Algocean-Qwen3.6-35B-A3B-mlx-OptiQ

Qwen/Qwen3.6-35B-A3B, quantized with OptiQ by aisamdasu for Apple Silicon; mixture-of-experts.

communityMoE

On disk36.0 GB

Memoryabout 43 GB

Use it

Install

pip install mlx-optiq

Serve it

An OpenAI- and Anthropic-compatible server on this Mac. Every app below can use it.

optiq serve --model aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ

OptiQ Code

The terminal coding agent uses whatever model optiq serve is running.

optiq serve --model aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ
optiq code

OptiQ Lab

optiq lab --model aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ

Claude Code

optiq run claude -m aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ

Codex

optiq run codex -m aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ

OpenCode

optiq run opencode -m aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ

Mistral Vibe

optiq run vibe -m aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ

Hermes Agent

optiq run hermes -m aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ

OpenClaw

optiq run openclaw -m aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ

optiq 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("aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ")
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	36.0 GB
Memory needed	about 43 GB
Images	no, text only
Downloads	31 in the last month
Updated	2026-07-08
Hugging Face	aisamdasu/Algocean-Qwen3.6-35B-A3B-mlx-OptiQ