iCreatAI News
Back to main site
Skip to content
Only First party#Deployment
Nov 9Mon
  1. Lovable Blog33

    The Radish AI team joins Lovable to scale agentic capabilities for businesses

    Lovable has acquired the team behind Radish AI, which built an agentic infrastructure platform for building, deploying, and scaling complex AI agents with privacy and security built in. The team includes Nirman Gupta (formerly of Meta and Bloomberg), Ish Singh (a two-time founder who led AI and data products at Google, JPMorgan, and Goldman Sachs), and Adrian Zgorzałek (12 years at Meta). They will be split between Lovable's Stockholm and London offices, with Gupta leading engineering on agents for businesses from the London office.

  2. Lovable Blog38

    How we moved our frontend off Firestore and built Lovable Realtime

    Lovable built its own WebSocket-based realtime service, Lovable Realtime, to replace browser-side Firestore listeners, letting its Go backend control payload shape and re-evaluate authorization continuously across Firestore and Spanner data. The service handles peaks of over 2 million concurrent subscriptions and delivered about 2.61 billion data and patch frames in a month, with update latency of 66 ms at p50 and 249 ms at p99. The rollout used shadow traffic and a gradual feature-flag ramp to 100% of users without a user-facing incident.

  3. NVIDIA Newsroom37

    NVIDIA, Rebellions and SK Telecom Announce Collaboration to Advance Sovereign AI Infrastructure With NVIDIA NVLink Fusion

    NVIDIA, Rebellions and SK Telecom announced a collaboration, supported by Korea's Ministry of Science and ICT, to advance Korea's sovereign AI infrastructure using NVIDIA NVLink Fusion. Rebellions plans to connect its next-generation R200 NPUs to the NVIDIA platform and use the NVIDIA MGX rack architecture, while SK Telecom plans to deploy these systems in its AI data center alongside its proprietary models and public-facing services, including those under Korea's "AI for All" project. Rebellions' R200 systems are expected to be ready for deployment in 2028.

  4. NVIDIA Newsroom36

    NVIDIA Dynamo Recipes Bring Korea’s Leading Open Models to Production

    NVIDIA is releasing Dynamo-optimized deployment recipes for four Korean open models: A.X K2 from SK Telecom, K-EXAONE 2.0 from LG AI Research, Solar Open 2 from Upstage, and Motif-3 from Motif Technologies. NVIDIA Dynamo is an open source, distributed inference-serving framework that handles request routing and communication for multi-node deployments at data center scale. Optimized for NVIDIA B200 GPUs, the recipes deliver up to 2x performance out of the box.

  5. NVIDIA Newsroom34

    From Chips to Skills, NVIDIA and Korea Expand Sovereign AI Collaboration

    NVIDIA and South Korea's Ministry of Science and ICT are expanding their collaboration across the country's sovereign AI ecosystem, from AI semiconductors and data center infrastructure to talent, startups, models, agents and security. The collaboration lets Korean NPU and XPU companies explore participation in the NVIDIA NVLink Fusion ecosystem and includes NVIDIA training, certification programs, hackathons and technical seminars. NVIDIA also announced a collaboration with SK Telecom and Rebellions, supported by MSIT, to advance sovereign AI infrastructure for Korean models and workloads using NVLink Fusion.

  6. NVIDIA Newsroom36

    Korea Advances Physical AI Across Industries With NVIDIA

    NVIDIA is helping advance Korea's physical AI ecosystem, with industry leaders applying accelerated computing, AI and simulation across semiconductor manufacturing, robotics and autonomous vehicles. Samsung Electronics is expanding its NVIDIA collaboration in semiconductor engineering, including a case where AI homogenization with coarser grids and GPU acceleration cut simulation time from about 18 hours on a 32-core CPU to 41 seconds. Kakao Mobility plans to build a level 4 autonomous driving system using NVIDIA Hyperion and Halos, while LG Electronics demonstrated NVIDIA Isaac GR00T-powered robotic manipulation with its CLOiD robot in a manufacturing environment.

