prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24

prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24, quantized with OptiQ by pty819 for Apple Silicon.

community

On disk0.6 GB

Memoryabout 2 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 pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24

OptiQ Code

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

optiq serve --model pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24
optiq code

OptiQ Lab

optiq lab --model pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24

Claude Code

optiq run claude -m pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24

Codex

optiq run codex -m pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24

OpenCode

optiq run opencode -m pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24

Mistral Vibe

optiq run vibe -m pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24

Hermes Agent

optiq run hermes -m pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24

OpenClaw

optiq run openclaw -m pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24

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("pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))

Details

Family	Qwen · Qwen docs
Architecture	Qwen3.5
Size on disk	0.6 GB
Memory needed	about 2 GB
Images	no, text only
Downloads	26 in the last month
Updated	2026-06-13
Hugging Face	pty819/prism-qwen3.5-reranker-0.8b-optiq-5bpw-cal24