2026
10,781 articles · 136 days · 10 categories · avg 79 per day
SpecSlot Architecture Cuts FID by 2.84 Against JiT-CM and Saves 3.9% GFLOPs
SpecSlot gives tokens fixed frequency roles and sends only low-frequency ones into global attention. On ImageNet-512, FID is 2.84 lower than JiT-CM with 3.9% fewer…
A drone with the 1B OpenFlyV2 model flies to its target on its own in 48% of real flights, without training on them
Shanghai AI Lab's OpenFlyV2 packs scene understanding and quadcopter control into a single 1B-parameter on-board model. In real flights, it reaches the target 48%…
L3G-World Generates 3D Worlds in Real Time at 9.02 Frames/s on an H100
L3G-World stores a scene in compact memory built from 3D Gaussians with VAE latents. It generates a consistent, explorable world one frame at a time at 9.02…
AOS-Bench tests AI agents: quality drops 13-23 points under parallel workloads
CuraLoc Pinpoints Code Bugs 59.3% More Accurately Than Rivals While Using 10x Fewer Tokens
A 2015 CNN Beat GPT-6 Astra and Claude Fable 5.1 at Skin Cancer Diagnosis
METR Experiment Recalculated: On Complex Tasks and Reviews, Developers Using AI Lose 24% of Their Hours
Google Taught Language Models to "Sleep" and Move Knowledge from Memory into Weights
Google researchers have built a framework called SLEEP: a model first accumulates new information in short-term memory, then writes it into its weights during a…
Keystone removes the model name from code: open-source LLM gateway adds 0.64 ms
Keystone's open-source Go gateway routes requests by difficulty, not model name. Fallbacks, dollar limits and a 250 ms judge add just 0.64 ms of latency.
Anthropic files for IPO: $16.3 billion in revenue in six months, ticker ANT
YADRO bans its LLM from touching numbers and ships 10+ reports a day with no humans in the loop
Anthropic reveals how it builds Claude with Claude: the brain separate from the hands
AI Agents Translated a 15,000-Page Proof Into 43 Million Lines of Code in 24 Days — and Found a 40-Year-Old Error
Liquid AI releases LFM2.5-24B-A2B: an agent model with 2B active parameters
Liquid AI has open-sourced the weights of a 24B-parameter MoE model with only 2B active. It runs at 184 tok/s on an M5 Max laptop and at over 21K tok/s on a single…
PyTorch 2.15 speeds up MoE models on Mac by 5–6.5x and adds FP8
Hugging Face's PyTorch team rewrote critical MPS ops in Metal: Qwen3.6-35B-A3B on an M3 Ultra jumped from 9.4 to 61.3 tokens/s, and Wan2.2 peak memory fell from 270…
AgentOrch: Pre-Planned AI Agents Now Trail Step-by-Step Execution by Just 8.5 Points
Researchers compared two ways AI agents handle workflows: planning everything upfront versus checking results step by step. The step-by-step approach is still more…
Moonshot releases Kimi K3.1 Pro: 2.8 trillion parameters, weights due November 4
Switzerland's Apertus 1.5 Gains Vision, Hearing and a 262K-Token Context Window
NVIDIA Releases 120B Nemotron 3.5 Super VL and Bolt Models for Local Agents
Goodfire Researcher Claims Neural Networks Are Far Simpler Than Interpretability Experts Thought
aiedit hides individual lines of code from an AI agent: 5 levels and 857 tests
A Python beginner built the aiedit editor with Claude: the model sees the project but not the locked lines. Code, not a prompt, enforces it: 11K lines, 857 tests…
vps-forge: root-enabled agents on a VPS get automatic rollback, self-bans drop from 7 of 60 to 0
The open-source Agent Skill vps-forge arms a 5-minute dead man's switch before the agent touches anything. Across 120 Opus 5.5 runs, scores rose from 79–92 to…
rwkv-quant: keeping one matrix in bf16 removes 39% of RWKV-7 2.9B's quantization loss
OpenAI Releases GPT-6.1 Terra: Half the Price, Twice the Accuracy in Code
Qwen3-8B on the Intel Core Ultra 5 226V NPU: 10.7 tok/s at 15 W and 24 dB of Noise
AWS Launches Multi-Turn RL in SageMaker: gpt-oss-20b Climbs From 15% to 69% for $339
Codex on GPT-6.1 Sol trims its window to 828 400 but still accepted an 888-thousand-token input
The config says model_context_window = 1 000 000, yet Codex reports 828 400. A 890 547-token test shows it is a history threshold, not a limit on a single input.
12 Pentesting Tools vs. Juice Shop: The Best AI Agent Found 5 Holes Out of 116
The author ran 6 AI agents, 4 DAST scanners and a Claude + Burp MCP combo against OWASP Juice Shop's 116 vulnerabilities. The hybrid won with 9/10; headline…
Google Releases Gemini 4 Argon: $2 per Million Tokens and First Place on the Vals Index
Gemini 4 Argon is live in the API and Gemini app: $2/$10 per million tokens, 95% off cached input, #1 on the Vals Index. Chrome, Gemma and quantum kernels got…
Meta open-sources Muse Spark Open: an 816B-parameter MoE with a 1M-token context
Six AI Agents Forecast Earnings Per Share: Three Matched the Analysts
Modal Raises $750M at a $15.75B Valuation and Reaches $1B in ARR
