Qwen3.6-35B-A3B-OptiQ-MLX
Qwen/Qwen3.6-35B-A3B, quantized with OptiQ by UcoWorker for Apple Silicon; mixture-of-experts.
communityMoEOn disk22.0 GB
Memoryabout 27 GB
Use it
Install
pip install mlx-optiqServe it
An OpenAI- and Anthropic-compatible server on this Mac. Every app below can use it.
optiq serve --model UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLXOptiQ Code
The terminal coding agent uses whatever model optiq serve is running.
optiq serve --model UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLX
optiq codeOptiQ Lab
optiq lab --model UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLXClaude Code
optiq run claude -m UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLXCodex
optiq run codex -m UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLXOpenCode
optiq run opencode -m UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLXMistral Vibe
optiq run vibe -m UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLXHermes Agent
optiq run hermes -m UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLXOpenClaw
optiq run openclaw -m UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLXoptiq run starts optiq serve if it is not running, sets the agent up for this session only, and stops the server when the agent exits.
mlx-lm, text only
from mlx_lm import load, generate
model, tokenizer = load("UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLX")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))Details
| Family | Qwen · Qwen docs |
|---|---|
| Architecture | Qwen3.5 MoE |
| Base model | Qwen/Qwen3.6-35B-A3B |
| Size on disk | 22.0 GB |
| Memory needed | about 27 GB |
| Images | no, text only |
| Downloads | 26 in the last month |
| Updated | 2026-04-16 |
| Hugging Face | UcoWorker/Qwen3.6-35B-A3B-OptiQ-MLX |