AI in week 32.
39 new tools and models from 8 analysed videos; 17 confirmed available in the EU.
Videos analysed
New 3D editors, open medical AI, AI symphony, Qwen 3.8, Wan Animate 2: AI NEWSAIAI Search· 9 Aug 2026 · 17 tools
Nader Dabit: Devin, Open-source, Harnesses, Software Factories, and more!MAMatthew Berman· 7 Aug 2026 · 1 tools
AI News: The AI Stories Everyone's Freaking Out AboutMAMatt Wolfe· 7 Aug 2026 · 12 tools
What is Google even doing?MAMatthew Berman· 7 Aug 2026 · 4 tools
Master Codex with these 15 TipsMAMatthew Berman· 5 Aug 2026 · 11 tools
The BEST local AI video generator is here! Minimax H3 tutorialAIAI Search· 5 Aug 2026 · 7 tools
Open-source is WINNINGMAMatthew Berman· 4 Aug 2026 · 5 tools
New BEST AI video generator is here! Seedance 2.5 vs Minimax H3AIAI Search· 3 Aug 2026 · 7 tools
Language Models
All Language Models →GPT-6 Astra
Model · OpenAI
OpenAI's GPT-6 Astra, a frontier language model used for reasoning, agentic work and weight interpretation research.
Muse Spark 1.2
Model · Meta
Closed-source Meta model for coding and agentic workflows with 1M-token context and multimodal input, offered via API.
ClinFusion
Model · Alibaba DAMO Academy
Open-weights vision-centric multimodal LLM for 2D and 3D medical image understanding, in 8B and 32B sizes.
Big Bang v1
Model
Open-weights 35B-A3B agentic model post-trained from Qwen3.6-35B-A3B for search, coding and scientific research.
Nemotron
Model · Nvidia
NVIDIA's family of open models for agentic AI: reasoning, vision, speech, retrieval and safety models.
Kimi 3 Max
Model · Moonshot AI
Moonshot AI's large language model, described as open-source with roughly 2T+ parameters, benchmarked by Artificial Analysis.
Video Generation
All Video Generation →LeapTalk
Model
Research model for real-time, audio-driven talking-head video generation from a reference image, using one-step inference.
Wan Animate 2
Model · Alibaba
Alibaba's diffusion-transformer model that animates a character image from a driving video, with a real-time Lite variant.
FLUX 3 Video
Model · Black Forest Labs
Black Forest Labs' video generation model in the FLUX 3 family, with native audio and clips up to 20 seconds.
Lumina by BytePlus
App · BytePlus
BytePlus creative platform for generating video and other media with ByteDance and partner generative models.
Voice & Audio
All Voice & Audio →VocalRender
Model
Score-native singing voice synthesis system that turns lyrics and a MIDI melody into expressive vocals.
Codex Voice Mode
App · OpenAI
Native voice mode in OpenAI's Codex that lets you speak to start and control coding agent threads.
Music
All Music →SymphonyGen
Model
Research model that generates orchestral music by first composing a harmony skeleton, then expanding it into a full score.
ACE-Step
Model
Open-source diffusion-based music generation foundation model with lyric-aligned vocals and editing controls.
3D & Worlds
All 3D & Worlds →Hunyuan3D Buffalo 1.0
Model · Tencent
Tencent's unified multimodal model for 3D understanding, text-to-3D generation, instruction-based editing and part generation.
MAC (Multi-Agent CAD)
Agent
Open-source multi-agent framework that generates editable CAD parts and assemblies from text prompts using build123d.
Arty
Model · Nvidia
Nvidia platform for generating character animations from text prompts, used as a source of reference motion.
Mixamo
App · Adobe
Adobe's web service for auto-rigging 3D characters and applying a free library of motion-capture animations.
Agents & Automation
All Agents & Automation →Long Horizon Harness
Library
Backend-independent runtime that splits long-horizon agent tasks into manager, executor and auditor roles.
Ask Maps
App · Google
Conversational AI assistant inside Google Maps for place search, trip planning and agentic tasks like food pickup orders.
Codex Thread Delegation and Search
App · OpenAI
Codex feature letting one thread search, summarize and link other threads and delegate work to them.
Coding
All Coding →Muse Code
App · Meta
Meta's coding product tied to its Muse model family, reportedly offered through paid subscription tiers.
ChatGPT Sites
App · OpenAI
OpenAI feature that builds and hosts web apps with login and a database from within ChatGPT and Codex via @sites.
Codex Environments and Connections
App · OpenAI
OpenAI Codex features for running coding tasks in reusable cloud environments and reaching a desktop Codex remotely.
Codex Quota and Banked Resets
App · OpenAI
Usage quota system for OpenAI Codex, with weekly limits per ChatGPT plan and expiring banked resets that refill quota.
Productivity
All Productivity →ChatGPT Education Plugins
App · OpenAI
OpenAI education plugins for ChatGPT Work and Codex, for K–12 teachers, college educators and college students.
ChatLLM
App · Abacus AI
Subscription AI assistant from Abacus.AI that bundles 100+ LLMs, image/video generators and an agent for apps, slides and docs.
Robotics
All Robotics →UBTECH swarm intelligence (Cruiser Y1)
Model · UBTECH
UBTECH swarm intelligence system for coordinating fleets of wheeled industrial humanoid robots (Cruiser Y1) in warehouses.
Xiaomi Robotics 1
Model · Xiaomi
Open-weights Vision-Language-Action robot foundation model from Xiaomi, pre-trained on 100K+ hours of UMI trajectories.
Persona AI Gen 1 humanoid welding
Hardware · Persona AI
Humanoid robot from Persona AI, shown performing welding tasks via VR teleoperation for hazardous industrial work.
Hardware & Devices
All Hardware & Devices →Jony Ive OpenAI smart speaker
Hardware · OpenAI
Rumored battery-powered, doughnut-shaped AI smart speaker without a display, from OpenAI and Jony Ive.
Google TPU
Hardware · Google
Google's custom AI accelerators, offered through Google Cloud for training and inference of machine learning models.
Research
All Research →WeatherNext 2
Model · Google DeepMind
Google DeepMind weather forecasting model producing ensemble forecasts, including cyclone track and intensity up to 15 days ahead.
Attention Is All You Need
Research · Google
The 2017 Google paper that introduced the Transformer, a neural architecture based solely on attention mechanisms.
Platforms & Infra
All Platforms & Infra →Sign in with ChatGPT
API · OpenAI
OpenAI feature letting third-party apps sign users in with ChatGPT accounts, with AI usage billed to the user.
ComfyUI-Spectrum-MiniMax-H3
Library
ComfyUI custom node that speeds up MiniMax H3 audio-video sampling by forecasting skipped transformer steps.
AI Toolkit
Library · Ostris
Open-source GUI and CLI training suite for fine-tuning image and video diffusion models, including LoRAs, on consumer GPUs.
SageAttention
Library
Windows-wheel fork of SageAttention, a quantized attention kernel library that speeds up diffusion and video model inference.
Wan2GP
App
Open-source local web app for running video, image, audio, music and 3D generative models on low-VRAM GPUs.