Datawhale Rebuilds Stanford's CS336 Into a Chinese Open-Source LLM Course
A Chinese-language adaptation of Stanford's CS336 bundles 16 chapters and 6 hands-on assignments covering the full LLM stack, from BPE to GRPO.
// Intelligence
Today in AI
A focused feed of models, agents, research and open-source releases for people building with AI.
Updated
Community-Built PRISM Merges MiniMax-H3 Quality With Eight-Step Speed
A community-published merge for MiniMax-H3's Ref2VA video model uses SVD denoising and geodesic interpolation to fuse a quality finetune with an 8-step turbo variant.
Comfy-Org Rescues Stable Diffusion 1.5 With 1.5M-Download Mirror
Comfy-Org has published a repackaged archive of Stable Diffusion v1.5 on Hugging Face, preserving the original weights after RunwayML pulled its page.
OrcaRouter's OrcaCyber Nano 1.0 Lets Security Teams Analyze Malware Locally
OrcaRouter dropped a 27B vision-language model tuned for reverse engineering, vuln analysis, and secure code review, packaged as GGUF to run fully offline.
gpu-unlock Gives Nvidia RTX 3090 Ti Nearly 80% More AI Throughput on Linux
An open source Linux kernel module removes the FP16 and BF16 throttle on Ampere GeForce cards, roughly doubling tensor throughput on an RTX 3090.
Teaching Small AI Models to Learn From Their Own Mistakes
A new literature review unifies the fragmented field of on-policy distillation, giving the student-rollout training loop a shared f-divergence framework and three-axis taxonomy.
Harvey Turns Claude Subscriptions Into a Full Open-Source Sales Agent
Harvey is an open-source autonomous sales agent that runs cold email campaigns on your Claude or ChatGPT subscription with no per-token API bill.
Aporia Cuts AI Coding Mismatches by 79% by Making Hidden Decisions Visible
A new VS Code extension forces coding agents to ask you design questions first, cutting mental-model mismatches by 79 percent in a user study.
Nanjing University's AutoTavern Beats Top Hearthstone Players With 91% Decision Accuracy
An LLM research team built a Hearthstone Battlegrounds simulator and trained a reinforcement learning agent that beats the top 10 ranked humans on the live ladder.
Michigan Researchers' MCLR Bakes Guidance Into Training, Cutting Diffusion Inference Costs
A new training objective called MCLR reproduces classifier-free guidance effects at training time, revealing CFG as implicit contrastive alignment.
Moonshot's Kimi K3.1 Pro Tops Every Public API Model in Security Benchmarks
Moonshot's new reasoning model takes the top open spot on Artificial Analysis Cyber Index at 64%, trailing only a trusted-access OpenAI model.
Comfy-Org Makes Qwen-Image Work in ComfyUI Without Complex Setup
Comfy-Org maintains drop-in single-file safetensors for Alibaba's Qwen-Image family, bundling the base model, the 2512 refresh, and a distilled variant for ComfyUI.
Vals AI Finds the Same Open-Weight Model Costs 2.3x More Depending on Provider
Vals AI's new Inference Provider Index shows the same open-weight model can cost 2.3x more and run 5x slower depending on where you host it.
Cohere Locks Enterprise AI Prompts Behind Hardware-Encrypted Model Vault
Cohere's Model Vault now runs on hardware-isolated NVIDIA H100 enclaves on Azure, letting enterprises verify their data stays private even from Cohere.
OpenAI Cuts Live Internet Access From Frontier Model Training After Security Breach
OpenAI pulled frontier models offline, built a cached web for them to train in, then turned its own cyber agents loose to break the walls.
Epoch Finds Half of 21 Major AI Benchmarks Are Flawed
Epoch AI expanded its benchmark audit program, flagging popular evals like SWE-bench Verified and Humanity's Last Exam as Flawed under a public rubric.
Anthropic Lets Claude Code Run Hours-Long Jobs Without Dropping Them
Claude Code CLI now supports MCP Tasks for durable long-running jobs and Elicitations for mid-task user prompts, with other Claude products coming next.
NVIDIA's Cosmos 3 Cuts Robot Test Drives by 90% With Synthetic Data
NVIDIA expands Cosmos with data factory blueprints, training recipes and faster inference, with Porsche, Mercedes-Benz, Siemens and others building on top.
Black Forest Labs' FLUX 3 Klein Runs Image Generation on 8GB GPUs
Black Forest Labs drops a 5.6B open-weights image model that runs on 8GB VRAM, tops open benchmarks, and generates 1MP images in 4 seconds.
humans& Beats FlashInfer's Best MoE Kernel by 25% on Blackwell GPUs
Humans& rebuilt the NVFP4 decode kernel on Blackwell by keeping weights at 4 bits while promoting activations to BF16, gaining both speed and accuracy.
DeepSeek's V4.1 Flash 1017 Tops Public Cybersecurity Rankings at $0.42 per Task
DeepSeek V4.1 Flash 1017 tops the Artificial Analysis Cyber Index among open models, with zero safety refusals and a 53 point jump on CyberGym-E2E-AA.
Moonshot's Kimi K3.1 Pro Challenges Claude With 2.8 Trillion Parameters
Moonshot's 2.8T parameter MoE now runs full Unreal projects, parametric CAD, and finance modeling with 65% fewer hallucinations.
LlamaIndex's OCRia-1 Parses PDFs 75x Faster Than Gemini at a Third the Cost
LlamaIndex's new 4B vision-language model tops ParseBench by 12 points and parses a typical page in 0.2 seconds, available now on OpenDocRouter.
Liquid AI's d1.1 Cuts Decision Costs to $0.02 per Million Tokens
Liquid AI's d1.1 narrows the gap to d1-ultra while cutting cost by more than half, keeping input-only billing at $0.02 per million tokens.
No stories match these filters.