  7. Artificial Analysis Articles40

    Endpoint Accuracy Index v1.1: Tighter long context intervals and expanded model coverage

    Artificial Analysis released Endpoint Accuracy Index v1.1, which updates AA-LCR-25 to dataset version 1.1 with corrected answer keys, an updated equality-checker LLM and 25 repeats per question, narrowing the average 95% confidence interval from ±4.7 to ±2.6 points. Coverage now spans 103 serverless endpoints across six open weights models, adding GLM-5.3 Flash and Muse Spark Open and replacing DeepSeek V4 Pro with DeepSeek V4.1 Flash 1017, while GLM-5.2 leaves. Of the 103 endpoints, 24 score statistically below their self-hosted reference, with causes grouped as output token caps, reduced reasoning, tool call handling differences, and request limits or stream behavior.

Nov 8Sun
Nov 7Sat
  1. vLLM Blog38

    Migrating vLLM to the LibTorch Stable ABI

    vLLM describes migrating its C++ backend to the LibTorch Stable ABI so that updating PyTorch no longer requires rebuilding vLLM, and as of v0.33.0 it publishes LibTorch ABI Stable CUDA wheels. The migration involved moving AutogradFunctions and meta functions to Python, building a separate stable library with the TORCH_TARGET_VERSION flag, registering ops via STABLE_TORCH_LIBRARY with TORCH_BOX, and migrating seven LibTorch-dependent dependencies including FlashMLA, DeepGEMM, FlashAttention-2/3, DeepSelect, FlashKDA and QuTLASS. Verification used torch-abi-audit, now in vLLM CI, and the post notes tradeoffs such as the limited stable API subset and small host-side overhead, with CPU and ROCm builds planned next.

  2. GitHub Blog · AI & ML36

    Beyond code generation: A developer’s guide to agentic engineering

    GitHub reports over 1.2 billion commits in one week this October, more than 400% above a year earlier, while a Faros AI study found pull requests spent 441% longer in review as teams moved from low to high AI adoption. The post outlines an agentic engineering workflow across Plan, Build, Verify, and Observe, describing Copilot features such as Plan Mode, cloud sandboxes, session continuity, stacked pull requests, Copilot code review, Agentic Autofix, and Agent Merge. It advises starting with one frustrating piece of work and putting an agent on it before expanding across the process.

  3. NVIDIA Technical Blog30

    Speed Up Image Generation with FLUX 3 Image Klein on NVIDIA Blackwell GPUs

    The article addresses inference latency when deploying diffusion models in design tools, image editors, and visual production workflows, noting that generating an image requires multiple sequential passes through a model with billions of parameters and that higher resolutions increase attention costs. It introduces FLUX 3 Image Klein, an image editing and generation model from Black Forest Labs (BFL), and covers speeding it up on NVIDIA Blackwell GPUs.

  4. ElevenLabs Blog30

    Voice AI infrastructure: The 2026 stack explained

    This guide describes the layers of a production voice AI stack—speech recognition, turn-taking, language models, orchestration, tools and MCP, knowledge and memory, speech synthesis, telephony, and guardrails, testing, and monitoring—and explains how ElevenAgents implements each, using Eleven v4 as the example voice layer. It covers two build modes (Custom and Realtime), Scribe v2 Realtime transcription at roughly 150 ms, a proprietary turn-taking model, workflow orchestration nodes, MCP tool approval modes, RAG latency reduced from 326 ms to 155 ms, and Eleven v4 Turbo figures of about 100 ms inference latency and about 150 ms time to first speech. It also details Guardrails 2.0, simulation and tool-call tests, A/B experiments, and observability metrics.

  5. OpenRouter Announcements38

    How Sentry Capped Runaway AI Bills Without Restricting Access

    Sentry moved its internal AI usage onto OpenRouter in July 2026, retiring roughly 500 Anthropic and OpenAI keys over about three weeks and switching off direct access to both providers in August. Keys are bound to users or workspaces so spend can be traced to an owner and model, and per-person and per-workspace budgets cap runaway bills. Sentry says the change gave it tighter control and wider model access at the same time, with 45+ apps and 147 models used in the first three months.

