majentik
/

Leanstral-TurboQuant-MLX-2bit

kv-cache-quantization

weight-quantization

theorem-proving

Mixture of Experts

Model card Files Files and versions

Leanstral-TurboQuant-MLX-2bit

Commit History

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

9a7963e
verified

majentik commited on 26 days ago

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

db6fcae
verified

majentik commited on Apr 17

Add MLX quantized model weights

9fbfbd1
verified

majentik commited on Apr 13

Add model card (weights pending mlx_lm mistral3 architecture support)

642eaf9
verified

majentik commited on Apr 13

initial commit

af6ecc8
verified

majentik commited on Apr 13