for Qwen3.5 2B

The least expensive GPU in our catalog that runs it without offloading is the Intel Arc A380 (launch price $139) at Q8_0.

Quality	Hardware	Quant	Memory	Speed	Runs
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking.	Intel Arc B580	`F16`	5.8 GB / 12.0 GB	44.2 tok/s	Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 3060 12GB	`F16`	5.8 GB / 12.0 GB	45.7 tok/s	Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 4070 SUPER	`F16`	5.8 GB / 12.0 GB	63.9 tok/s	Runs great
40Fair quality (C) · 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 4060	`Q8_0`	3.5 GB / 8.0 GB	74.9 tok/s	Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	AMD Radeon RX 9070 XT	`F16`	6.0 GB / 16.0 GB	69.9 tok/s	Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 4070 Ti SUPER	`F16`	6.0 GB / 16.0 GB	85.3 tok/s	Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 4080 SUPER	`F16`	6.0 GB / 16.0 GB	93.4 tok/s	Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 5060 Ti 16GB	`F16`	6.0 GB / 16.0 GB	56.8 tok/s	Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 5070 Ti	`F16`	6.0 GB / 16.0 GB	114 tok/s	Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 5080	`F16`	6.0 GB / 16.0 GB	122 tok/s	Runs great

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.