今日已更新 373 条资讯 | 累计 24659 条内容
关于我们

标签：#opensource

找到 1622 篇相关文章

AI 资讯

Hillock: A brain-inspired, CPU-bound memory gate for local LLMs

Hi everyone, I've been hacking on a local personal memory system called Hillock . Honestly, it's very much a work in progress and it isn't some flawless breakthrough, but I wanted to see if we could build a lightweight, completely offline memory layer for local LLMs without the overhead of running a heavy neural vector database or wasting precious VRAM. The project is named after the biological Axon Hillock —the exact gatekeeper region of a human neuron that sums up incoming electrical charges and decides whether to fire (open the gate) or remain silent (block). How the architecture works: The Ground Truth (SQLite) : Stores hard facts as simple database triples (Subject-Predicate-Object) so the system has a solid symbolic foundation. The Synapses (Hebbian Plasticity) : Tracks which concepts co-occur during a conversation to dynamically build gradient-free associative weights. The Context (Hyperdimensional Computing) : Maintains a 10,000-dimensional leaky context vector that rolls, binds, and accumulates history. This helps the system resolve pronouns (like "he/she") and decide when to block a query to prevent hallucinations. The Honest Benchmarks (Yes, it breaks!) I wrote a tough, 30-sentence scientific benchmark with complex sentence structures and hard negatives (like asking what Einstein discovered when the text only mentions Curie discovering radioactivity and Einstein working with her). Running Qwen 1.5B locally on my computer, here is how it actually did: Extraction Precision : 10.6% Extraction Recall : 22.7% Retrieval Accuracy : 30.0% Gate Accuracy : 30.0% Why are these scores low? Because a tiny 1.5B model completely trips over complex English grammar during ingestion (it gets confused and creates weird predicates). However, the actual HDC vector-matching itself is incredibly stable. I enforce a Constant-Component-Count of exactly 3 components per fact, which balances the vector norms and keeps retrieval highly reliable once the facts are actually in the dat

2026-06-15 原文 →
AI 资讯

Making a fleet of self-hosted LLM agents trustworthy

Originally published at llmkube.com/blog/making-self-hosted-llm-agents-trustworthy . Cross-posted here for the dev.to audience. Running a single local LLM node is a solved problem. You write an InferenceService, the operator schedules it, llama.cpp or MLX serves it, and you get an OpenAI-compatible endpoint. We have been doing that for months. Running a fleet of them is where it stops being easy. My fleet is heterogeneous on purpose: CUDA pods in the cluster, and Apple Silicon Macs sitting off-cluster on the homelab network, each one running two separate agents (one for inference, one for the agentic coding harness). The day I shipped 0.8.4 to that fleet, I learned exactly how it does not scale. I updated each Mac by hand. The control plane had no idea what version any agent was running. And the launchd reload I used to restart an agent was a silent no-op on an already-loaded service, so the old binary kept running while I believed I had updated it. I found that out by hand-inspecting a process tree. Three machines made it annoying. Thirty would make it impossible, and the whole pitch for sovereign, on-prem AI is that you run a lot more than three. So the last stretch of work on LLMKube was not about a faster runtime or a bigger model. It was about making the fleet trustworthy : able to update itself safely, and unable to lie to the control plane about its own state. Here is what that took. Helm and brew for the edge The fix is a new cluster-scoped CRD, AgentRelease , and a self-update path in the agents themselves. You describe the release you want once, the operator rolls it out, and the agents pull and apply it. The design borrows directly from prior art that already solved this for Kubernetes nodes: Rancher's system-upgrade-controller, k0s autopilot's per-platform SHA-256 staging, and Teleport's outbound-only poll model. The properties that make it safe to leave running: Declarative and approved. An AgentRelease names the agent, the version, and the per-platform

2026-06-15 原文 →
AI 资讯

Why I Built a New Memory Plugin for Hermes Agent

Hermes Agent already has memory, and that matters. It keeps local context, it improves over time, and it works without forcing you into a cloud service. It also supports several external memory providers. I still built hermes-mempalace , because none of the existing options fit my setup quite right. I wanted something: local-first isolated by Hermes profile verbatim, not just extracted facts easy to inspect on disk simple enough to trust over time That last part is the important one. I did not want a memory layer that turns conversations into an opaque pile of embeddings or summaries you cannot really audit. I wanted actual transcripts, mined into a readable structure, with no hidden server in the middle. Why the existing options were not enough Hermes already gives you a few paths: built-in memory and session context external providers for different use cases enough flexibility to adapt, if you are willing to bend your workflow around them And to be clear, some of those options are good. But ... I run Hermes on a headless machine at home. And I use separate profiles for different contexts. And I do not want conversation content depending on a cloud API or a separate service unless there is a very good reason. So, the best fit had to check a few boxes: [x] no API key [x] no external server [x] no extra runtime I did not already want/install [x] storage isolated by HERMES_HOME [x] memory we can actually read later That let to MemPalace , or https://mempalaceofficial.com/ (hopefully, that's the right one!) What hermes-mempalace does hermes-mempalace wires MemPalace into the Hermes memory provider interface. It follows the same lifecycle as the rest of Hermes memory providers: system_prompt_block() adds a short memory reminder to the prompt. prefetch() can run a MemPalace search before the first model call. sync_turn() buffers completed turns without slowing the chat loop. on_session_end() writes buffered turns to markdown and mines them into the palace. shutdown() flu

2026-06-15 原文 →
AI 资讯

The Only Productivity Hack That Actually Worked for Me

The Only Productivity Hack That Actually Worked for Me I've tried every website blocker in existence. They all have the same flaw: one click and they're off. That settings toggle might as well not exist when the urge to procrastinate hits. I needed something I couldn't override. So I built the opposite. The Problem Every blocker follows the same pattern: it blocks, you unblock, you procrastinate. The issue isn't discipline — it's that your future self and your present self want different things. And that future self will happily undo anything you set up. What I needed was a system where both selves agreed on the rules upfront and then neither could break them. The Solution kblocker is a Linux kernel module that hooks into netfilter and drops TCP connections to whatever sites I configure. The key feature isn't the blocking — it's how you turn it off. When blocking is enabled, kblocker generates a 128-bit key. With PGP mode, it automatically encrypts that key to people I trust and then erases it from kernel memory. The raw key no longer exists anywhere on my system. To disable the blocker, one of those people has to decrypt it and send it back to me. I outsourced my willpower. When I want to focus, I run: sudo kblockerctl enable 120 Two hours of blocked distractions. If I feel the urge to procrastinate, I can't — I'd have to text a friend, explain why, wait for them to decrypt it, and paste the result. By then the urge is gone. The Result I've gone from losing entire afternoons to YouTube to actually finishing things. The blocker has caught me mid-reflex to type youtube.com more times than I can count. It's free if anyone wants it: github.com/Dan-J-D/kblocker If you keep breaking your own focus tools, this is for you.

2026-06-14 原文 →
← 上一页 下一页 →