  6. ElevenLabs Blog22

    Share your ElevenMusic creations directly to SoundCloud

    ElevenMusic launched a SoundCloud integration that lets users connect their SoundCloud account under Linked Accounts in profile settings, then add a description, genre, and tags to an original track and click "Save to SoundCloud" to upload it to their profile. SoundCloud Artist Pro members can get three months of ElevenMusic free, then $9.99/mo unless canceled; the offer is 18+, ends 31 Oct 2027, and terms apply.

  7. NVIDIA Technical Blog22

    Accelerating Reinforcement Learning Weight Refit with the Reshard Collective

    This post discusses how reinforcement learning post-training generates trajectories by running inference with the current policy model, and how disaggregated setups run inference and training in parallel on separate GPU resources for efficiency. Because the policy model is updated only on the training GPUs, the system must periodically refresh the weights on the inference side, and the post introduces the Reshard Collective to accelerate this weight refit.

  8. AWS Machine Learning Blog40

    Serverless MCP server for AI agent data lake access on Amazon ECS Fargate

    Amazon's North America Sort Center team built a serverless Model Context Protocol (MCP) server on Amazon ECS Fargate that acts as a centralized, agent-agnostic gateway between AI agents and the enterprise data lake. Adding a dataset requires only a metadata JSON file, which the pipeline indexes into Amazon Bedrock Knowledge Bases, while Amazon Cognito authentication, SELECT-only Athena tools, Bedrock Guardrails, and CloudWatch logging provide centralized controls. The architecture separates the tool layer from the agent layer so any MCP-compatible client can share the same tools and access controls in a multi-tenant environment.

  9. AWS Machine Learning Blog58

    SageMaker Inference now supports Multi-Instance GPU to maximize GPU utilization for model endpoints

    Amazon SageMaker Inference now supports NVIDIA Multi-Instance GPU (MIG) partitioning on inference component endpoints, letting you divide a physical GPU into up to seven hardware-isolated partitions and place a different model in each. Partitioning is declared via AcceleratorPartitionLayouts on the endpoint config and AcceleratorPartitionType on each inference component, with no new API operations or IAM actions. In testing, a mig-3g.40gb partition of an H100 delivered 53–89% more output tokens per second and 18–34% lower cost per million output tokens than a full A100-40GB above a workload-dependent concurrency crossover.

  10. Databricks Blog38

    Real-time fraud detection with Feature Views

    Databricks describes how Feature Views in its Feature Store let teams define a feature once and use the same definition for point-in-time-correct training data and managed online-serving pipelines, reducing training-serving skew and data leakage. The walkthrough follows a $900 payment from a Kafka stream through rolling and Sawtooth window aggregations, an on-demand deviation feature, model training and Model Serving, with Unity Catalog providing governance and lineage.

  11. Databricks Blog30

    How public sector organizations can govern AI at scale

    Databricks outlines why public sector AI projects stall on governance questions—model access, data residency, and spend visibility—and describes how its Unity Gateway provides centralized controls over which models, coding agents, and tools teams can use, with per-user or per-team spending caps, platform-level compliance policies, and geographic data residency for hosted foundation models. The piece argues that agencies which solve governance centrally before scaling adoption can match each mission to an appropriate model while keeping data, users, and budgets under control.

  12. AWS Machine Learning Blog42

    Extend Claude Code to remote sessions on Amazon Bedrock AgentCore with security controls

    This post explains how to run Claude Code as a remote agent on Amazon Bedrock AgentCore runtime, where each task executes in its own isolated microVM session inside your AWS account and returns a GitHub pull request for human review. It walks through the accompanying sample-remote-coding-agent, covering the architecture (API Gateway, Lambda, DynamoDB, SQS, AgentCore runtime) and security controls including Bedrock Guardrails prompt screening, short-lived GitHub tokens via AgentCore Identity, private VPC networking, and least-privilege IAM. The sample suits scoped features, bug fixes, and tests on trusted repositories, but not untrusted code or open-ended work, and the agent never merges pull requests.

  13. AWS Machine Learning Blog36

    Taxbit Assist: A grounded, role-aware AI agent on Amazon Bedrock AgentCore

    Taxbit built Taxbit Assist, an embedded AI assistant for its digital-asset tax compliance dashboard, on Amazon Bedrock AgentCore using the Strands Agents SDK, with grounded and cited answers, role and organization awareness, no tax advice, and regional data residency. The post details the architecture, a curated knowledge base with semantic chunking and reranking, identity bound in code rather than by the model, guardrails-based PII handling, and continuous production quality measurement. Taxbit reports the assistant has run in the US and EU since May 2026, with model cost averaging under 20 cents per conversation and about three-quarters of input tokens served from prompt cache.

