Qwen-Image-2.1 gets Day-0 support in OpenVINO
Day-0 OpenVINO support from @inteldevs! 🥳 Qwen-Image-2.1 can now run with optimizations on Intel hardware. One open-weights checkpoint supports both generation and editing. 👇
All news · Today9 items
#Deployment and engineering
All collected and published items, ordered by time. Picks are a selected subset.
Qwen · X33 Kimi · X32 Kimi K3 launches on Amazon Bedrock
Kimi K3 is now on Amazon Bedrock! Use Bedrock's access, encryption, and audit controls to run coding, document analysis, and extended agent workflows. Explicit prompt caching is supported. Start building with K3 on AWS 👉 https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-moonshot-ai-kimi-k3.html
Simon Willison71 TypeSafe AI releases Jev, a System One model that outputs only floating-point decisions
TypeSafe AI released Jev last week, a new category it calls a System One model (the author prefers “decision models”). It accepts text input but returns floating-point values for categories, yes-or-no decisions, scores, and confidence rather than text.
Latent SpacePicks76 TypeSafe AI founder Diogo Almeida on Jev and RLCD: building System 1 models for software, not chat
In a Latent Space interview, TypeSafe AI founder Diogo Almeida introduces Jev as a System 1 model designed for code consumption and optimized for intelligence per dollar rather than chat experience. Its key innovation is unpublished RLCD (Reinforcement Learning for Calibrated Decisions), which uses calibrated probabilities instead of human-rating feedback.
Why it matters: Almeida revisits the strategic split between RLHF and RLCD and explains why the model is optimized for reliability rather than benchmark scores.
NVIDIA Blog28 NVIDIA launches DSX Ready certification to vet power and cooling products for AI factories
NVIDIA has launched DSX Ready, a certification program that checks whether partner products meet the requirements of NVIDIA's DSX AI factory reference design. The first categories are battery energy storage systems (BESS) and coolant distribution units (CDUs).
NVIDIA Blog22 Why NVIDIA says scaling physical AI requires a full-stack safety system
NVIDIA has introduced Halos, which it calls the first and only full-stack safety system for physical AI, spanning design, verification, and deployment.
NVIDIA Blog20 Egypt's NVIDIA AI ecosystem moves toward production as DLI enrollment grows more than tenfold in a year
At an event at the Grand Egyptian Museum, NVIDIA showcased Egypt's AI ecosystem moving from skills development to production deployments. Enrollment in NVIDIA's Deep Learning Institute in Egypt has grown more than tenfold in a year.
NVIDIA Blog36 NVIDIA on AI safety: engineering controls at every layer of the agent stack
NVIDIA argues that AI safety is an engineering problem requiring enforceable controls, clear ownership, and verification evidence, and says safety boundaries must operate independently of agent reasoning. Its open-source NVIDIA OpenShell safety runtime enforces policies outside an agent's reach and provides sandboxed execution. Cisco DefenseClaw and JFrog have built governance and skill-scanning capabilities on top of it.
NVIDIA Blog17 NVIDIA spotlights five companies using AI to advance clean energy at New York Climate Week
At New York Climate Week, NVIDIA highlighted five companies using AI to advance clean energy. ThinkLabs AI uses digital twins and agents to reduce the evaluation time for grid-connection applications from 30–45 days to two minutes.
Hugging Face Blog48 Hugging Face releases tokenizers v1: benchmarks show major encoding and decoding speedups
Hugging Face has released tokenizers v1, producing exactly the same token IDs as v0.23 while running up to dozens of times faster. The project splits a single crate into a workspace and introduces bitcannon, which replaces regex tokenization with SIMD bit operations, along with an allocation-free merge loop, thread-local word cache, and native multithreaded parallelism.
Simon Willison18 Simon Willison releases llm-keys-ui 0.1, an API-key setup interface for Codex Remote
Simon Willison has released the llm-keys-ui 0.1 plugin, which configures API keys on a remote machine without requiring users to paste them into the ChatGPT app.
Z.ai 智谱 · X60 Zhipu explains how GLM-5.3 helped build and optimize inference infrastructure for GLM-5.3-Flash
Zhipu describes how GLM-5.3 was used to help build and optimize the inference infrastructure serving GLM-5.3-Flash. The system went from first working run to production readiness in under two weeks, and end-to-end throughput tripled from the initial baseline.
