aharshms/simulator
Decoder only, Encoder only, Diffusion model simulation for SLO, Memory, Infra calculations. For model inference and training
Repositories
aharshms/done-and-dusted
A personal to do list application. It shows to do list in a funny way each time you open a new tab in chrome.
aharshms/awesome-gpu-virtualization-methods
A curated awesome-list of techniques, tooling, and research for virtualizing and sharing GPU resources across workloads and environments—covering approaches for partitioning, multiplexing, VM-based acceleration, and remote/disaggregated GPU access with links to key papers, docs, and open-source projects.
aharshms/OSBRtest
Bud runtime is a very opinionated End to End compound AI lifecycle management platform. The platform is intended to be the foundational layer for any organisation irrespective of its size and skillset looking to adopt GenAI
aharshms/docs
aharshms/vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
aharshms/Oidc.Server
Flexible OpenID Connect and OAuth 2.0 server-side implementation for modern ASP.NET projects
aharshms/Awesome-LLM-Inference
📖A curated list of Awesome LLM Inference Paper with codes, TensorRT-LLM, vLLM, streaming-llm, AWQ, SmoothQuant, WINT8/4, Continuous Batching, FlashAttention, PagedAttention etc.
aharshms/Awesome-Efficient-LLM
A curated list for Efficient Large Language Models