  14. AWS Machine Learning Blog50

    How Greenlight Financial built an autonomous SOC solution on Amazon Bedrock AgentCore

    Greenlight Financial's Luna, built on Amazon Bedrock AgentCore runtime and memory, autonomously resolves more than 90 percent of its 3,000–4,000 monthly security and risk cases, using a tiered supervisor architecture with over 30 specialized agents and episodic memory for continuous learning. The system triages dark web breach-card alerts, identity anomalies, and financial-risk events in seconds, escalating complex cases to a Tier 2 supervisor and human analysts. Greenlight later extended Luna into financial risk triage within two months by adding new specialized agents and a risk-domain supervisor to the existing infrastructure.

  15. AWS Machine Learning Blog50

    Govern AI agent access with Okta Cross App Access and Amazon Bedrock AgentCore – Part 1

    This AWS post describes using Okta Cross App Access (XAA) as the governance plane and Amazon Bedrock AgentCore Identity as the execution plane so an AI agent can act on behalf of signed-in users, with Okta deciding which agent may act for which user and AgentCore redeeming Okta's single-use ID-JAG for a short-lived access token. It walks through registering the agent in Okta, configuring the credential provider and workload identity, enforcing per-user authority via the groups claim in MCP servers, and reports measured results such as revocation taking effect in about 1 second and AgentCore Identity adding about 0.6 seconds to a roughly 30-second request.

  16. AWS Machine Learning Blog36

    Govern AI agent access with Okta Cross App Access and Amazon Bedrock AgentCore – Part 2

    Part 2 of this AWS post moves the Part 1 Okta Cross App Access solution from one agent and two MCP servers to a topology with Amazon Bedrock AgentCore Gateway as a single front door that redeems Okta's ID-JAG for each target, default-deny Cedar policies in AgentCore Policy binding tools to agents and delegated scopes, and a second agent on Amazon EKS under its own Okta identity. It also describes tracing decisions with AgentCore Observability and CloudWatch Omni alongside Okta's System Log, reports validation results and latency figures, and compares XAA with per-user OAuth consent for applications that cannot accept an ID-JAG.

  17. Google Developers Blog60

    Agent Distraction

    The post reports that 13% of agent file-reading sessions expose the agent to inline TODO(bug/...) or FIXME(bug/...) references, and in 6% of encounters the agent takes follow-up action on that bug ID, 83% of which are unprompted by the developer. Cross-referencing these detours against live bug status, it finds 41% chase already-closed or obsolete tickets while 57% involve still-open bugs where the agent expanded scope, and a fixed-effects regression shows unprompted bug detours add 70% more steps, 77% more tokens and 105% more wait time per session. The authors built agents that scan codebases and remove confirmed-obsolete bug references, reporting thousands removed over six months with over 95% acceptance from human reviewers.

    Why it matters: Offers a quantified, data-backed look at how stale code comments misdirect coding agents, plus a concrete remediation approach.

  18. NVIDIA Newsroom40

    How NVIDIA Open Source Tools Are Helping Researchers Build Next-Generation 6G Networks

    NVIDIA's open source Aerial and Sionna tools, whose adoption has grown 10x in a year with over 1 million downloads of its open source 5G/6G tools and use by more than 8,000 research and academic institutions, are being used by researchers on projects including goal-oriented semantic communications for robots at King's College London, a neural receiver tested by SoftBank, demand-predicting base station sleep at the University of Leeds, and GPU-accelerated channel models in CTTC's 5G-LENA simulator.

  19. Factory Blog36

    Extending the Factory Organization Model with User Groups

    Factory introduced User Groups, which sync from an identity provider via Directory Sync and let admins set usage limits and model access per group, mapped into chosen organizations. Settings resolve per request in a hierarchy where an individual override wins, then the user's group settings in that organization, then the organization default; for limits the highest group limit applies, and for model access a model is allowed if any of the user's groups allows it. Groups are configured in the Enterprise Admin Console, group limits can be managed through the Organization API up to 100 groups per request, and changes are recorded in the audit log.

