Ajah
Self-hostable LLM safety layer with RAG verification
What it does
Self-hostable gateway between your app and any LLM provider. Built after Helicone went into maintenance mode. What makes it different: - Verifies RAG responses against source documents - Flags hallucinations with zero latency added - Traces multi-agent sessions visually - Masks PII before storage - Costs attributed by user, feature, agent step 9 providers. docker-compose up. MIT license. No data leaves your server.
Does the same job
all alternatives →- ALAny-LLM – Lightweight router to access any LLM Provider2025 · github.com · ▲125
We built any-llm because we needed a lightweight router for LLM providers with minimal overhead. Switching between models is just a string change : update "openai/gpt-4" to "anthropic/claude-3" and you're done. It uses official provider SDKs when available, which helps since providers handle their own compatibility updates. No proxy or gateway service needed either, so getting started is pretty straightforward - just pip install and import. Currently supports 20+ providers including OpenAI, Anthropic, Google, Mistral, and AWS Bedrock. Would love to hear what you think!
- ASAmla Sandbox – WASM bash shell sandbox for AI agentsJan 2026 · github.com · ▲146
WASM sandbox for running LLM-generated code safely. Agents get a bash-like shell and can only call tools you provide, with constraints you define. No Docker, no subprocess, no SaaS — just pip install amla-sandbox
- AOAirgapped Offline RAG – Run LLMs Locally with Llama, Mistral, & Gemini2024 · github.com · ▲9
I've built an airgapped Retrieval-Augmented Generation (RAG) system for question-answering on documents, running entirely offline with local inference. Using Llama 3, Mistral, and Gemini, this setup allows secure, private NLP on your own machine. Perfect for researchers, data scientists, and developers who need to process sensitive data without cloud dependencies. Built with Llama C++, LangChain, and Streamlit, it supports quantized models and provides a sleek UI for document processing. Check it out, contribute, or suggest new features!
- HYHelicone (YC W23) – OSS LLM Observability and Development Platform2025 · github.com · ▲29
Hey HN, we're Justin and Cole, the founders of Helicone (https://helicone.ai). Helicone is an open-source platform that helps teams build better LLM applications through a complete development lifecycle of logging, evaluation, experimentation, and release. You can try our free demo by signing up (https://helicone.ai/signup) or self-deploy with our new fully open-source helm chart (https://helicone.ai/selfhost). When we first launched 22 months ago, we focused on providing visibility into LLM applications. With just a single line of code, teams could…
- OSOpen-source project that use LLM as deception system2025 · ▲10
Hello everyone I wanted to share a project I've been working on that I think you'll find really interesting. It's called Beelzebub, an open-source honeypot framework that uses LLMs to create incredibly realistic and dynamic deception environments. By integrating LLMs, it can mimic entire operating systems and interact with attackers in a super convincing way. Imagine an SSH honeypot where the LLM provides plausible responses to commands, even though nothing is actually executed on a real system. The goal is to keep attackers engaged for as long as possible, diverting them from your real…
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, May 2026
the whole month →
Parallel agents, diff reviewer, and multi-model comparisons
Dev tools · May 2026 · kilo.ai
- NW
Hey HN, Henry here from Cactus. We open-sourced Needle, a 26M parameter function-calling (tool use) model. It runs at 6000 tok/s prefill and 1200 tok/s decode on consumer devices. We were always frustrated by the little effort made towards building agentic models that run on budget phones, so we conducted investigations that led to an observation: agentic experiences are built upon tool calling, and massive models are overkill for it. Tool calling is fundamentally retrieval-and-assembly (match query to tool name, extract argument values, emit JSON), not reasoning. Cross-attention…
Life & fun · May 2026 · github.com
- FM
Dev tools · May 2026 · github.com