Turn local language models into Jev-style structured decision models. Get results from text and images with prefill alone—no token-by-token decoding required.
7 daysStars49024h gain+6Score2.1
Search
Find trending repositories by name or description.
5 results for “prefill”
RankRepository7 daysStars24h gainScore
DeepSeek-V4.1-Flash on 2x NVIDIA DGX Spark with the TensorFold engine: 1.8-1.9x code, 2.5x multi-stream, 1.7-1.95x prefill vs the vLLM kit, exact speculative decoding. Work in progress.
7 daysStars14124h gain0Score0.1
CLI tool to automatically prime a Lancache with Steam games
7 daysStars42924h gain-Score0.0
W8A8/W4A8 inference + optimized SDPA on Apple Silicon — unlocking unused INT8 TensorOps in M5 for 1.2–1.9× faster LLM prefill, plus FlashInfer-inspired GQA decode attention for up to 1.6× SDPA speedup, built as MLX custom primitives.
7 daysStars43124h gain-Score0.0
Entrpi/ds4, a Blackwell CUDA perf fork of antirez/ds4 on NVIDIA DGX Spark: one-command install, ~3x upstream prefill, ~1.5x decode, DSpark, and full continuous batch support
7 daysStars18524h gain-Score0.0