  20. Baseten Blog70

    Introducing Serverless Loops: frontier RL training for everyone

    Baseten announced the general availability of Serverless Loops, a Tinker-compatible training SDK for supervised fine-tuning and reinforcement learning on frontier open models at 256k context length, with one-click deployment to Baseten Dedicated Inference or weight download in Hugging Face Safetensors format. The release cites early users including NVIDIA, Datadog, LangChain Labs and Ambience Healthcare, and reports NVIDIA's post-trained Nemotron Super 3.5 VL improving intent classification accuracy from 76.8% to 94% and reaching 87% validation F1 on receipt extraction.

    Why it matters: Useful for readers weighing managed post-training options, as it pairs a GA launch with named early adopters and concrete reported results.

Nov 6Fri
  1. Modal Blog27

    How Snorkel runs realistic agent simulations with Modal

    Snorkel uses Modal to power inference for open-weight models it evaluates and to provide GPU and CPU sandboxes for millions of agent simulations per month, peaking at tens of thousands of sandboxes an hour. Modal's VM sandboxes let Snorkel recreate production-like environments, including multi-service systems and GPU-enabled workloads, and its memory bursting feature cut total sandbox costs by around 20% by avoiding over-provisioning. Snorkel also uses Modal's Shared and Dedicated Endpoints to stand up inference for new and fine-tuned models, and plans to evaluate Modal as a training solution.

  2. ElevenLabs Blog55

    Give your Unity NPCs a voice with ElevenAgents

    ElevenLabs released an ElevenAgents Unity SDK that connects a Unity scene to an ElevenLabs voice agent, listening through the player's microphone, playing the agent's voice through an AudioSource, and running the session via the Conversation API. The same code runs in the Editor, desktop builds, and WebGL, with iOS, Android, and XR in beta. The SDK supports dynamic variables, overrides, contextual updates, client tools, and main-thread events, and is installed via the git URL https://github.com/elevenlabs/unity.git#v0.1.0.

  3. OpenRouter Announcements67

    Introducing the New Auto and Auto Code Routers

    OpenRouter launched a rebuilt Auto Router on a new technical foundation that it says outperforms the previous version by up to 4× on benchmarks, with a routing step 3× faster at p90 (0.65s vs 1.96s), alongside a new coding-specialized Auto Code router (openrouter/auto-code) that scores in line with Claude Opus 5.5 across five benchmarks while costing 74% less per task. The new routers weigh eleven factors per request combined with live pricing, provider health, and cache status, and work within governance policies such as guardrails, zero data retention, provider restrictions, allowed model lists, and BYOK keys. The slug openrouter/auto is unchanged, and the cost_tier, allowed_models, and excluded_models settings still work.

    Why it matters: Useful for readers tracking model-routing products, since it details concrete latency and cost claims and explains how routing differs from provider routing.

  4. Browser Use Blog25

    Best Proxy Providers for AI Agents (2026)

    The article compares seven rotating residential proxy providers (Decodo, Oxylabs, FloppyData, Roundproxies, Byteful, DataImpulse, Webshare) on IP pool size, location targeting, sticky session length, entry pricing, trials and traffic expiry, and reports results from browser-agent benchmark tasks such as product research, text extraction, form filling and session continuity. It also covers built-in versus bring-your-own proxies on browser platforms, setup guidance for custom proxies, and advice on checking session continuity, location coverage, target-site access and cost per completed task. The author notes the samples are small and differently timed and do not establish a universal reliability or speed ranking.

  5. Cloudflare Blog34

    The post-quantum future of web privacy

    Cloudflare explains how Privacy Pass, which relies on classical cryptography like RSA and elliptic curves, will need post-quantum replacements, and explores MOLE (Moderation Of unLinkable Endorsements), a design whose endorsements hide the issuer's identity. The post describes how the VOLE-in-the-Head proof system combined with multivariate signatures such as UOV and MAYO yields blind signatures like PoMFRIT (6.7KB, reducible to about 4.3KB with UOV), and how these building blocks can be reconfigured into MOLE endorsements and credentials. It also covers optimizations (fewer hashes, TurboSHAKE, smaller Keccak state, multivariate commitments and PRGs) and notes that multivariate cryptanalysis is still immature, so parameter changes and further study are expected.