GitHub Blog · AI & MLPicks78 GitHub uses Copilot to rewrite the Copilot runtime in Rust
Using the GitHub Copilot app and Copilot CLI, GitHub completely rewrote the Copilot agent runtime from TypeScript into more than 800,000 lines of production Rust. AI agents wrote most of the code, merged incrementally into main across 128 pull requests rather than in a single cutover.
Why it matters: Drawing on firsthand experience, the author reviews the 800,000-line Rust rewrite and shares agent-orchestration and review practices that can be reused.
NVIDIA BlogPicks62 NVIDIA Vera Rubin NVL72 makes its MLPerf Inference v6.1 debut with leading performance
NVIDIA submitted preview results for Vera Rubin NVL72 for the first time in MLPerf Inference v6.1, reporting up to 3.7× the throughput of GB300 NVL72 on Qwen3-VL and up to 2.5× on DeepSeek-R1.
Why it matters: MLPerf Inference v6.1 reports throughput, scaling efficiency, and software-optimization gains for Vera Rubin NVL72 versus GB300 NVL72, offering a basis for judging inference infrastructure value.
NVIDIA Blog42 Emerald AI, Google, and NVIDIA launch AI Energy Management Alliance
Emerald AI, Google, and NVIDIA have announced the AI Energy Management Alliance (AEMA), which aims to help data centers adjust power use dynamically to grid conditions. The alliance uses technology-neutral, performance-based requirements focused on measurable factors such as response speed, duration, predictability, and emergency behavior. It also sets out pre-interconnection load-reduction and emergency-response obligations, common technical standards and data sharing, faster paths to grid connection, and cost allocation based on actual system impact.
Latent Space70 TypeSafe releases Jev decision model, claiming over 100× the speed and 200× lower cost of small frontier LLMs
TypeSafe has released Jev, a non-autoregressive model trained with RLCD for decisions, classification, routing, and scoring. The company claims it is more than 100× faster and over 200× cheaper than small frontier LLMs, with no charge for output tokens.
NVIDIA Blog38 University of Manchester uses NVIDIA Earth-2 to forecast air pollution across the UK
A University of Manchester team trained a UK-wide air-pollution model using Earth-2 CorrDiff, a generative model from NVIDIA Earth-2. Training took two days on a single eight-GPU node on Isambard-AI, producing forecasts at a resolution of 2–3 square kilometers.
NVIDIA BlogPicks62 NVIDIA details DSX platform: MaxLPS delivers 24% more token throughput on the same power budget
At the AI Infra Summit, NVIDIA presented its DSX platform. DSX MaxLPS was validated on Lambda's five-rack, 19-node HGX B200 cluster, where it increased token throughput by 24% under the same power budget—from about 4 million to 5 million tokens per second—and improved performance per watt by 23%.
Why it matters: The article reports DSX MaxLPS measurements on a Lambda cluster, allowing readers to judge possible throughput gains under a fixed power budget.
NVIDIA BlogPicks72 NVIDIA shares efficiency progress for Vera Rubin and DSX at AI Infra Summit
At the AI Infra Summit, NVIDIA announced several advances in Vera Rubin and its DSX platform, shifting AI infrastructure measurement from peak performance to useful agentic tokens per megawatt.
Why it matters: The article reports several measured token-output figures per megawatt for Vera Rubin and DSX, showing how efficiency metrics for AI factories are changing.
NVIDIA Blog34 How Children's Hospital of Philadelphia uses open NVIDIA AI for heart modeling and surgical simulation
Children's Hospital of Philadelphia (CHOP) uses MONAI, an open-source medical-imaging framework co-founded by NVIDIA, to turn existing CT, MRI, and 3D ultrasound scans into anatomically accurate pediatric heart models in seconds. The process previously took a skilled researcher four hours.
GitHub Blog · AI & ML34 How GitHub uses Copilot to turn marketing campaigns into code, from planning through follow-up
GitHub's Japan and Korea marketing lead automated event operations with GitHub Copilot. GitHub Issues serve as work units, issue forms as registration forms, labels as triggers, and Actions as task runners. A single issue automatically sets up an event, screens applicants daily, and handles wrap-up when the event ends.
