Skip to content
Sign inLaunch

Week 39, 2026

Discover the AI products launched Sep 21 – Sep 27 UTC. Final results.

Ranked by votes, comments, and recency24 launches

  1. 1
    Mini-AGI

    Dynamic continual learning model trained on 8GB VRAM

    mini-AGI is a byte-level language model that trains from scratch on a single 8 GB VRAM GPU and keeps learning from a continuous stream of text. Weights live on disk and are paged onto the GPU as needed, so model size is bounded by disk spac

    0 commentsY▲ 277 on HNLLMs & InfrastructureResearch & DataOpen Sourceby AI Launch Team

  2. 2
    Cactus Needle 3

    8-29MB automation models can match DeepSeek V4 Flash

    Needle 3 is an 8-29 MB foundation model from Cactus for phones, wearables, robots, smart home devices, cars and microcontrollers. It trades general chat ability for strong tool calling, structured extraction and text embeddings, running ful

    0 commentsY▲ 236 on HNLLMs & InfrastructureFreemiumby AI Launch Team

  3. 3
    LLM Attention Visualization

    A visualization of the attention mechanism in LLMs.

    An interactive visualization of the attention mechanism in a small large language model. Hover or tap any generated token to see which earlier tokens influenced it, with opacity scaled by attention weight across all heads and layers. Exampl

    0 commentsY▲ 177 on HNResearch & DataLLMs & InfrastructureFreeby AI Launch Team

  4. 4
    JevBench

    A reproducible benchmark for typed decision models

    JevBench, from Benchmark Heaven, compares Jev-class decision models across intelligence, calibration, speed and cost. It sits alongside Benchmark Heaven's broader model rankings, which can be filtered by provider, lab, hosting region and da

    0 commentsY▲ 153 on HNResearch & DataLLMs & InfrastructureFreeby AI Launch Team

  5. 5
    Kinesis

    Control your Mac with the Meta Neural Band

    Kinesis is a highly experimental macOS app that lets you control your Mac with a Meta Neural Band. Swipe between desktops, open Mission Control, control music, and pinch and turn for volume or brightness. You choose the gesture mappings, pr

    0 commentsY▲ 122 on HNProductivityOpen Sourceby AI Launch Team

  6. 6
    PacBench

    How well can models one-shot a Pac-Man game?

    PacBench tests how well a model and harness can recreate Pac-Man from one prompt: "Create a Pac-Man game in a single html page." Entries from different models and harnesses can be sorted by score, cost, time and tokens, with run data for ea

    0 commentsY▲ 78 on HNCoding & Dev ToolsResearch & DataFreeby AI Launch Team

  7. 7
    Chess Post-mortem Skills

    Claude Code skills that turn your chess games into readable post-mortems

    Claude Code skills that turn one of your chess games into a post-mortem you can read: plain-language explanations of your mistakes, checked against Stockfish, an annotated PGN, an HTML viewer and a narrated video. You can also record yourse

    0 commentsY▲ 75 on HNCoding & Dev ToolsProductivityOpen Sourceby AI Launch Team

  8. 8
    Lossless-memory

    A personal AI memory that never summarizes

    lossless-memory is a long-term memory system for a personal AI assistant that never summarizes. It keeps every raw line, timestamps everything, and searches by time first and words second, returning results in chronological order. A small i

    0 commentsY▲ 67 on HNLLMs & InfrastructureProductivityOpen Sourceby AI Launch Team

  9. 9
    Jevstiller

    Distill Jev into a local model – with a disagreement bound

    Jevstiller distills a hosted Jev classifier into a local model, with a measured bound on how often the local model disagrees with Jev. It runs as a drop-in proxy, so batch jobs avoid a network call per answer while keeping answers close to

    0 commentsY▲ 63 on HNLLMs & InfrastructureOpen Sourceby AI Launch Team

  10. 10
    Pizza Bot

    An inbox for AI agents that work in the background

    Pizza Bot is an inbox for long-running AI work. Start or schedule a task, walk away, and come back to finished work in Unread and decisions waiting in Action. Agents keep running when you disconnect, and the same app works on desktop, web a

    0 commentsY▲ 61 on HNAI AgentsProductivityOpen Sourceby AI Launch Team

  11. 11
    OtoDock

    Self-hosted company OS: Claude Code and Codex agents working in departments

    OtoDock is a self-hosted, multi-tenant platform for running collaborative AI agents across a company. Agents connect to your tools, work in departments, delegate to each other and keep working on their own. They run on your Anthropic and Op

    0 commentsY▲ 50 on HNAI AgentsProductivityFreeby AI Launch Team

  12. 12
    Geiger

    See every AI agent on your machine and what it can touch

    Geiger is a read-only command that inventories every AI agent, harness, MCP server, plugin and AI extension on a machine, and explains in plain language what each one can touch. Run npx geiger-scan with no install, account or telemetry. Fi

    0 commentsY▲ 49 on HNAI AgentsCoding & Dev ToolsOpen Sourceby AI Launch Team

  13. 13
    jevals

    Replacing LLM judges with typed Jev decisions

    jevals runs evals and guardrails for AI agents using Jev-style decision models instead of an LLM judge. All the checks for a trace, such as tool choice, groundedness, answer relevancy, prompt injection and PHI, go out as one request that co

    0 commentsY▲ 47 on HNLLMs & InfrastructureAI AgentsOpen Sourceby AI Launch Team

  14. 14
    Sunk Cost

    How long until a local LLM rig pays for itself?

