Research
MiniMax Just Launched M3.1 Flash: 428B MoE Agentic Model at $0.15/M Input
MiniMax has released M3.1 Flash, a 428B-parameter MoE model with about 23B active parameters and a 1M-token context. API pricing is $0.15/$0.60 per 1M tokens, with weights due Oct…
The Download: preserving the Arctic permafrost, and playful AI agents
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Inside the quest to combat Arctic fires a…
Inside the quest to combat Arctic fires and save the planet’s frozen carbon
In the middle of an especially warm July in 2021, a wildfire ignited in a thick forest of black spruce and peeling paper birch trees on the northern campus of the University of Al…
How far can we push computing?
MIT Technology Review Explains: Let our writers untangle the complex, messy world of science and technology to help you understand what’s coming next. You can read more from the s…
NVIDIA Open Sources Run:ai Under Apache 2.0 for Kubernetes GPU Orchestration
NVIDIA has released the full Run:ai GPU orchestration platform as open source under Apache 2.0. The release includes KAI Scheduler, CUDA 13.4 GPU fractioning, quota management, an…
Mistral AI Releases Timestral: A 209M Forecasting Model Family That Outperforms Google’s TimesFM-3
Mistral released Timestral, a time series foundation model family from 5M to 209M parameters. The 209M model leads GIFT-Eval, TIME and fev-bench among zero-shot models and runs on…
TinyFish Makes Search Free: What It Really Costs an AI Agent to Read the Web
An AI agent that reads the web pays 2 kinds of cost. The first is the retrieval charge: what a search or fetch API bills per request. The second is model processing: content passe…
Google Open-Sources Harness2, a Recursive Agent Harnessing Framework
Google Research released Harness2, an Apache-2.0 framework that rewrites an agent's harness at test time. One refinement step lifted GLM-5 by 11.5 points on WorkBuddy-Bench with n…
How to Make Zero-Training Tabular Predictions with NVIDIA Structured Data Models
In this tutorial, we implement Structured Data Models, NVIDIA’s new library that puts several pretrained foundation models for tables behind one PyTorch interface, and use it to m…
Meet Voxel51 3.0: A Unified Multimodal Data Platform for Physical AI
Voxel51 3.0 merges search, curation, annotation and evaluation of multimodal sensor data into one platform for physical AI teams. It adds Voxel51 Agent with 45+ MCP tools, dataset…
Meet Kyutai REKI: The 2.4B VLA Model Bringing Efficient Memory to Robot Manipulation
Kyutai, with Inria and École des Ponts, has released REKI, a 2.4B vision-language-action model (VLA) for robot manipulation. The REKI vision-language-action model is pretrained on…
DeepSeek Just Released V4.1-Flash-1017, Scores 77.0 on DeepSWE, Opens Search API
DeepSeek has just released DeepSeek-V4.1-Flash-1017, a new checkpoint of its V4.1-Flash model. It is live on the DeepSeek website, app and API. DeepSeek claims stronger scores, fa…
20 Jev Alternatives: The Decision Models Taking On TypeSafe in 2026
TypeSafe launched Jev on September 15, 2026. It is a decision model, not a chatbot. You define the allowed answers, and Jev returns a label, a score or a yes/no probability. In ab…
StepFun Releases Step 5 Preview Open Weights: 600B MoE, 27B Active, 1M Context
StepFun has released Step 5 Preview as open weights. FP8 and BF16 checkpoints are live on Hugging Face and ModelScope. Step 5 Preview is a sparse Mixture-of-Experts model from Ste…
Perplexity Open-Sources SPACE, a Firecracker MicroVM Sandbox for AI Agents
Perplexity has open-sourced SPACE, the sandbox platform behind every task in Perplexity Computer. The Perplexity SPACE sandbox gives each agent task its own Firecracker microVM. C…
The Download: failed social media bans, and extending pets’ lives
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Kids are outsmarting social media bans. T…
Can’t bear to say goodbye to your pet? Biotech’s got you covered.
I don’t remember much about Peppy, my first and only pet cat. I can conjure the sensation of stroking her while she purred, recall how she used to get stuck in trees (just like in…
Love is infrastructure
I. Meet cute Barnabas and Mainspring grew up near at hand, about an hour apart with good traffic, but for a long time they weren’t close. They traded glimpses and occasional hello…
This implant can restore sight for patients with macular degeneration
Age-related macular degeneration (AMD) affects more than 5 million people around the world. And until now, no treatment could restore their vision. The Prima system, which was app…
Kids are outsmarting social media bans. That’s not a problem we need to solve.
Social media bans are stacking up. In July, France joined Australia in restricting minors from accessing social media—barring children under 15 from major platforms, before the ba…
Liquid AI Releases d1-ultra: A Multimodal Decision Model That Scores 65 on the AA Decision Index
Liquid AI released d1-ultra, a text and image decision model that returns calibrated probabilities in 1 forward pass. It scores 65 on the Artificial Analysis Decision Index and co…
Meta Releases Muse Spark Open: 816B Open-Weight MoE With 1M Context
Meta released Muse Spark Open, an 816B-parameter open-weight MoE model with 42B active parameters and a 1M-token context. It leads open models on MCP-Atlas and runs on 2 M5 Ultra …
NVIDIA Releases Nemotron 3.5 Super VL, a 120B Open Multimodal Agent Model
NVIDIA released Nemotron 3.5 Super VL, an open 120B-parameter multimodal MoE model with 12B active parameters, 1M-token context and NVFP4 weights that run on 1 DGX Spark, plus 2 B…
CopilotKit Releases CLiF: A Continual Learning Platform That Turns Agent Traces and User Actions into Skills, Evals, and Fine-Tuning Data
Introduction Teams shipping AI agents at scale keep running into the same pattern. Even agents built on top-tier models like Fable and Astra struggle to carry out complex, end-to-…