Popular repositories Loading

1. llama.cpp-turboq-mtp llama.cpp-turboq-mtp Public

   Qwen4-Exp (Qwen3.8-Flash-Next) SWA + MTP + PLE-on-iswa — bounded deep-context decode w/ long-range recall, TBQ4 4.25 bpv KV, DSpark + RotorQuant. Main (master) = Qwen4-Exp build; the Qwen35 SWA hyb…

   C++ 95 14

2. Strata-DSV4.1 Strata-DSV4.1 Public

   DeepSeek-V4.1-Flash on one gaming PC (RTX 4090 + 90 GB RAM): ~250 tok/s exact prefill, ~26 tok/s decode. 152 GiB GSQ mixed-tier experts across VRAM / pinned RAM / NVMe. Built on Strata.

   C++ 14 1

3. GLM-4.7-Flash-Rapport GLM-4.7-Flash-Rapport Public

   Fast GLM-4.7-Flash wrapper for Claude Code with Google Search MCP and Vision Proxy

   JavaScript 10 5

4. llama.cpp-TurboQuant-DSV4 llama.cpp-TurboQuant-DSV4 Public

   llama.cpp fork: TurboQuant KV (TBQ3/TBQ4) + RotorQuant + MTP + DSV4 native TBQ4 KV via dequant-at-read (1M context on 96GB, zero swap)

   C++ 2

5. ComfyUI-H3O ComfyUI-H3O Public

   MiniMax H3 node pack by Indra's Mirror — H3 Prompt Enhancer + H3 Aspect Ratio Detector

   Python 1

6. CM-Building-Solutions CM-Building-Solutions Public

   C&M Building Solutions - Quality Carpentry & Cabinet Making Melbourne

   CSS