iCreatAI News
Back to main site
Skip to content
Only First party#Open source ecosystem
Oct 30Fri
  1. AWS Machine Learning Blog58

    How MoEngage turned over 400 APIs into agent tools with AgentCore Gateway

    MoEngage used Amazon Bedrock AgentCore Gateway to expose roughly 400 internal REST APIs across seven AWS Regions as permission-scoped agent tools, avoiding over 20 separate MCP servers and per-API networking. Cedar policies filter tools server-side per agent identity, managed egress over Amazon VPC Lattice keeps APIs private, and teams expose operations by annotating OpenAPI specs rather than writing MCP code. MoEngage estimates the approach saved about six months of platform engineering effort, with new API onboarding now taking hours instead of days.

Oct 29Thu
  1. Hugging Face Blog80

    Welcome Le Chonk: Mistral Large 4 on Hugging Face

    Mistral Large 4 ("Le Chonk") is a multimodal, multilingual Mixture-of-Experts model with about 1T total and 49B active parameters and a 1M-token context window, released in FP8 and NVFP4 variants under the Mistral Ecosystem Open-Weight License. It combines a 72-layer text decoder using MLA with a DeepSeek Sparse Attention indexer and a 1.6B Pixtral vision encoder, and the post reports 82% on the Artificial Analysis Cyber Index vulnerability reproduce-and-patch test and 93% on Cybench. Day-0 support covers Transformers, vLLM, llama.cpp, TRL fine-tuning, Inference Endpoints and Dell Enterprise Hub, with GGUFs from Unsloth and the llama.cpp team.

    Why it matters: A detailed, hands-on release guide that pairs architecture specifics and benchmark claims with runnable deployment and fine-tuning recipes for a large open-weight model.

  2. Hugging Face Blog44

    Announcing Project ProtoStar: Making Publicly Funded Science AI-Ready

    Project ProtoStar is a new initiative bringing together universities, industry, and the open science community to release publicly funded scientific datasets, models, and code in AI-ready, machine-actionable formats on open infrastructure. Founding partners include Carnegie Mellon University, Purdue University, Georgia Institute of Technology, Arizona State University, and The George Washington University through the Open Forum for AI, the Astera Institute, Hugging Face as infrastructure partner, and ORCA as community partner. The first open challenge, hosted by Carnegie Mellon University from November 2–13, 2026, asks participants to use historical weather data from 210 U.S. cities through 2018 to predict 2019–2026 climate conditions and propose a climate adaptation policy idea.

  3. PyTorch Blog47

    PyTorch 2.15 Release Blog

    PyTorch 2.15 makes the rewritten NCCL backend the default for all nccl process groups, with a flight recorder that dumps diagnostics on hangs and TORCH_DIST_USE_NCCL2=0 or nccl-legacy as rollback paths. It adds one-sided transports for torch.distributed with a NIXL backend, finer FSDP2 and pipeline parallelism memory and communication controls, NVFP4 decode improvements on Blackwell, FP8 and faster decode on Apple Silicon, and broader platform support including FlyDSL on AMD gfx950, ROCm 10.0 wheels, CUDA 13.2 as the default wheel, and torch.compile on Python 3.15. The release contains 3,117 commits from 583 contributors since 2.14, and Python 3.10 and macOS 14 are no longer supported.

  4. AWS Machine Learning Blog36

    Distributed training with Ray Train DLCs and EFA on EKS

    This AWS post shows how to fine-tune Qwen2.5-1.5B with PyTorch FSDP across eight NVIDIA L4 GPUs on two EFA-enabled g6.12xlarge instances on Amazon EKS, using the published Ray Train DLC (public.ecr.aws/deep-learning-containers/ray:train-ml-cuda-v1.1) without building any image. KubeRay creates the Ray head and worker pods from one RayCluster manifest, while hardware, transport and per-job Python packages are configured from outside the image via resource requests, environment variables and Ray runtime_env. The sample repository includes idempotent deploy scripts that locate g6.12xlarge capacity, provision the cluster and node group, install the GPU and EFA device plugins and KubeRay, and submit the job from inside the head pod, with the script failing if NCCL did not use EFA.

  5. Baseten Blog55

    The fastest, lowest-cost batch transcription with Qwen3-ASR

    Baseten optimized Qwen3-ASR 1.7B, an Apache 2.0 open-weight speech recognition model supporting over 30 languages and 22 Chinese dialects with a 5.4% average WER on the Open ASR Leaderboard, for batch transcription on a single RTX PRO 6000. The optimized stack transcribes 7x more audio per second than out-of-the-box vLLM at high concurrency, reaching 2,645x real-time factor on 5-minute clips and 3,128x on hour-long files, with median latency for 5-minute clips at 64 concurrent requests falling from 50 seconds to 7.3 seconds. Baseten reports a cost of about $0.0013 per audio hour, 8x lower than the same model on vanilla vLLM and roughly 277x less than popular closed-weight providers.

