AI News

Fine-Tuning Tool-Calling LLMs: A Complete Guide Using XYZ-Aquila-SFT and Qwen3

Implement an end-to-end fine-tuning pipeline for tool-calling language models. This tutorial covers parsing trajectories, structured tool-call extraction, Qwen-compatible ChatML rendering, and efficie

Chert

<p> Vapi for FaceTime: AI video agents in a few lines </p> <p> <a href="https://www.producthunt.com/products/chert?utm_campaign=producthunt-atom-posts-feed&

How to Build a Simple AI Web Scraper with Python

Turn any webpage into a lightweight LLM-powered QA engine by cleaning HTML, converting content to Markdown, and returning focused answers while reducing token usage.

5 Fun Agentic AI Papers to Read

If you read only five papers on AI agents, make them these.

Z.ai Ships GLM-5.3 Without Retraining the Base Model: Better at Complex Coding and Long-Horizon Tasks

Z.ai released GLM-5.3 on August 14, 2026. The model reuses the 743B GLM-5.2 base unchanged. Every reported gain comes from scaled post-training: more long-horizon task environments, more environment t

HarnessRouter Community Edition

<p> Open-source unified interface for agent harnesses </p> <p> <a href="https://www.producthunt.com/products/epsilla?utm_campaign=producthunt-atom-posts-fee

Meet Needle 2: An Open 45M-Parameter Tool-Calling Model That Ships as a 14MB Binary and Runs a Full Session in 28MB of RAM

Cactus Compute released Needle 2, an open 45M-parameter model for tool calling, device use, and structured extraction. The full model is a single 14MB binary that runs a session in about 28MB of RAM.

Create a Reasoning-Focused LLM: A Practical Guide to Streaming, Curating, and Fine-Tuning the SupraLabs Reasoning Corpus

This tutorial provides a complete workflow for building a compact, reasoning-focused language model. By streaming the SupraLabs reasoning corpus from Hugging Face, we apply quality filters and curate

State of Open Models: Summer 2026 Observations

Chronock

<p> Scheduling and calendar sync, all in one </p> <p> <a href="https://www.producthunt.com/products/chronock?utm_campaign=producthunt-atom-posts-feed&ut

Google AI Just Released Gemini 3.7 Flash: A Coding and Agent Model at $0.75/1M Input Tokens

Google has released Gemini 3.7 Flash, a refinement of Gemini 3.6 Flash with algorithmic improvements to its reasoning core. It handles text, images, audio, and video across a 1M-token context window w

Record, train, and deploy from one place with Strands Agents, LeRobot, and Hugging Face Storage Buckets

Liquid AI Releases LFM2.5-VL-3B: A 3B Vision-Language Model That Reads Screens, Grounds Objects, and Calls Tools On-Device

Liquid AI released LFM2.5-VL-3B, a 3.1B-parameter vision-language model built for on-device deployment. It averages 80.7 on ScreenSpot-v2 and lifts RefCOCO grounding from 57.1 to 87.9. Function callin

Building a Streaming Local AI Agent

Streaming gets used in two different ways when people talk about AI agents. Straighten out your understanding here.

Constraining Output Space for SLM Narrow Automation Optimization

This article will kick off a series on narrow automation optimization for SLMs, and as the first entry will cover one of the more most useful techniques for doing so: constraining the output space ins

Dyna Robotics Introduces Dyna-2: A World-Action Model Pre-Trained on 1 Million Hours of Human Video

Dyna Robotics has released Dyna-2, a world-action model pre-trained on more than one million hours of egocentric human video. The technical report establishes three results: a scaling law on human dat

SpaceXAI Releases Grok 4.6: A 500K-Context Frontier Model Tuned for Long-Running Agents, Coding, and Knowledge Work

SpaceXAI released Grok 4.6 on August 12, 2026 — a post-training upgrade over Grok 4.5, not a larger base model. It ties GPT-5.6 Sol Max at 61 on the Artificial Analysis Intelligence Index, ships 500K

What We Learned by Reproducing 2,200 papers from ICML

AllenAI Open Instruct Tulu 3 Post-Training with SFT, DPO, RLVR, GRPO, and Verifier-Based Evaluation

Build a custom LLM post-training pipeline using AllenAI’s Open Instruct framework. This comprehensive guide walks through Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Reinfo

Introducing OlmoEarth embeddings: Custom embedding exports from OlmoEarth Studio for downstream analysis

nenspace

<p> the lo-fi of LLMs: your mind, made larger </p> <p> <a href="https://www.producthunt.com/products/nenspace?utm_campaign=producthunt-atom-posts-feed&u

LFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge

Building an End-to-End Data Science Portfolio Project

Most portfolios stop at a notebook. Take yours all the way.

