A developer shared Booth on Hacker News as a Show HN project, a lightweight checkpoint library for saving LLM outputs. It is an early-stage open-source tool aimed at developers who need to persist and resume LLM workflow state.
A lightweight web client for Claude and LMStudio that maintains memory across sessions, allowing users to have persistent conversations with AI models. It solves the problem of context loss in chat interfaces by storing and recalling conversation history.
A lightweight web client for Claude and LMStudio that maintains memory across sessions, allowing users to have persistent conversations with AI models. It solves the problem of context loss in chat interfaces by storing and recalling conversation history.
MindCache is an open-source memory system that gives LLMs persistent, structured memory across conversations. It helps developers build agents that remember context, user preferences, and past interactions, solving the problem of LLMs having no long-term memory.
MindCache is an open-source memory system that gives LLMs persistent, structured memory across conversations. It helps developers build agents that remember context, user preferences, and past interactions, solving the problem of LLMs having no long-term memory.
Coalent is an LLM answer cache that automatically invalidates cached responses when the underlying source documents change. It solves the problem of stale or outdated LLM-generated answers by tying cache validity to document updates, useful for developers building document-grounded LLM applications.
Coalent is an LLM answer cache that automatically invalidates cached responses when the underlying source documents change. It solves the problem of stale or outdated LLM-generated answers by tying cache validity to document updates, useful for developers building document-grounded LLM applications.
A developer released open-source AMDGCN kernels aimed at optimizing LLM inference. The kernels target AMD hardware to improve performance.
A developer released open-source AMDGCN kernels aimed at optimizing LLM inference. The kernels target AMD hardware to improve performance.
OpenSmith is an open-source tool for tracing LLM pipelines locally, without sending data to any cloud service. It helps developers debug and monitor LLM applications by providing a local trace viewer, solving the problem of relying on external services for observability.
OpenSmith is an open-source tool for tracing LLM pipelines locally, without sending data to any cloud service. It helps developers debug and monitor LLM applications by providing a local trace viewer, solving the problem of relying on external services for observability.