Choose a recipe, stage its weights, and run the command from the matching GB10SOR checkout. Multi-Spark lanes also need the private local cluster configuration described in the repository. Setup guide →
Use the repository quickstart for DGX OS or NixOS. Multi-Spark lanes also require the matching private local cluster configuration.
nvidia
GLM-5.3-Flash-NVFP4
nvidia/GLM-5.3-Flash-NVFP4
One-command Nix ready
Engine
vLLM 0.28.1rc1.dev580+g385dce36b
Served context
32,768 tokens
Request limit
4
Parallelism
TP4 / PP1
Checkpoint weights
190.4 GiB
Recipe YAML
Recipe & author credits
Use the exact checkpoint revision and deployment profile. Combining recipes is a planning choice; it does not establish combined compatibility.
Contributors: NVIDIA model/recipe team; vLLM and SGLang upstream