Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw

ornith-ai/Ornith-1.5-35B-A3B, quantized with OptiQ by ahmedihamdy for Apple Silicon; mixture-of-experts, takes images.

communityvisionMoEMTP head

On disk27.5 GB

Memoryabout 33 GB

Imagesyes

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 ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw

OptiQ Code

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

optiq serve --model ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw
optiq code

OptiQ Lab

optiq lab --model ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw

Claude Code

optiq run claude -m ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw

Codex

optiq run codex -m ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw

OpenCode

optiq run opencode -m ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw

Mistral Vibe

optiq run vibe -m ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw

Hermes Agent

optiq run hermes -m ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw

OpenClaw

optiq run openclaw -m ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw

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("ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))

Details

Family	Qwen · Qwen docs
Architecture	Qwen3.5 MoE
Base model	ornith-ai/Ornith-1.5-35B-A3B
Size on disk	27.5 GB
Memory needed	about 33 GB
Images	yes, through optiq serve and the Lab
MTP head	included
Downloads	451 in the last month
Updated	2026-08-29
Hugging Face	ahmedihamdy/Ornith-1.5-35B-A3B-OptiQ-static-5.5bpw