lEWFkRAD/qwen38-rtx-pro-6000
Reproducible Qwen3.8 deployments on one RTX PRO 6000: Flash-Next W4 PLE, SGLang patches, benchmarks, and field reports.
GitHub repository with 7 stars and 1 forks.
Language: Python
Topics: benchmarks, blackwell, llm-inference, quantization, qwen, sglang