Oct 28Wed
  1. Pydantic AI Articles22

    I received the AAIF Outstanding Technical Contribution Award

    The author received the Agentic AI Foundation's Outstanding Technical Contribution Award at AGNTCon + MCPCon North America for their work as one of the two maintainers of the MCP Python SDK, including the v2 release shipped alongside the 2026-07-28 MCP specification revision with a new MCPServer API and multi-round-trip requests. They also sponsored SEP-1288 for the WebSocket transport and thanked colleague Laís Carvalho for the nomination and Pydantic for funding their open source work. The post notes that the Logfire MCP server runs on the same Python SDK and is where this work lands first.

Oct 27Tue
  1. Trail of Bits Blog36

    Backporting security fixes to stable releases

    Trail of Bits argues that stable and LTS software often lacks backported fixes for known CVEs, leaving users exposed to exploitation by AI agents, and cites kernel and Debian/Ubuntu CVE counts as a measure of the gap. The post describes using Codex with GPT-5.6 to find fixes that apply cleanly to stable kernel trees, with 142 fixes covering 90 CVEs now applied, and says the team has begun sending backported patches to Linux distributions. It invites open source projects and distributions to apply at trailofbits.com/patch-the-planet for help.

  2. n8n Blog55

    Introducing n8n 3.0

    n8n released version 3.0.0 BETA, focused on internal changes for reliability, stability and security, with a stable release planned for early December. The release includes breaking changes such as removal of legacy nodes (Function, Item Lists, Cron, Interval), Docker-only self-hosting, safer security defaults, and a Migration Report under Settings to identify affected workflows and settings. Upgrading to 3.x is one-way, so users are advised to test on a staging instance and take a full backup first.

  3. Cloudflare Blog74

    Cross-request KV cache exposure in the GLM-5.3 Workers AI model, and our remediation with SGLang

    Cloudflare reports that from August 27 to September 9, 2026, 155 responses from the GLM-5.3 Workers AI model (@cf/zai-org/glm-5.3) to prompts longer than 681,280 tokens contained data from other customers' requests, caused by a prefill KV pool smaller than the 1,048,576-token context window combined with an SGLang bug that let rejected over-length requests proceed through disaggregated prefill/decode handoff. The model was taken offline on September 9 and restored the same day after a corrected configuration and an engine patch, and Cloudflare says it found no evidence of deliberate exploitation. Cloudflare recommends operators of disaggregated SGLang upgrade to v0.5.22 or later and verify KV pool size against context length on every node.

    Why it matters: A detailed first-party incident report that is useful for anyone operating disaggregated LLM inference, with concrete configuration checks and upstream patch references.

  4. Meta Engineering40

    Meta Networking at the 2026 OCP Global Summit: Open Innovation for Scale-Up, Scale-Out, and Scale-Across AI Networks

    At the 2026 OCP Global Summit, Meta presented networking work supporting its AI models and gigawatt-scale Prometheus cluster, including the liquid-cooled IcePack/IceCube rack with 102.4T switch blades, the Flattened Planar Fabric for scale-out, BAG super-spine and SRv6 backbones for scale-across, and Tahan, an Ethernet scale-up switch for MTIA 300. Meta also introduced OSPA, an OCP initiative co-led with Microsoft and Alibaba to standardize the switch platform layer with a v1.0 spec submitted to OCP, and said its MetaRoCE RDMA transport is going open source, starting with a user-space version on GitHub. Recorded talks and slide decks from the event are available on the 2026 OCP Global Summit website.

Oct 23Fri
  1. AWS Machine Learning Blog34

    Train Qwen3 on AWS Trainium with torchtitan

    This post walks through training Qwen3 three ways on a single trn2.48xlarge using torchtitan-neuron: via Hugging Face Transformers v5 as a modeling backend, via torchtitan's native Qwen3 by changing config.model_spec, and via NKI mixture-of-experts kernels from the torchtitan-neuron recipes, all using the same torchrun launch pattern. It also shows how to load and save published safetensors checkpoints, attach a custom NKI rope_hf kernel through torchtitan's override registry, and scale with data and tensor parallelism across 64 logical NeuronCores.

  2. NVIDIA Blog34

    NVIDIA Open Models Power a New Wave of European Public Sector Projects

    NVIDIA partners and European governments are adopting open models for public-sector AI, including the European Commission's deepset-based AI@EC platform, Poland's healthcare and employment agentic systems, and Germany's Materna Sinfonie and secunet sovereign cloud. New Nemotron-Personas datasets for the UK and Germany provide 1 million synthetic personas each for testing public-service models, while France's CANUT selected LINAGORA and Docaposte for a sovereign workspace and Greece's KIEFER builds an agentic factory on DGX SuperPOD. NVIDIA's Open Models Codefest named Pleias, Zehnlab.ai, Disseqt, Algoritmica GmbH and ABA Technology as its top five teams.

