Wësche(@WescheNex1q):Running @ViC305's GLM-5.3 753b recipe on 4 Sparks took four fixes on our cluster: build the CUDA kernels before loading (warm-up OOM) one NFS copy of the 319GB pack hand-written hosts (cloned machine-ids) swap off Result: 84.4. All scripts are open: https://t.co/c3qGayNxpS Credits: @ViC305 (quant + recipe), TensorFold, the DFlash2 drafter @Zai

Wësche

@WescheNex1q

Day time artist and night time AI enthusiast. Building & benchmarking frontier LLMs on 4x DGX Spark clusters + Mac. Creator of Vesica Studio. Houston

参加 January 2013

731 フォロー中 3.1K ファン

Wësche@WescheNex1q

2026.10.08 19:12

Running @ViC305's GLM-5.3 753b recipe on 4 Sparks took four fixes on our cluster: build the CUDA kernels before loading (warm-up OOM) one NFS copy of the 319GB pack hand-written hosts (cloned machine-ids) swap off Result: 84.4. All scripts are open: Credits: @ViC305 (quant + recipe), TensorFold, the DFlash2 drafter @Zai_org