AI in week 26.

GS

GPT-5.6 Sol

Model · OpenAI

29

OpenAI's GPT-5.6 Sol frontier model, with an Ultrafast API tier on Cerebras reaching up to 750 output tokens per second.

Frontier LLM reportedly from Anthropic, priced at $10/M input and $50/M output tokens; details unverified.

DR

DeepSeek R1

Model · DeepSeek

2

DeepSeek's open-weights reasoning LLM (671B MoE, 37B active) trained with large-scale reinforcement learning.

SF

Sakana Fugu

Model · Sakana AI

2

Sakana AI's orchestration API that routes tasks across a pool of LLMs through one OpenAI-compatible endpoint.

GM

GPT-5.4 mini

Model · OpenAI

1

Smaller, lower-cost OpenAI language model in the GPT-5.4 family, offered through the OpenAI API and ChatGPT.

S2

Seed 2.1

Model · ByteDance

1

ByteDance Seed's closed multimodal model family (Pro and Turbo) aimed at agentic, real-world productivity tasks.

UO

Unlimited OCR

Model · Baidu

1

Baidu's open-weights vision-language OCR model for parsing very long, multi-page documents.

Proof-of-concept image generator from Unconventional AI that uses coupled oscillators instead of diffusion.

DA

DanceOPD

Research · ByteDance

1

ByteDance research framework that distills multiple specialist image models into one via on-policy velocity matching.

S2

Seedance 2.5

Model · ByteDance

10

ByteDance Seed's video generation model supporting multi-reference inputs and one-take creation.

S2

Seedance 2.0

Model · ByteDance

3

ByteDance Seed's multimodal video generation model, producing video from text and reference inputs.

MA

Manim

Library

2

Community-maintained Python library for programmatically creating mathematical and explanatory animations.

RE

Remotion

Library · Remotion

2

React framework for creating and rendering videos programmatically, with a Studio, Player and agent skills.

DS

Domain Shuttle

Model

1

Subject-driven text-to-video model that keeps multiple reference characters and objects consistent across art styles.

HH

Happy Horse 1.1

Model · Alibaba

1

Alibaba's AI video generation model with native audio and lip sync, text, image and reference-to-video up to 1080p.

WS

Wan Streamer

Model · Alibaba

1

Alibaba research model for real-time conversational avatars using one transformer over text, audio and video.

OP

OpenMontage

Agent

1

Open-source agentic video production system that turns an AI coding assistant into a video studio.

VO

Voicebox

App

1

Open-source, local-first voice studio for voice cloning, text-to-speech, dictation and agent voice output.

LI

Lift4D

Research

1

Research method that reconstructs complete 4D (3D plus time) scenes from a single monocular video using Gaussian Splatting.

CT

Claude Tag

Agent · Anthropic

2

Anthropic feature that lets Slack teams tag Claude as a teammate to take on tasks and work in the background.

Computer-use AI agent from Brett Adcock, founder of Figure, reported to be in private beta.

DE

DeerFlow

Agent · ByteDance

1

ByteDance's open-source super agent harness that runs sub-agents, memory and sandboxes for long-horizon tasks.

SK

SkillSpector

Library · NVIDIA

1

NVIDIA's open-source security scanner for AI agent skills, checking repos, URLs, zips and files for vulnerabilities.

AC

Anthropic Cybersecurity Skills

Library

1

Open-source library of 800+ cybersecurity skills for AI coding agents, mapped to MITRE ATT&CK, NIST CSF and other frameworks.

GS

GStack

Library · Garry Tan

2

Open-source set of Claude Code slash-command skills that act as a virtual engineering team, by Garry Tan.

O1

Ornith 1.0

Model

1

Open-weights family of agentic coding models (9B to 397B) that learns to generate its own task-specific scaffolds.

UA

Understand Anything

Library · Egonex AI

1

Claude Code plugin that builds an interactive knowledge graph of a codebase or wiki using a multi-agent pipeline.

GR

Graphify

Library · Graphify Labs

1

Open-source CLI and agent skill that turns code, docs, PDFs and media into a queryable local knowledge graph.

TS

Taste Skill

Library · Leonxlnx

1

Agent Skill that guides coding agents to build company websites with one design system, real photos and browser-checked output.

FD

Frontend Design skill

Library · Anthropic

1

Anthropic agent skill that steers Claude toward distinctive, intentional frontend design instead of generic AI-looking UI.

CM

Codebase Memory MCP

Library · DeusData

1

Local MCP server that indexes code, docs and infrastructure into a knowledge graph for AI coding agents.

Multi-model AI workspace giving access to ChatGPT, Claude, Gemini and other models, with a no-code agent builder.

SS

Stop Slop

Library

1

MIT-licensed skill file that instructs Claude or other LLMs to remove common AI writing patterns from prose.

H5

HIW 500

Research · BitRobot Foundation

1

Open dataset of 500+ hours of whole-body teleoperated Unitree G1 humanoid data recorded in real homes.

OpenAI's custom in-house AI accelerator chip, codenamed Jalapeño, discussed in industry analysis.

IS

IBM sub-1nm chip technology (nanostack)

Hardware · IBM

1

IBM Research's 0.7 nm (7 angstrom) chip technology using a 3D "nanostack" transistor architecture, announced June 2026.

MR

Meta Ray-Ban glasses new styles

Hardware · Meta

1

New Ray-Ban Meta AI glasses styles, including the Meta Adventurer and Meta Fury, with camera, audio and Meta AI.

OI

OpenAI inference chip

Hardware · OpenAI

1

Custom AI inference chip announced by OpenAI and Broadcom, built to serve ChatGPT-style workloads faster and cheaper.

AU

Autodata

Research · Meta

1

Meta research framework where an AI agent acts as a data scientist, iteratively building training and evaluation datasets.

Research ultrasound method using injected microbubbles to map blood flow in the living human brain in 3D.

PE

PerceptionDLM

Model · ByteDance

1

Diffusion-based vision-language model from PKU MSALab and ByteDance that captions multiple image regions in parallel.

TA

The Atlantic AI Watchdog music training database

App · The Atlantic

1

Searchable Atlantic database showing which music and videos were used to train AI models such as Suno and Udio.