A diagram of the BRAID framework showing a Markov decision process with text and image generation loops connected by…

BRAID Fuses Text-Image Reasoning Into One RL Objective

BRAID unifies multi-turn text-image reasoning as a Markov decision process, enabling joint RL optimization of both modalities with a single objective.

x.com/Jul 11, 2026/3 min read

generative-aireinforcement-learningmultimodal

A robotic arm in a modern kitchen reaches toward a bunch of bananas on a countertop, surrounded by other fruit and…

AI ResearchBreakthrough

100

BAAI Orca World Model Matches π0.5 With No Action Labels

BAAI's Orca world model matches specialized π0.5 on five robotics tasks, trained on 125,000 hours of video without action labels, predicting abstract world states.

the-decoder.com/Jul 11, 2026/3 min read/Widely Reported

roboticsworld modelsai

NVIDIA researcher at computer screen showing NVFP4 quantized Kimi-K2.7-Code model on Hugging Face, with Blackwell…

AI Research

90

NVIDIA Releases FP4 Quantized Kimi-K2.7-Code with 1T Parameters

NVIDIA released FP4 quantized Kimi-K2.7-Code on Hugging Face, a 1T-parameter model for Blackwell GPUs with claimed accuracy retention.

x.com/Jul 10, 2026/3 min read

hardwarequantizationai

Two business executives shake hands in a modern office, with Tencent and Manus logos visible on a digital screen…

Policy & EthicsBreakthrough

100

Tencent in Talks for Majority Stake in Manus at $2B After Meta Deal Blocked

Tencent is buying a majority stake in AI agent startup Manus at $2B valuation after Beijing blocked Meta's acquisition. Manus reported ~$500M annual revenue.

the-decoder.com/Jul 10, 2026/3 min read/Widely Reported

chinametaai agents

Products & Launches

95

OpenAI GPT-5.6 Sol matches Fable 5 at 1/3 cost, adds multi-agent API

OpenAI's GPT-5.6 Sol nearly matches Claude Fable 5 on aggregate benchmarks at one-third the cost, with new multi-agent and tool-calling APIs.

scmp.com/Jul 10, 2026/3 min read/Widely Reported

anthropicagentic aiai models

A group of professionals in business attire shaking hands at a Hong Kong financial district event, with a digital…

Funding & BusinessBreakthrough

80

Chinese AI Firms Raise $20B in Hong Kong Amid US Chip Curbs

Chinese tech firms raised $20B in Hong Kong for AI and semiconductor expansion, driven by US chip export controls.

news.google.com/Jul 10, 2026/3 min read

us-china tradesemiconductorsai funding

A line chart showing Anthropic's annualized revenue rate climbing from $400M to $550M daily growth, with a $69B peak…

Funding & Business

100

Anthropic ARR Hits $69B, Daily Revenue Growth Accelerates to $550M

Anthropic ARR hits $69B per Yipit, daily revenue growth accelerates to $550M. Secondary market valuation reaches $1.2T.

nextbigfuture.com/Jul 10, 2026/3 min read/Widely Reported

anthropicfundingbusiness

Mark Zuckerberg at a Meta event, smiling while presenting Muse Spark 1.1 on a large screen showing code, with a…

AI Research

100

Meta Muse Spark 1.1 Debuts in AI Coding Battle; Zuck Post Hits 12M Views

Meta released Muse Spark 1.1 for agentic coding tasks. Zuckerberg's post got 12M views in 12 hours; no benchmarks disclosed.

x.com/Jul 10, 2026/3 min read/Widely Reported

metaai modelscoding assistants

AI Research

77

JPMorgan AI Agents Beat 60/40 Portfolio in Backtests

JPMorgan's AI agents outperformed the 60/40 portfolio in backtests, signaling a shift toward autonomous asset allocation by major financial institutions.

bloomberg.com/Jul 9, 2026/3 min read

financejpmorganai agents

A software engineer reviews code on a large monitor displaying benchmark tasks, with a broken task highlighted in…

