FareedKhan-dev/kimi-k3-in-c
kimi-k3-in-c is a portable C99 inference engine designed to run the Kimi K3 2.78-trillion-parameter LLM on a single CPU with minimal RAM, focusing on extreme memory efficiency without GPUs or exter...
Awesome Infra for AI › Inference Optimization
insight-platform/Savant
Savant is a high-level, open-source framework designed to accelerate the development of real-time, streaming, and efficient multimedia AI applications, primarily focusing on computer vision and video analytics. It leverages the Nvidia DeepStream SDK to provide an abstraction layer that simplifies the creation of dynamic, fault-tolerant inference pipelines for Nvidia GPUs (Turing, Volta, Ampere, Ada, Hopper, Blackwell) and Jetson devices (Xavier, Orin, Nano, NX, AGX). Savant enables developers to build production-ready applications with maximum performance on Nvidia equipment, reducing time-to-market. It supports various computer vision tasks such as detection, classification, segmentation, tracking, and custom pre- and post-processing. The framework allows for the creation of heterogeneous pipelines with different models and data sources, supports hybrid edge/datacenter applications, and integrates with OpenTelemetry and Prometheus for monitoring and tracing. By abstracting low-level DeepStream complexities, Savant allows engineers to focus on computer vision logic rather than infrastructure, providing a robust architecture for distributed edge/datacenter computer vision applications.
https://github.com/insight-platform/Savant
Also in Inference Optimization
zilliztech/GPTCache
GPTCache is a library that creates a semantic cache for LLM queries, significantly reducing API costs and improving response times by storing and reusing previous LLM responses.
vllm-project/vllm-ascend
vLLM Ascend is a community-maintained hardware plugin for seamlessly running vLLM and various large language models on Ascend NPUs, optimizing inference performance.
youssofal/MTPLX
MTPLX is a macOS-native inference engine for Apple Silicon that leverages multi-token prediction (MTP) for accelerated local LLM serving, offering significant speed improvements.
open-compress/claw-compactor
Claw Compactor is an LLM token compression engine that uses a 14-stage Fusion Pipeline for content-aware, reversible compression to reduce LLM inference costs and optimize context windows.
siliconflow/onediff
OneDiff is an acceleration library for diffusion models, providing out-of-the-box performance optimizations for popular UIs and libraries like Hugging Face Diffusers and ComfyUI.
ovg-project/kvcached
kvcached is a KV cache library that introduces virtual memory abstraction for LLM serving on shared GPUs, enabling elastic and demand-driven KV cache allocation for improved GPU utilization under d...
nobodywho-ooo/nobodywho
NobodyWho is an efficient, on-device inference engine that enables local execution of LLMs and SLMs across various platforms without requiring API keys.