repoGitHubTrust 82 · PrimaryPublished 3d agoLive · yesterday

marzukia/qMLX

qMLX: Custom inference engine for Qwen 3.5 122B on Apple Silicon, extending MLX with hybrid attention support, SSD-backed KV cache, and RYS layer duplication for efficient large-model serving on consumer hardware.

Lineage graph

Paper → model → repo connections mined from source citations (Tier-1 exact match).

Why these links exist

Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.

PossiblePossibly related (embedding) · 60%A barebones CPU-only inference engine for Qwen 3, written from scratch in pure C →
PossiblePossibly related (embedding) · 54%Hardware startup unveils inference accelerator →
PossiblePossibly related (embedding) · 53%OpenAI and Broadcom unveil LLM-optimized inference chip →
PossiblePossibly related (embedding) · 51%Gemma 4 12B - MLX Kernel →
PossiblePossibly related (embedding) · 51%OpenAI and Broadcom announce chip designed for LLM inference at scale →

Covers

newsA barebones CPU-only inference engine for Qwen 3, written from scratch in pure C newsHardware startup unveils inference accelerator newsOpenAI and Broadcom unveil LLM-optimized inference chip newsGemma 4 12B - MLX Kernel newsOpenAI and Broadcom announce chip designed for LLM inference at scale

Related across the graph

newsOpenAI and Broadcom announce chip designed for LLM inference at scale newsOpenAI and Broadcom unveil LLM-optimized inference chip newsHardware startup unveils inference accelerator newsA barebones CPU-only inference engine for Qwen 3, written from scratch in pure C newsGemma 4 12B - MLX Kernel

Topics