31
We ran AI agent swarms in Minecraft and Mindustry
Swarms of Claude and GPT agents in Minecraft and Mindustry.
New
The latest blog posts to take off on Hacker News. 3,236 posts.
32
Grace Blackwell needs a new scheduler: 27% more usable GPUs from the same racks
NVL72 puts all 72 GPUs in a rack on one NVLink fabric, but only rack-local jobs get the speedup. How SkyPilot keeps multi-node training and inference on NVLink with topology-aware placement, in-place recovery from GPU failures, and automatic defragmentation.
33
Zombie stats: Where the numbers AI quotes come from
We asked five AI assistants 200 questions that call for a number and followed 1,285 of the statistics they cited back to the source.
34
We're releasing the weights of Mistral Large 4
The weights of Mistral Large 4 (Le Chonk) are now open. A 1T parameter multimodal model, and best open-weight model for coding, trained in Europe.
35
Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5
Google’s Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5, and second only to Claude Opus 5.5 on the Artificial Analysis Coding Agent Index
36
37
Gemini 4 Carbon
Gemini 4 Carbon is our strongest model with state-of-the-art cybersecurity and long horizon capabilities.
38
Dayjob: Tax – evaluating AI agents on 50 professional tax assignments
DAYJOB: Tax evaluates AI agents across 50 realistic tax environments, testing their ability to navigate messy records, exercise professional judgment, and deliver reliable tax work.
39
The Agent SDK
The Agent SDK is the framework we use to build our own managed agents like Bugbot, Security Reviewer, and Rollouts. Define agents in code and deploy them with one command.
40
41
Early progress on accelerating drug development with Claude
Over the past several months, we’ve partnered with six organizations to accelerate curing disease. We’ve also started drug discovery efforts of our own, including programs for rare childhood forms of muscular dystrophy. In this post, we provide an update on our progress.
42
Introducing LFM3: Every Modality for Specialized Intelligence
LFM3 is Liquid AI’s new open-weight model family: 2B-class text, vision, audio, omni, and decision models built to fine-tune and run fast on any device.
43
Genie 4
Genie 4 introduces real-time interactive world simulation with native audio, advanced physics, natural human motion, and steering powered by Gemini models.
44
Cross-request KV cache exposure in the GLM-5.3 Workers AI model (155 responses)
A configuration mismatch in GLM 5.3 on Workers AI triggered an SGLang bug that exposed KV cache pages across requests. Here is how we addressed the issue.
45
Three new discoveries in mathematics, computer science, and physics
OpenAI shares model results on the Kakeya conjecture, Exponential Time Hypothesis, and quantum chromodynamics—and how researchers can verify and build on them.
46
GPT-6.1 Terra
Delivering fast and capable computer use at scale
47
Concerning behavior from a popular open-weight AI model
Discover how untrusted open-weight AI models can potentially leak data, even when self-hosted. Learn the security risks and how to protect your enterprise.
48
Fixing Tool-Call Accuracy in vLLM
Tool call accuracy is an important part of serving open-weight AI models. When your tool calls aren’t accurate, the model builds up on nonsensical context. We’ve been carrying our own fork of vLLM to hit 100% in tool call schema accuracy with Kimi and DeepSeek. Fortunately, vLLM 0.31 added support for constrained…
49
Brown student maps the hidden global labor powering AI
Cunyan Ma’s interactive tool exposes the power structure that extracts low-wage labor from data workers in the Global South to make everyday AI algorithms function.
50
A 10-year-old CNN outperforms frontier LLMs for skin cancer diagnosis
I am a dermatologist. I tested the ability of GPT-6 Astra and Claude Fable 5.1 to diagnosis skin cancer and found that they were less accurate than I expected. I compared to a 10-year old CNN and to my surprise found that this was superior.
51
Cerebras Deploys Trillion-Parameter AI Models
Learn how Cerebras maps trillion-parameter frontier models across WSE-3T wafers to deliver scalable inference at 1,000–2,000+ tokens per second.
52
No signs of AI payoff outside the tech sector
Despite massive investments and widespread hype around artificial intelligence, the technology has delivered zero financial payoff to corporate America outside of tech. Subscribe for daily updates.
53
Dream-1
Today, we're releasing Dream-1, a World Action Model that controls robots it has never seen, performs tasks it was never shown, and runs from $3/hour.
54
Securing our frontier model training environment
OpenAI explains how we strengthened containment for frontier model training and evaluation, including offline environments and high-compute red-teaming.
55
Kimi K3.1 Pro and K3.1 Pro Highspeed
Introducing Kimi K3.1 Pro, our most capable 3T-class open-weight model with 1M-token context and major advances in coding, 3D modeling, and financial research.
56
Gemma-4-26B-A4B with 39-69% less thinking at same accuracy
This is the fifth model in the ThinkingCap series. We reduced how much Google's gemma-4-26B-A4B-it thinks, without compromising the quality of its answers.
57
Gaussian Splats part 3: There's always a bigger scene
Continuing our Gaussian splatting development series, Director of Technology Sam Birley breaks down how we bypassed GPU Scene instance caps, implemented PlayCanvas Streamed SOG LODs, and managed GPU memory budgets for production-ready photogrammetry… bringing us to a point where the renderer has fin
58
The same AI request emits 8.5 times more CO2 in Texas than in Sweden
Swedish AI provider Berget AI releases an open-source method for calculating the carbon emissions of every individual AI request, developed with researchers from SEI, WattTime and Klimatkollen — and customers can now follow their emissions per request in the console.
59
LLMs are making formal software verification feasible at a much larger scale
Claude formally proved security properties in the compiled code of 123 open-source libraries. Along the way, it found bugs that had eluded detection for years.
60
Every AI model has the same "controversial" take on humans
We asked 14 AI models for their most controversial opinion of humans, three times each. 35 of 41 answers were the same one: we care more about belonging than about the truth.