Top stories in AI today: – Nous Research's AI crushes elite math exam
– Microsoft maps how people use Copilot
– Fix bugs and ship features from Slack
– AI ring gives ‘external memory’ – 4 new AI tools, community workflows, and more Read more: https://
therundown.ai/p/open-source-
ai-crushes-elite-math-exam
…
OPEN SOURCE
-
AI Research Advances, Copilot Updates, and Smart Ring Innovation
By
–
-
Heretic 1.1: Open-Source LLM Abliteration Library Evolution
By
–
Abliterate LLMs with Heretic 1.1
— Maxime Labonne @ ICLR (@maximelabonne) 11 décembre 2025
It's cool to see this project evolving into a solid open-source library
The new viz feature shows how the abliteration process gradually groups the residual vectors into two nice clusters 👀 pic.twitter.com/PLYmRPuYB7Abliterate LLMs with Heretic 1.1 It's cool to see this project evolving into a solid open-source library The new viz feature shows how the abliteration process gradually groups the residual vectors into two nice clusters
-
OpenAI hints at ‘garlic’ model ahead of GPT-5.2
By
–
BREAKING : OpenAI hints at “garlic”, a rumoured codename of their upcoming model. GPT-5.2 is expected tomorrow
-
AGENTS.md Moves to AAIF with New Logo Launch
By
–
http://
AGENTS.md started as a practical way to share instructions with coding agents like Codex, and it’s been amazing to see the community adopt it so quickly. Excited to see it live beyond OpenAI. To mark the move to the AAIF, we put together a simple logo. -
Structured Prompting Improves AI Outputs
By
–
Here's why these work: The pattern: these techniques all add structure AROUND the generation step. Bad prompting: "Do the thing" Engineer prompting: "Plan how you'll do the thing, do the thing, verify the thing" Models are prediction engines. Give them a better scaffold and
-
Technique 4: Differential Prompting
By
–
Technique 4: Differential Prompting
— God of Prompt (@godofprompt) 10 décembre 2025
Engineers don't ask for one output. They ask for two versions optimized for different criteria, then pick or merge. This exploits the model's ability to hold multiple solution strategies.
Template:
Generate two versions of [output]:
Version… pic.twitter.com/yHkRJkECqf— Technique 4: Differential Prompting Engineers don’t ask for a single output. Instead, they request two versions optimized for different criteria, then select or merge them. This approach leverages the model’s ability to explore multiple solution strategies. Template: Generate two versions of [output]: Version 1:
—
[Content optimized for criterion A] Version 2:
—
[Content optimized for criterion B] -
Metacognitive Scaffolding for Error Prevention
By
–
Technique 3: Metacognitive Scaffolding Instead of asking for an answer, engineers ask the model to explain its reasoning process BEFORE generating. This catches logical errors at the planning stage. Template: Before you [generate output], first:
1. List 3 assumptions you're -
Multi-Shot Technique with Failure Cases
By
–
Technique 2: Multi-Shot with Failure Cases Everyone uses examples. Engineers show the model what NOT to do. This creates boundaries that few-shot alone can't establish. Template: Task: [what you want] Good example:
[correct output] Bad example:
[incorrect output]
Reason it -
Constraint Techniques for Prompts
By
–
Technique 1: Constraint-Based Prompting Most prompts are too open-ended. Engineers add strict constraints that force the model into a narrower solution space, eliminating 80% of poor outputs before they occur. Template: Generate [output] with these non-negotiable constraints:
– -
Mistral AI Launches Devstral 2 Coding Models and Vibe CLI
By
–
Congrats to @MistralAI on the big launch 🚀 You can check out Mistral Vibe in Zed today – just download it and add your API key! Mistral AI (@MistralAI) Introducing the Devstral 2 coding model family. Two sizes, both open source. Also, meet Mistral Vibe, a native CLI, enabling end-to-end automation. 🧵 — https://nitter.net/MistralAI/status/1998407332502405347#m
→ View original post on X — @arthurmensch, 2025-12-09 18:14 UTC