One-flag setup for local Qwen3.8: builds a CUDA llama.cpp with HauhauCS FastMTP speculative decoding, downloads model/template/sidecar, serves on localhost — PowerShell & bash.
-
Updated
Aug 25, 2026 - Shell
#
fastmtp
Here are 2 public repositories matching this topic...
Benchmarks and serving notes for Qwen3.5-27B fine-tune on Apple Silicon via MLX. Companion to dgx-spark-nvfp4-serving.
benchmarks quantization dora mlx fine-tuning apple-silicon mac-studio llama-cpp llm-inference speculative-decoding mlx-lm qwen3 qwen3-5 m3-ultra fastmtp
-
Updated
Apr 21, 2026 - Python
Add this topic to your repo
To associate your repository with the fastmtp topic, visit your repo's landing page and select "manage topics."