OpenAI News38 How OpenAI scaled Habitat into an online storage platform for one billion ChatGPT users
OpenAI evolved Habitat from a Python library into a globally distributed storage platform supporting one billion ChatGPT users and 22 million requests per second. The platform meets ChatGPT's online-storage needs at scale.
Mistral AI26 Cloudera and Mistral partner to bring dedicated sovereign intelligence to enterprise data
Mistral and Cloudera are partnering to integrate Mistral models into Cloudera's hybrid data platform. Enterprises can run inference in private or public clouds, on premises, or in fully air-gapped environments while retaining full control. They can also train their own models on proprietary data in controlled environments while keeping ownership of both data and models. Cloudera's platform hosts 30 exabytes of customer-managed data.
Hugging Face Blog49 Asynchronous GRPO with LoRA on Hugging Face Jobs: one bucket, one proxy, no NCCL
TRL v1.14's AsyncGRPOTrainer can now train and sync only LoRA adapters to vLLM. Rank-1 adapters are just a few megabytes and can be transferred through a Storage Bucket mounted across multiple Hugging Face Jobs, without NCCL.
Mistral AI58 How Mistral uses AI agents to migrate 40,000 lines of legacy Fortran
Mistral helped a European energy operator migrate a 40,000-line Fortran 77 reservoir simulator to C++. The codebase had no test suite or centralized documentation. The team first built a numerical-equivalence testing framework, then used a Skill.md to guide agents in exporting Fortran state snapshots and validating them in C++. A custom parser generated the call tree, and Vibe CLI coordinated hundreds of agents with Mistral OCR to fill documentation gaps.
Kimi · X40 Kimi Work launches remote control
Remote Control is now available in Kimi Work. Keep Kimi Work running on your computer and use your phone to move tasks forward. Work with Kimi Work wherever you are.
GitHub Blog · AI & MLPicks71 GitHub previews Project HydraFusion for multi-model orchestration
GitHub has released a research preview of Project HydraFusion, enabled in GitHub Copilot CLI with `/experimental` and selectable through `/model`. Usage is billed per token at each model's standard rate.
Why it matters: GitHub reports quality and cost comparisons across three orchestration modes and three benchmarks, helping readers weigh multi-model orchestration for coding tasks.
Hugging Face Blog40 Hugging Face releases NeoMME, 260M and 800M multimodal multilingual encoders
Hugging Face has introduced the NeoMME family of multimodal multilingual encoders in 260M and 800M sizes. A single bidirectional Transformer directly processes text tokens and raw image patches, without a separate vision tower or causal language model.
Hugging Face BlogPicks62 Hugging Face releases @huggingface/kernels with 207 WebGPU kernels and a Fleet benchmarking tool
Hugging Face has released @huggingface/kernels, a JavaScript library for loading and running optimized WebGPU kernels from the Hub, along with an initial collection of 207 kernels. Each is published in a separate Apache-2.0 repository with a manifest, correctness tests, benchmarks, and a WGSL shader template.
Why it matters: Hugging Face has released 207 WebGPU kernels with contracts, tests, and benchmarks, offering a view of how reusable the foundations of browser inference have become.
Hugging Face BlogPicks62 Hugging Face introduces gr.Workflow in Gradio, turning AI pipelines into a drag-and-drop canvas
Hugging Face has introduced gr.Workflow in Gradio, representing multi-step AI pipelines as a graph with three node types—references, operators, and agents. Gradio provides a drag-and-drop canvas where every node can run and every intermediate result is visible.
Why it matters: The article details gr.Workflow's node types, REST endpoints, and deployment path, helping readers assess how to implement multi-step AI pipelines.
