mlx-community/Qwen3.5-4B-MLX-4bit
Run locally on Apple devices with Mirai
- Type
- local
- From
- Alibaba
- Quantization
- MLX 4-bit
- Parameters
- 4B
- Size
- 2.2 GB
1Choose framework
2Run the following command to install Mirai SDK
spm https://github.com/trymirai/uzu.git
3Apply code
| 1 | import Uzu |
| 2 | |
| 3 | public func runChat() async throws { |
| 4 | let engineConfig = EngineConfig.create() |
| 5 | let engine = try await Engine.create(config: engineConfig) |
| 6 | |
| 7 | guard let model = try await engine.model(identifier: "mlx-community/Qwen3.5-4B-MLX-4bit") else { |
| 8 | return |
| 9 | } |
| 10 | for try await update in try await engine.download(model: model).iterator() { |
| 11 | print("Download progress: \(update.progress())") |
| 12 | } |
| 13 | |
| 14 | let messages = [ |
| 15 | ChatMessage.system().withText(text: "You are a helpful assistant"), |
| 16 | ChatMessage.user().withText(text: "Tell me a short, funny story about a robot") |
| 17 | ] |
| 18 | let session = try await engine.chat(model: model, config: .create()) |
| 19 | let stream = await session.replyWithStream(input: messages, config: .create()) |
| 20 | var message: ChatMessage? = nil |
| 21 | for try await update in stream.iterator() { |
| 22 | switch update { |
| 23 | case .replies(let replies): |
| 24 | message = replies.last?.message |
| 25 | case .error(let error): |
| 26 | print("Error: \(error)") |
| 27 | } |
| 28 | } |
| 29 | print("Text: \(message?.text() ?? "empty")") |
| 30 | } |
Other local models from Alibaba
trymirai/Qwen3.5-0.8B-M
438.0 MB800M
trymirai/Qwen3.5-0.8B-L
785.3 MB800M
trymirai/Qwen3.5-2B-M
1.0 GB2B
trymirai/Qwen3.5-2B-L
1.9 GB2B
trymirai/Qwen3.5-4B-M
2.3 GB4B
trymirai/Qwen3.5-4B-L
4.2 GB4B
Qwen/Qwen3-0.6B
1.1 GB600M
Qwen/Qwen3-0.6B-MLX-4bit
313.3 MB600M
Qwen/Qwen3-0.6B-MLX-8bit
597.4 MB600M
Qwen/Qwen3-1.7B
3.2 GB1.7B
Qwen/Qwen3-1.7B-MLX-4bit
883.2 MB1.7B
Qwen/Qwen3-1.7B-MLX-8bit
1.7 GB1.7B
Qwen/Qwen3-4B
7.5 GB4B
Qwen/Qwen3-4B-Instruct-2507
7.5 GB4B
Qwen/Qwen3-4B-Thinking-2507
7.5 GB4B
Qwen/Qwen3-4B-MLX-4bit
2.0 GB4B
Qwen/Qwen3-4B-MLX-8bit
3.9 GB4B
Qwen/Qwen3-8B
15.3 GB8B
Qwen/Qwen3-8B-MLX-4bit
4.1 GB8B
Qwen/Qwen3-8B-MLX-8bit
7.9 GB8B
Qwen/Qwen3-14B
27.5 GB14B
Qwen/Qwen3-14B-MLX-4bit
7.3 GB14B
Qwen/Qwen3-14B-MLX-8bit
14.2 GB14B
Qwen/Qwen3-32B
61.0 GB32B
Qwen/Qwen3-32B-MLX-4bit
16.2 GB32B
Qwen/Qwen3-32B-MLX-8bit
31.5 GB32B
Qwen/Qwen3.5-0.8B
1.4 GB800M
mlx-community/Qwen3.5-0.8B-MLX-4bit
424.5 MB800M
mlx-community/Qwen3.5-0.8B-MLX-8bit
783.0 MB800M
Qwen/Qwen3.5-2B
3.5 GB2B
mlx-community/Qwen3.5-2B-MLX-4bit
1.0 GB2B
mlx-community/Qwen3.5-2B-MLX-8bit
1.9 GB2B
Qwen/Qwen3.5-4B
7.9 GB4B
mlx-community/Qwen3.5-4B-MLX-8bit
4.2 GB4B
Qwen/Qwen3.5-9B
16.7 GB9B
mlx-community/Qwen3.5-9B-MLX-4bit
4.7 GB9B
mlx-community/Qwen3.5-9B-MLX-8bit
8.9 GB9B
Qwen/Qwen3.5-27B
50.1 GB27B
mlx-community/Qwen3.5-27B-4bit
14.1 GB27B
mlx-community/Qwen3.5-27B-8bit
26.6 GB27B
Qwen/Qwen3.6-27B
53.5 GB27B