AI
Artificial intelligence: LLMs, agents, diffusion, vision, NLP and the latest from top labs.
209 published articles
Product launch
Claude Sonnet 5 is here, and Anthropic is betting on science and enterprise
Anthropic released Claude Sonnet 5, a faster and more capable model for coding and agents, along with Claude Science for researchers and Claude Tag for teams. The company also published a policy framework on AI exponential growth.
2026-06-30
Artificial Intelligence
DeepSeek's new model makes efficiency the AI arms race front line
DeepSeek has released a new LLM that sharpens its competition with major AI labs by betting on efficiency over brute force. The model aims to offer strong performance with optimized resource usage, reflecting a broader industry pivot toward cost-effective AI.
2026-06-30
Anthropic launch analysis
Claude Sonnet 5 just made the Opus price gap harder to justify
Claude Sonnet 5 promises near-Opus-level reasoning and agentic performance at lower cost. Early testers report dramatic improvements in autonomous task completion and follow-through on complex, multi-step operations. With introductory pricing through August 2026, the model makes the Opus price gap harder to justify.
2026-06-30
effective context, output ceilings, and the hidden tax of long windows
Your AI model says it can read 1 million tokens. It's lying. Here's the real math.
All four frontier LLMs advertise 1M+ token contexts, but effective recall, output limits, and real-world cost differ sharply. DeepSeek V4 Pro leads in output ceiling and cost, Gemini excels under 200K tokens, and Claude Opus wins on caching for interactive code review. This analysis breaks down the numbers from April 2026.
2026-06-30
Agent Benchmark Analysis
ProgramBench: every public AI scores 0% on the hardest coding test yet
ProgramBench challenges AI agents to reconstruct programs from binaries alone, without source code or issue descriptions. All public models fail to fully resolve any task, exposing weaknesses in probing, architecture, and stopping judgment. The benchmark is a stress test for coding agents moving beyond patch-based workflows.
2026-06-29
Synthetic Data
Synthetic data generation just ditched its weakest link: the boss
Matrix is a decentralized framework that uses serialized messages passed through distributed queues for multi-agent synthetic data generation. By eliminating the central orchestrator, it achieves 2 to 15 times higher throughput on identical hardware.
2026-06-06
Local AI
Ollama 0.30 just made local AI cheaper than cloud inference for more people
Ollama 0.30 boosts NVIDIA inference by up to 20%, enables Vulkan GPU support by default for AMD and Intel devices, and expands GGUF model compatibility, including fine-tuned models from Hugging Face and support for tool-calling with coding agents.
2026-06-05
AI Infrastructure
Nvidia's 55-billion-token trick just rewrote the math on agentic AI costs
Nemotron 3 Ultra, Nvidia's sparse 550B model with 55B active parameters, promises to cut inference costs by up to 30% versus peer open models while keeping frontier reasoning. With a 1M-token context window and native NVFP4 quantization, it targets the cost of multi-step agentic workflows.
2026-06-04
Open-source framework
Stanford just made local AI agents work, and made the cloud look optional
OpenJarvis 1.0 from Stanford's Hazy Research and Scaling Intelligence labs runs personal AI agents locally via Ollama, with cloud access as an optional add-on. It ships with presets for morning briefings, cross-document research, and local code assistants, all while tracking energy cost and latency beside accuracy.
2026-05-28
Cloud Infrastructure
Alibaba laid bare the 14-round problem that's bleeding AI agents dry
Alibaba Cloud's ANOLISA is a three-layer OS upgrade for the agent era, cutting token use by 30% and improving execution time by a similar margin. Designed for high-density agent deployment, it addresses security, observability, and inefficiency at the kernel level.
2026-05-26
Alibaba Cloud
Alibaba's Anolisa v0.3 gives AI agents a rollback button and a security net you can measure
Alibaba Cloud's Agentic OS, Anolisa, reaches version 0.3 with a security module, real-time cost savings visible down to the token, and workspace snapshots that undo agent actions in milliseconds.
2026-05-07
Local AI Inference
Ollama rewrote its Apple engine. The M5 results are finally competitive.
Ollama previews MLX-backed inference on Apple silicon, pushing Qwen3.5-35B-A3B to 1,851 tok/s prefill on M5. The shift brings NVFP4 quantization for production parity and a smarter cache for agentic workflows.
2026-03-30