majentik
/

Qwen3.5-27B-RotorQuant-MLX-8bit

kv-cache-quantization

8-bit precision

Model card Files Files and versions

Qwen3.5-27B-RotorQuant-MLX-8bit

Commit History

docs: Tier 2 polish — variant matrix + quant trade-off

8d8a577
verified

majentik commited on 25 days ago

chore(card): enrich YAML frontmatter (pipeline_tag, language, library_name, inference)

f2a537c
verified

majentik commited on Apr 17

Add MLX 8-bit quantized Qwen3.5-27B with RotorQuant KV cache

e4515b3
verified

majentik commited on Apr 13

initial commit

66efb42
verified

majentik commited on Apr 13