lna-lab

3 projects47 starsacross 1 gallery

View profile on GitHub

blackwell-geforce-nvfp4-gemm

NVFP4 inference on Blackwell GeForce (RTX 5090/5080/5070 Ti/RTX PRO 6000) — SM120 patches for vLLM + FlashInfer + CUTLASS. 175 tok/s on Qwen3.6-35B MoE. Maintained by lna-lab on GitHub, where it has earned 22 stars from the community.

gemma4-12b-vllm-sm120

Reproducible recipe: serve abliterated Gemma-4-12B (gemma4_unified) at 50-118 tok/s on no-NVLink Blackwell (SM120) via vLLM nightly + ModelOpt FP8/NVFP4 + MTP spec-decode. Maintained by lna-lab on GitHub, where it has earned 15 stars from the community.

GGUF-to-NVFP4-SM120

Lna-Lab production pipeline: GGUF -> modelopt-format NVFP4 + working MTP head for vLLM on RTX PRO 6000 Blackwell (SM120). Stages 2 (NVFP4) and 3 (MTP graft) are Lna-Lab originals; stage 1 (GGUF->bf16) reuses li-yifei/gguf-to-nvfp4. Maintained by lna-lab on GitHub, where it has earned 10 stars from the community.