e-accelerate

Popular repositories Loading

1. pd-bridge pd-bridge Public

   Forked from chadhurley25075-png/pd-bridge

   Heterogeneous prefill/decode for DeepSeek-V4-Flash: CUDA prefill (DGX Spark, vLLM) -> Metal decode (Mac Studio, oMLX) over plain 10GbE

   Python 1

2. Qwen3.8-27B-DFlash2-Triton-Blackwell Qwen3.8-27B-DFlash2-Triton-Blackwell Public

   Qwen3.8-27B on SGLang for Blackwell SM120: DFlash2 block-16 + triton. 268 tok/s on one RTX PRO 5000 (48GB), full 262K context. Measured, with honest sample sizes.

   Shell

3. GLM-5.3-Flash-EXL3-2x-DGX-Sparks GLM-5.3-Flash-EXL3-2x-DGX-Sparks Public

   Forked from MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Sparks

   GLM-5.3 Flash EXL3 for 2-4x DGX Sparks

   Python

4. glm53-flash-2x-recipe glm53-flash-2x-recipe Public

   GLM-5.3-Flash EXL3 on 2x GB10 (ASUS GX10 + DGX Spark), dual-rail: tuned, measured recipe on MiaAI-Lab's kit

   Python