Mistral AI · X38 Mistral and HUMAIN announce strategic partnership
Today we announced a strategic partnership with @HUMAIN spanning AI infrastructure, advanced model development, and deployment of AI solutions in Saudi Arabia and the broader region. We will jointly develop localized frontier AI models, initially focused on cybersecurity, voice, and strong performance in Arabic-language markets. https://mistral.ai/news/mistral-x-humain/
Mistral AI36 Mistral and HUMAIN partner to advance sovereign AI in Saudi Arabia and the Middle East
Mistral and HUMAIN have announced a strategic partnership covering AI infrastructure, advanced model development, and AI deployment, initially focused on cybersecurity and voice, with plans for frontier models that perform strongly in Arabic. The partnership is worth hundreds of millions of euros. Mistral will explore using HUMAIN's data-center infrastructure, and the two companies will develop a joint go-to-market strategy for regulated industries in Saudi Arabia.
Thinking Machines · X47 We want to improve Inkling’s agentic performance. To help us understand its real-world behavior, we are making it available for free on OpenRouter (only with agentic harnesses) for the next few weeks, starting now. We’ll use the data, disassociated from accounts, to better it.
We want to improve Inkling’s agentic performance. To help us understand its real-world behavior, we are making it available for free on OpenRouter (only with agentic harnesses) for the next few weeks, starting now. We’ll use the data, disassociated from accounts, to better it.
Hugging Face Blog34 How Hugging Face Inference Endpoints, Jobs, and Buckets power Papers with Code search
Papers with Code's hybrid search system is powered by Hugging Face Jobs, Storage Buckets, and Inference Endpoints. It currently maintains embeddings for more than 110,000 papers from arXiv and Daily Papers.
Hugging Face Blog56 Liquid AI releases LFM2.5-DSpark draft models, accelerating inference by up to 3.2×
Liquid AI has released DSpark draft-model checkpoints for LFM2.5-1.2B-Instruct, LFM2.5-2.6B, and LFM2.5-8B-A1B. Speculative decoding increases decoding speed without changing output quality, with up to 3.18× higher GPU throughput and up to 2.87× speedup on-device.
Mistral AIPicks71 Mistral releases Agentic Search, boosting retrieval accuracy by up to about 3×
Mistral has released Agentic Search, a multi-step retrieval layer for complex enterprise documents. It gives the model five tools—search, open, navigate, read, and grep—to inspect, navigate, and verify sources before answering.
Why it matters: The source reports accuracy, token usage, and latency for a multi-step retrieval loop on FinanceBench and OfficeQA Pro, showing one possible alternative to conventional RAG.
Jeremy Howard · X50 Sentence Transformers v6.0 adds support for ColBERT-style late interaction models
Jeremy Howard highlights support for @answerdotai’s ColBERT model in Sentence Transformers, describing it as amazingly fast, extremely accurate, and only 30M, with local embedding index creation and querying directly in Python. In the quoted announcement, @tomaarsen releases Sentence Transformers v6.0 and introduces MultiVectorEncoder as a first-class model type for ColBERT-style training, inference, and interpretation alongside dense, sparse, and reranker models.
Hugging Face Blog48 Hugging Face introduces constraint-aware GPU allocator, improving cluster utilization by up to 33 percentage points
Hugging Face built a constraint-aware GPU allocator and compared it with a FIFO scheduler across seven benchmark scenarios. With the same hardware and workloads, GPU utilization improved by up to 33 percentage points, while priority-weighted throughput rose in every scenario by up to 105%.
Hugging Face BlogPicks62 Closing the recording-to-training-to-deployment loop with Strands Agents, LeRobot, and Hugging Face Storage Buckets
A Hugging Face blog post presents a streaming-data loop with Strands Robots: the same `Robot()` records a LeRobotDataset and syncs it to a Storage Bucket, then `stream_dataset()` reads directly from the Hub for training without downloading the full dataset first. To deploy a trained checkpoint back to hardware, users only need to switch the mode to `real`.
Why it matters: The Strands Robots example breaks down a robotics data loop and shows how recording, streaming training, and deployment can share one LeRobot format.
Mistral AIPicks62 Mistral launches regional inference endpoints and Priority Tier, and adds third-party open models including GLM-5.2
Mistral has announced three steps toward AI sovereignty. Mistral Regional Endpoints are now generally available, letting customers choose whether inference runs in Europe or the United States. Mistral Priority Tier is in public preview, offering custom rate limits and SLA-backed uptime commitments for mission-critical workloads.
Why it matters: Mistral brings regional inference endpoints, a priority service tier, and third-party open models together on one infrastructure platform, offering a view of sovereign AI deployment.