Oct 22Thu
  1. NVIDIA Blog40

    Leading European Financial Services Organizations Embrace Open Models to Build Proprietary Intelligence

    At GTC Berlin, European banks, payment companies and data providers showcased production-scale AI built on NVIDIA accelerated computing and open weight models running on their own infrastructure, including Revolut's PRAGMA foundation model serving an initial 80 million customers and Starling Bank's agentic model trained on U.K. sovereign cloud. Open weight models let institutions like bunq, NatWest, Modulr and Santander build proprietary intelligence at controlled cost, while LSEG, RavenPack, HSBC and Stripe reported speedups such as 10x faster tick-data analysis and 42x faster data pipelines for fraud detection.

  2. NVIDIA Blog60

    NVIDIA Cosmos Opens New Tools to Train, Evaluate and Deploy World Models

    NVIDIA expanded its Cosmos platform with new training tools, simulation checkpoints and Cosmos Data Factory blueprints, including five cloud-native blueprints for tasks such as autonomous vehicle dataset generation and defect image generation. Cosmos Framework now extends training, optimization and deployment tools as agent skills to open models like GR00T, Alpamayo, QWEN and Pi0.5, while new Cosmos 3 Nano simulation variants and updated NIM microservices offer up to 2x faster inference. The post cites Porsche, ANYbotics, Ultralytics and ADNOC as users of the platform.

    Why it matters: The post details concrete tooling, model support and named customer deployments, useful for readers tracking physical AI and world-model infrastructure.

Oct 21Wed
  1. NVIDIA Blog60

    Bringing the Spark to Europe: Local Intelligence Powers Next Wave of Agents

    At GTC Berlin, NVIDIA showcased its local AI hardware lineup: RTX Spark for personal agents on Windows PCs, DGX Spark for Linux developers, and DGX Station for enterprise workloads, with a Windows version due in December. Announcements included Underdog Desktop Agent for RTX Spark, DGX Spark 64GB systems rolling out Oct. 23, support for clusters of up to eight DGX Sparks, NVIDIA Sync Model Launcher, and the new Recursive Flint local coding model. NVIDIA also added day-zero support for new Nemotron, GLM, DeepSeek, Qwen, MiniMax and FLUX.3 models, plus accelerations for llama.cpp and vLLM.

    Why it matters: It offers a detailed roundup of NVIDIA's local AI hardware and software announcements for developers and enterprises.

  2. NVIDIA Blog62

    NVIDIA Agent Toolkit Expands, Grounding Enterprise Agents in Company Data

    NVIDIA has expanded its open source Agent Toolkit, adding the Agent Knowledge Foundation with CUDA-X libraries and an Auto-Ontology library, Nemotron Retriever and Kumo open models, the OpenShell secure agent runtime, NemoClaw blueprints and tools such as NeMo Relay and NeMo Compass. The company says enterprises can use Agent Knowledge Foundation without moving data, rebuilding workflows or changing governance controls, and cites adoption by Palantir, Databricks, Snowflake, SAP, ServiceNow, Zoom, AWS and others.

    Why it matters: Useful for readers tracking how vendors are packaging data grounding, runtime security and governance for enterprise agents.

Oct 7Wed
  1. Google DeepMind72

    EmbeddingGemma 2: an open, lightweight multimodal embedding model

    Google DeepMind announced EmbeddingGemma 2, a 740-million-parameter model built on the Gemma 4 architecture and released under the Apache 2.0 license, which maps text, code, images, video, and audio into a unified embedding space for on-device use. It offers modular encoders (as little as 270M parameters for text-only, with optional 170M vision and 300M audio encoders), Matryoshka Representation Learning to truncate vectors from 768 down to 128 dimensions, an 8K token context window, and on a Google Pixel 11 Pro requires as little as ~191MB active RAM for text-only weights and ~567MB for the full multimodal model. The company reports a 9.92-point improvement on MTEB Code (68.76 to 78.68) over EmbeddingGemma and says the model is available on Hugging Face and Kaggle, with support across tools including transformers, vLLM, llama.cpp, Ollama, and Qdrant.

    Why it matters: It documents a new open, commercially licensed embedding model with concrete size, memory, and benchmark figures relevant to developers building on-device search and RAG.

Sep 16Wed
  1. Mistral AI58

    Mistral and Mozilla are bringing open, private and multilingual AI to your web browser

    Mistral announced a partnership with Mozilla under which its models now power Firefox Smart Window (beta), Mozilla's AI browsing assistant, for users in France and North America, with the UK and Germany expected later this year. Mistral says it fine-tunes models on regional languages and dialects, and that Smart Window conversations are not saved on Mozilla's servers by default, with partners agreeing to zero data retention. Mozilla CEO Anthony Enzor-DeMeo and Mistral CEO Arthur Mensch framed the deal as two open source advocates working together on privacy, control and choice in AI-powered browsing.