AI Research

95

OpenAI Finds 30% of SWE-Bench Pro Tasks Are Broken, Pulls Endorsement

OpenAI finds ~30% of SWE-Bench Pro tasks broken, pulls endorsement. Human reviewers flagged 249 flawed tasks.

the-decoder.com/Jul 9, 2026/3 min read/Widely Reported

ai codingbenchmarksopenai

Side-by-side coding benchmark results comparing Databricks' GLM 5.2 and Anthropic Opus 4.8, with cost per task shown…

Open SourceBreakthrough

100

Databricks Defaults to Chinese Model GLM 5.2, Matches Opus at $1.28/Task

Databricks defaulted to GLM 5.2 after it matched Opus 4.8 at $1.28/task vs $1.94. The move signals enterprises building custom benchmarks and multi-vendor AI stacks.

the-decoder.com/Jul 9, 2026/3 min read/Widely Reported

open sourcecoding agentsmodel economics

A reflective orchestration agent interface showing DeepSeek V3.2 with a 67.25% pass@2 score on ARC-AGI-1, costing…

AI ResearchBreakthrough

86

DeepSeek V3.2 Agent Hits 67% on ARC-AGI-1 Without Fine-Tuning

Moghe & Chin achieve 67.25% pass@2 on ARC-AGI-1 using DeepSeek V3.2 in non-thinking mode at $0.62/task, with no fine-tuning. The work demonstrates agent architecture alone can lift a 15.50% baseline by ~52 points.

arxiv.org/Jul 9, 2026/3 min read

arc-agibenchmarksdeepseek

Nvidia, Hugging Face Open-Source Robot Models to Democratize Physical AI

Open SourceBreakthrough

98

Nvidia, Hugging Face Open-Source Robot Models to Democratize Physical AI

Nvidia and Hugging Face open-sourced robot models to democratize physical AI, providing pre-trained models and simulation tools on the Hugging Face hub.

aibusiness.com/Jul 8, 2026/3 min read/Widely Reported

hugging faceopen-sourcerobotics

Funding & Business

100

BofA Lends OpenAI $520M Ahead of Delayed IPO

BofA lent OpenAI $520M ahead of its IPO, now delayed to 2027. First bank loan for the $40B+ funded AI leader signals lender confidence.

news.google.com/Jul 8, 2026/3 min read/Multi-Source

financefundingai

A developer's laptop screen shows the Gemini API dashboard with a workflow diagram connecting an async agent node to…

Products & Launches

83

Google DeepMind adds async agents, MCP support to Gemini API

Google DeepMind added background execution and MCP support to Gemini API Managed Agents. Four new features target developers building long-running, stateful agent workflows.

the-decoder.com/Jul 8, 2026/3 min read/Multi-Source

geminimcpai agents

Four metagaming types need separate fixes or models learn…

AI ResearchBreakthrough

82

Four metagaming types need separate fixes or models learn to conceal it

A LessWrong taxonomy classifies AI metagaming into four types requiring separate fixes; blanket mitigation may teach concealment.

lesswrong.com/Jul 8, 2026/3 min read

taxonomyalignmentai safety

Mistral AI engineer tests Robostral Navigate on an autonomous warehouse robot navigating between shelves at a…

Products & Launches

100

Mistral AI Ships Robostral Navigate for Physical AI Push

Mistral AI released Robostral Navigate, a robotics navigation model for physical AI, targeting European industrial customers. The company is negotiating a €3B funding round at a €20B valuation.

bloomberg.com/Jul 8, 2026/3 min read/Widely Reported

roboticsintelmistral

Chinese AI engineers in a clean lab examine a silicon wafer, with chip design blueprints on a nearby monitor…

Big TechBreakthrough

100

DeepSeek, Zhipu AI Build Custom Inference Chips to Cut GPU Dependency

DeepSeek and Zhipu AI are developing custom inference chips to cut GPU costs. China's domestic chip budget share hit 46% in July 2026.

pandaily.com/Jul 8, 2026/3 min read/Widely Reported

custom siliconai hardwarechina ai

OpenAI GPT-5.6 launch event with a large digital display showing benchmark scores and a sleek AI interface…

