How to use Avifenesh/GLM-5.3-Flash-NVFP4 with memra:
# memra serves NVIDIA Blackwell workstation and consumer cards (sm_120a), with a # compile-gated Hopper lane. Prebuilt binaries need Linux x86_64 and driver 580+, # and no CUDA toolkit. curl -fsSL https://raw.githubusercontent.com/avifenesh/memra/main/tools/install.sh | sh
# One chat-templated generation. In a repo with several GGUF files, append # :<substring> to choose one, for example hf:Avifenesh/GLM-5.3-Flash-NVFP4:Q4_K_M MEMRA_CHAT=1 run-gen hf:Avifenesh/GLM-5.3-Flash-NVFP4 --prompt "Explain KV caches in one sentence."
# Or an OpenAI-compatible server on 127.0.0.1:8080. MEMRA_MODELS="model=hf:Avifenesh/GLM-5.3-Flash-NVFP4" memra-server
The community tab is the place to discuss and collaborate with the HF community!