5 Easy Ways to Install Python on Windows

Learn how to install Python on Windows using the Python Install Manager, WinGet, uv, Miniconda, or the official Python installer, and choose the best setup for beginners and Python development.

NVIDIA AI Releases Nemotron 3.5 Lightning: A 30B Open MoE with 3B Active Parameters, and NeMo Switchyard Model Router

NVIDIA's open 30B MoE targets the agent execution layer, with Switchyard routing each step to the cheapest capable model. The post NVIDIA AI Releases Nemotron 3.5 Lightning: A 30B Open MoE with 3B Act

Expeditione

<p> The Interactive 3D Encyclopedia </p> <p> <a href="https://www.producthunt.com/products/expeditione?utm_campaign=producthunt-atom-posts-feed&utm_medi

The Ultimate Guide to Contributing to Open Source Projects

This guide walks through what contributing to open source projects actually covers, how to pick a project that will actually respond to you, the exact git mechanics, and more.

Thinking of ACE? We Can Do It with Fewer Tokens

3 Visual Proofs of the Central Limit Theorem to Build Your Intuition

To build your intuition, this article shows three visual proofs that the classic bell curve appears in myriad situations.

Build Low-Latency Multilingual Voice Agents: Open Weights & Full Deployment Control with NVIDIA Magpie TTS

New Free eBook: Understanding Agentic AI, an Executive Briefing

Written for CEOs, CTOs, CIOs, and technology executives, our new free ebook "Understanding Agentic AI: An Executive Briefing" walks through the components every real agentic system is built from.

Specification Engineering: The New Skill After Prompt Engineering

Prompt engineering taught us how to ask better questions. Specification engineering teaches us how to define better work.

AirAlarm

<p> Wake at the end of your sleep cycle </p> <p> <a href="https://www.producthunt.com/products/airalarm?utm_campaign=producthunt-atom-posts-feed&utm_med

Making Knowledge Distillation Cheap Enough to Run at Scale

Meta is back with Muse Glimmer: local, agentic, multimodal, and open source

Attyn

<p> Bringing intelligence to your cursor </p> <p> <a href="https://www.producthunt.com/products/attyn?utm_campaign=producthunt-atom-posts-feed&utm_mediu

Clamshell

<p> Close your MacBook. Keep the work running. </p> <p> <a href="https://www.producthunt.com/products/clamshell?utm_campaign=producthunt-atom-posts-feed&amp

Zetik

<p> A chief of staff in your pocket </p> <p> <a href="https://www.producthunt.com/products/zetik?utm_campaign=producthunt-atom-posts-feed&utm_medium=rss

FileRouter

<p> Take control of files and editors </p> <p> <a href="https://www.producthunt.com/products/filerouter?utm_campaign=producthunt-atom-posts-feed&utm_med

Vidaya

<p> Healthspan score from your wearables, labs, and DNA. </p> <p> <a href="https://www.producthunt.com/products/vidaya?utm_campaign=producthunt-atom-posts-

Baseten on Hugging Face Inference Providers 🔥

Joy

<p> A Mac menu bar app that throws confetti for your wins </p> <p> <a href="https://www.producthunt.com/products/joy-celebrate-your-wins?utm_campaign=produc

CostLogic

<p> AI-powered construction takeoffs, estimates, and invoices </p> <p> <a href="https://www.producthunt.com/products/costlogic?utm_campaign=producthunt-atom

GPU Management: Why Idle GPUs Are the New Grounded Aircraft

NVIDIA Cosmos-H-Dreams: Bringing Real-Time Generative Simulation to Surgical Robotics

Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incident

Bringing Nunchaku 4-bit Diffusion Inference to Diffusers

Grabette: an open system to record robot-manipulation data

Blume

<p> AI-ready, Markdown-first documentation framework </p> <p> <a href="https://www.producthunt.com/products/blume-3?utm_campaign=producthunt-atom-posts-feed

Big Mike

<p> Your favorite uncle who knows sports & bets, on iMessage </p> <p> <a href="https://www.producthunt.com/products/big-mike?utm_campaign=producthunt-atom-p