Hy-MT2-7B-optiq-3_5bpw
Hy-MT2-7B-optiq-3_5bpw, quantized with OptiQ by QianheChen01 for Apple Silicon.
communityOn disk4.1 GB
Memoryabout 6 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 QianheChen01/Hy-MT2-7B-optiq-3_5bpwOptiQ Code
The terminal coding agent uses whatever model optiq serve is running.
optiq serve --model QianheChen01/Hy-MT2-7B-optiq-3_5bpw
optiq codeOptiQ Lab
optiq lab --model QianheChen01/Hy-MT2-7B-optiq-3_5bpwClaude Code
optiq run claude -m QianheChen01/Hy-MT2-7B-optiq-3_5bpwCodex
optiq run codex -m QianheChen01/Hy-MT2-7B-optiq-3_5bpwOpenCode
optiq run opencode -m QianheChen01/Hy-MT2-7B-optiq-3_5bpwMistral Vibe
optiq run vibe -m QianheChen01/Hy-MT2-7B-optiq-3_5bpwHermes Agent
optiq run hermes -m QianheChen01/Hy-MT2-7B-optiq-3_5bpwOpenClaw
optiq run openclaw -m QianheChen01/Hy-MT2-7B-optiq-3_5bpwoptiq 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("QianheChen01/Hy-MT2-7B-optiq-3_5bpw")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))Details
| Family | Hunyuan V1 Dense |
|---|---|
| Architecture | Hunyuan V1 Dense |
| Size on disk | 4.1 GB |
| Memory needed | about 6 GB |
| Images | no, text only |
| Downloads | 114 in the last month |
| Updated | 2026-08-24 |
| Hugging Face | QianheChen01/Hy-MT2-7B-optiq-3_5bpw |