frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: AI Council – multi-model deliberation that runs in the browser

https://github.com/prijak/Ai-council
5•prijak•1h ago
There's LLM Council and similar tools, but they use predefined model lineups. This one is different in a few ways that mattered to me:

*Bring your own models.* Mix Ollama (local), OpenAI, Anthropic, Groq, Google — or any OpenAI-compatible endpoint — in whatever combination you want. A council of DeepSeek-R1 + llama2-uncensored + mistral-nemo is a very different deliberation than GPT-4o + Claude + Gemini.

*Zero server, zero account, zero storage.* The app is purely static. API calls go directly from your browser to providers. Nothing touches a backend. No tokens, no sessions, no analytics. Your API keys never leave your machine.

*Runs on your own hardware.* If you have Ollama, you can run an entire council locally for free. I use a 5-member all-Ollama setup on an RTX 2070 (8GB VRAM) — sequential requests, slow, but completely private.

The deliberation process is 3 stages: 1. All members answer independently 2. Each member critiques anonymized responses from the others 3. A designated Chairman synthesizes a final verdict

A few things I found genuinely interesting: - Reasoning models (DeepSeek-R1, QwQ) emit <think> blocks mid-stream. Stripping these while showing a " Thinking…" indicator keeps the UX clean without losing answer quality. - The Contrarian persona on an uncensored model produces meaningfully different critiques than a safety-tuned model playing the same role. - Peer review across models catches blind spots that a single model arguing with itself won't surface.

GitHub: https://github.com/prijak/Ai-council.git

Comments

abdulhak•1h ago
it's a good project, there is a platform that uses the same technique that you used to gather info on certain topic from different llms and there is one orchestrator that is running them all, and compares the answers and give you a deep understanding and analyses of each model gave as response i don't remember the name of it but i think they launched recently. it's really good project keep up.

YAML-schema-router – content-based schema routing for YAML-language-server

https://github.com/traiproject/yaml-schema-router
1•drLuca•4m ago•1 comments

Show HN: TypstMe – Build ATS-proof resumes using Typst and WASM

1•yoangilliand•6m ago•0 comments

Ask HN: Can RAG be used for recommendation system?

1•pinner•10m ago•0 comments

Ask HN: What happens to all the hardware when data centers upgrade?

1•givemeethekeys•10m ago•0 comments

CSLib: A Focused Effort on Formalizing Computer Science in Lean

https://www.cslib.io/
1•md224•11m ago•0 comments

Steve Webber on Multics

https://www.youtube.com/watch?v=0gzjrF9r6QA
1•twoodfin•12m ago•0 comments

Unlocking full-body human motor learning at scale

https://amathislab.github.io/mm-blog/
1•moatmoat•12m ago•0 comments

A Bollywood Spy Thriller Crosses the India-Pakistan Divide

https://www.bloomberg.com/news/articles/2026-02-20/bollywood-epic-dhurandhar-breaks-records-in-in...
1•petethomas•13m ago•0 comments

Show HN: Velo – Open-source, keyboard-first email client in Tauri and Rust

2•Espired•21m ago•0 comments

If Chatbots Can Replace Writers, It's Because We Made Writing Replaceable

https://thewalrus.ca/if-chatbots-can-replace-writers-its-because-we-made-writing-replaceable/
2•eon01•21m ago•0 comments

What Is OAuth?

https://leaflet.pub/p/did:plc:3vdrgzr2zybocs45yfhcr6ur/3mfd2oxx5v22b
4•cratermoon•21m ago•0 comments

Napkin Math

https://github.com/sirupsen/napkin-math
2•fagnerbrack•22m ago•0 comments

The Rise and Fall of Scala: A Love Letter to the Language That Broke My Heart

https://medium.com/@naveensky/the-language-that-made-me-feel-like-a-genius-until-it-didnt-my-brea...
1•b-man•25m ago•0 comments

Cord: Coordinating Trees of AI Agents

https://www.june.kim/cord
6•gfortaine•28m ago•0 comments

Heaven Has a Physical Location, a Physicist Claims

https://www.popularmechanics.com/science/a70423234/location-of-heaven/
2•nikolay•29m ago•3 comments

Facebook Is Doing A-Okay

3•givemeethekeys•30m ago•1 comments

Show HN: Share Your Ideas with the Community

https://novadev.vip/share/
1•SuperGamer474•31m ago•0 comments

Precise exceptions in relaxed architectures [video]

https://www.youtube.com/watch?v=JkNgZz4bu1E
1•matt_d•37m ago•1 comments

An Archaeological Dig Through the Extropian Archives

https://mtabarrok.com/extropia-archaeology
1•jdkee•37m ago•0 comments

India's Sarvam 105B LLM tops OCR benchmarks

https://timesofindia.indiatimes.com/technology/tech-news/explained-what-is-indias-sarvam-ai-model...
1•dyauspitr•38m ago•0 comments

Show HN: I created a beautiful number animation library for React Native

https://number-flow-react-native.awingender.com/
1•WingenderA•39m ago•0 comments

Correcting the Financial Times Report about AWS, Kiro, and AI

https://www.aboutamazon.com/news/aws/aws-service-outage-ai-bot-kiro
1•nonfamous•40m ago•0 comments

Show HN: I built a 55K-word email marketing knowledge base and Claude Code skill

https://www.emailmarketingskill.com/
2•cosmoblk•47m ago•0 comments

Little PAIPer – Learn computer science through Peter Norvig's teachings

https://lilpaiper.ai
1•vismit2000•47m ago•0 comments

Show HN: Universal ROS bridge for AI agents – control robots with LLMs

https://github.com/webthree549-bot/agent-ros-bridge
1•webthree549•51m ago•0 comments

The last-mile data problem is stalling enterprise agentic AI

https://theagenttimes.com/articles/the-last-mile-data-problem-is-stalling-enterprise-agentic-ai-g...
1•Ross00781•55m ago•0 comments

CFO says "Everyone uses Excel, definitely not Sheets. Get over it". True?

1•0xWTF•56m ago•4 comments

Stripe is holding $50K and refunding our customers

3•JohannesSchip•57m ago•0 comments

Claws are now a new layer on top of LLM agents

https://twitter.com/karpathy/status/2024987174077432126
4•Cyphase•59m ago•3 comments

SleepFM model predicts 130 conditions (C-Index >0.75) from one night of sleep

https://www.nature.com/articles/s41591-025-04133-4
2•marojejian•1h ago•1 comments