Products & LaunchesBreakthrough

100

OpenAI GPT-5.6 Launches Thursday After US Gov't Lifts Ban

OpenAI's GPT-5.6 Sol launches Thursday after US gov't lifts ban. It beats Claude Mythos 5 on benchmarks at half the cost.

the-decoder.com/Jul 8, 2026/3 min read/Widely Reported

model pricinganthropicgpt-5.6

Steeve Morin, ZML founder, stands beside a server rack, gesturing toward hardware, with a data center background

Startups

82

ZML releases free LLM inference server supporting Nvidia

ZML released LLMD, a free inference server for LLMs supporting Nvidia, AMD, Google TPU, Apple Metal, and Intel Arc, aiming to reduce AI costs and break vendor lock-in.

techcrunch.com/Jul 8, 2026/3 min read

open sourceai infrastructurechips

A line graph titled 'LLM agent performance vs task length' shows a steep downward curve labeled 'nonlinear failure'…

AI Research

90

LLM agents fail nonlinearly as tasks lengthen, 27-paper synthesis finds

27-paper synthesis finds LLM agent failures compound nonlinearly with task length. Six failure clusters identified across 19 benchmarks.

arxiv.org/Jul 8, 2026/3 min read/Multi-Source

llm agentsbenchmarksai research

A bar chart showing CAD benchmark scores for 7 LLM models, with DeepSeek-V3.2 and Qwen3-235B-A22B at the top near…

AI ResearchBreakthrough

75

LLMForge: 7 Models Score 0.89 on CAD Benchmark; VLMs Fix Cylinders

LLMForge scores 0.89 on 97-design CAD benchmark. VLM critic achieves 100% watertight meshes but cylinders remain a failure mode.

arxiv.org/Jul 8, 2026/3 min read

benchmarkaicad

Products & Launches

100

Cohere Releases Arabic Speech Recognition Model Under Apache 2.0

Cohere released Arabic speech recognition model under Apache 2.0, claiming world's most accurate open-source version. No benchmark data or training details disclosed.

x.com/Jul 7, 2026/3 min read/Multi-Source

open-sourcearabiccohere

Lilian Weng, co-founder of Thinking Machines Lab, gestures during a presentation, with a slide showing a diagram of…

AI Research

94

Lilian Weng Argues Harness Design, Not Model Rewrites, Is Path to RSI

Lilian Weng argues RSI starts with harness design, not model rewrites, citing Sakana AI's The AI Scientist in Nature 2026 and two other projects.

x.com/Jul 7, 2026/3 min read

sakana airecursive self-improvementai research

A diagram illustrating GraphRAG memory design with MongoDB at center connecting to full-text, semantic, and graph…

AI Research

75

GraphRAG Memory Design: Retrieval Over Storage, MCP Integration

Agent memory design prioritizes retrieval over storage, using unified MongoDB and MCP server. GraphRAG enables multi-hop traversal via three strategies.

x.com/Jul 7, 2026/3 min read

knowledge graphsmcpai agents

Line chart comparing Ant's 1.1B LingBot-Vision model against Meta's 7B DINOv3 on 12 benchmarks, with Ant's model…

AI Research

100

Ant Group's 1.1B LingBot-Vision Beats Meta's 7B DINOv3 on 12 Benchmarks

Ant Group's 1.1B LingBot-Vision tops Meta's 7B DINOv3 on 12 spatial benchmarks, with 40% fewer FLOPs.

pandaily.com/Jul 7, 2026/3 min read/Widely Reported

computer visionai modelsfintech ai

Simplexity Robotics i7 Pro robots on a factory production line, with multiple units positioned near industrial…

Startups

88

Simplexity Ships 100 Robots 11 Months After Founding

Simplexity Robotics shipped 100 i7 Pro robots 11 months after founding. Claims all-scenario capability without reconfiguration, but lacks third-party validation.

pandaily.com/Jul 7, 2026/3 min read/Multi-Source

roboticsstartupsembodied ai