    Sunk Cost estimates how long a machine for running local AI models takes to pay for itself compared with paying for an API. Pick a Mac mini, Mac Studio, DGX Spark or Strix Halo box, see which open models fit and how fast they run, and adjus

    0 commentsY▲ 47 on HNLLMs & InfrastructureResearch & DataFreeby AI Launch Team

  15. 15
    Graphify C#

    Compiler-accurate Find Usages for coding agents

    graphify-csharp is a free, headless Roslyn/MSBuild indexer that gives coding agents compiler-accurate Find Usages for C#. It turns a solution into deterministic semantic evidence: exact callers, references, implementations, inheritance, ove

    0 commentsY▲ 46 on HNCoding & Dev ToolsOpen Sourceby AI Launch Team

  16. 16
    Moadim

    A scheduler for agents

    Moadim is an open-source loop engine that runs AI agents on a schedule. Define a loop with a prompt, a schedule and an agent, and it runs Claude, Codex, Hermes or Pi against your repo on every tick, each run in an isolated workbench with a

    0 commentsY▲ 37 on HNAI AgentsCoding & Dev ToolsOpen Sourceby AI Launch Team

  17. 17
    aclif

    Agent CLI framework – one grammar – canonical names across SaaS

    aclif is a framework for building command-line tools that AI agents use to reach SaaS platforms. An agent gets one tool with one grammar and canonical names that reach the same record the same way on any platform, plus self-describing comma

    0 commentsY▲ 35 on HNAI AgentsCoding & Dev ToolsOpen Sourceby AI Launch Team

  18. 18
    OpenLake

    Storage engine for KV-cache offload and LLM training

    OpenLake is a high-performance storage engine for LLM inference and GPU training. Its Infinity Core I/O Engine delivered 6.72 GiB/s writes and 11.55 GiB/s reads in the MLPerf Storage v3.0 Llama 3.1 8B checkpointing benchmark through its S3

    0 commentsY▲ 35 on HNLLMs & InfrastructureOpen Sourceby AI Launch Team

  19. 19
    Lanes Link

    A personal context MCP: one endpoint for every agent you use

    Lanes Link is a private, self-hostable MCP server that connects your accounts, memory, skills and secrets to every AI agent you use, behind Access Profiles you manage. Connect services like Gmail, Calendar, Drive, Notion, Linear, Slack and

    0 commentsY▲ 28 on HNLLMs & InfrastructureProductivityFreemiumby AI Launch Team

  20. 20
    skillbay

    A curated marketplace of AI agent skills, each reviewed by a person

    skillbay is a curated marketplace of AI agent skills: SKILL.md packages that teach coding agents how to do a job. Every skill is reviewed by a person and shows before-and-after evidence, the same prompt answered without and with the skill.

    0 commentsY▲ 28 on HNAI AgentsCoding & Dev ToolsFreemiumby AI Launch Team

  21. 21
    Otis

    A minimal AI agent that runs local models out of the box

    Otis is a minimal AI agent from Triangl Labs that runs local models out of the box. First shared by its maker on Show HN: https://news.ycombinator.com/item?id=49696084

    0 commentsY▲ 19 on HNAI AgentsLLMs & InfrastructureFreeby AI Launch Team

  22. 22
    RxFilm Studio

    Create and edit your product videos with an AI agent, in one native Mac app

    RxFilm Studio is a native macOS app for making films with an agent in the director's chair. Score with generated music cues, add multi-speaker narration, transcribe, translate and retime captions, create stills, and render on a real timelin

    0 commentsY▲ 19 on HNVideoAudio & VoiceFreemiumby AI Launch Team

  23. 23
    LLMobi

    Use AI on your Android phone with no internet

    LLMobi lets you download an AI model to your Android phone once and then use it with no internet connection, no account and no bill. First shared by its maker on Show HN: https://news.ycombinator.com/item?id=49526848

    0 commentsY▲ 18 on HNChatbots & AssistantsFreeby AI Launch Team

  24. 24
    49 Agents

    2D Canvas for Agents

    49 Agents is an open-source 2D agentic IDE: one zoomable canvas for all your agents, terminals, projects and machines. It shows Claude status on every pane, and adds a Monaco editor, git graph, issue tables, permission notifications and mar

    0 commentsY▲ 18 on HNCoding & Dev ToolsAI AgentsOpen Sourceby AI Launch Team