  6. Trail of Bits Blog40

    After Binance, check what your signing systems can approve

    This article examines the October 2026 Binance security incident, in which attackers who compromised a DevOps laptop obtained valid signatures from Binance's wallet-signing service and used an EIP-7702 delegation to authorize repeated transfers without further signatures. It argues that transaction execution and monitoring systems must understand newer Ethereum execution paths such as EIP-7702 delegations, session keys, Safe modules and ERC-4337 UserOperations, and offers a four-step checklist covering reduced signing access, treating delegation as an administrative privilege, limiting damage if signing is compromised, and detecting changes in wallet authority.

  7. Amp Chronicle60

    Spaces

    Amp launched spaces, a new way to interact with agents and your team in which a space is a conversation with a single agent whose job is to spawn and coordinate agents for you, so you no longer have to create and organize threads yourself. The post describes internal uses, including building Spaces v1 itself, running a service-accounts launch with monitoring and user refunds, and a security review that split the codebase into 16 areas. To start a space, use the + in the Spaces section of the sidebar, set a title and a project, and tell your agent what to work on.

    Why it matters: It offers a concrete look at a higher-level agent-orchestration workflow, illustrated with the team's own day-to-day examples.

  8. Amp Chronicle50

    Close the Loop

    Amp now offers a feedback widget that lets users send change requests directly from their website: enable it in project settings, add a script to the site's <head>, and allow the site's origin. Visiting the site with ?amp shows the annotation widget, where clicking page elements and writing instructions sends feedback that starts a new thread in the project, and the change is shipped from that thread like any other. After the first use, the widget appears on the site without ?amp until hidden.

  9. Manus Blog34

    How to Build Your First Game with AI: A Step-by-Step Guide

    This guide compares three no-code routes to a first game—AI game builders like Manus, visual event-sheet engines GDevelop and Construct 3, and block-based Scratch—and stresses that game design, not tooling, is the hard part. It provides a fill-in-the-blanks game brief template and a walkthrough of building a 2D robot platformer in Manus, from prompt and Game Blueprint through playtesting, tuning, and publishing a shareable link.

  10. vLLM Blog42

    VeRL-Omni v0.3.0: Async Training and Broader Multimodal Post-Training

    VeRL-Omni v0.3.0 adds asynchronous and decoupled RL execution paths across multimodal workloads, makes the TransferQueue/ReplayBuffer-based V1 trainer the default, and moves the rollout stack to vLLM 0.30.0 and vLLM-Omni 0.30.0. It includes end-to-end recipes for MiniMax H3 audio-video post-training with DiffusionNFT and MiniCPM-o 4.5 AVQA GSPO, where a 4×H800 experiment found separate-async about 9% slower per prompt at natural response lengths but roughly 6.7% faster at 1024 tokens. The release also expands model and algorithm coverage, adds Ascend and AMD ROCm Dockerfiles, and notes upgrade changes including V1 as default, named diffusion media artifacts, and fail-closed attention backends.

  11. Codex Releases22

    0.167.0

    Codex 0.167.0 adds an agent command center opened with `codex agents --no-daemon`, unread-result markers in the command center and session picker, cross-backend session switching from the terminal picker, local image attachment previews, and a Google Cloud Vertex AI setup wizard. Bug fixes keep saved chats visible across model provider changes, preserve transcript history when Ctrl+L clears the terminal, and resolve network approval prompts when their requests end. Documentation and chores cover managed network constraints under [permissions.network], DotSlash manifest updates, and reduced jemalloc arena counts for Linux musl builds.

  12. Sierra Blog40

    Accelerators take care of the basics

    Sierra describes Accelerators, prebuilt packages for high-volume workflows such as order tracking, appointment scheduling, claims intake and ID verification, which come with basic integrations and guardrails and are customized for a business using Ghostwriter. Accelerators are maintained by Sierra but track customer changes separately so updates can be reviewed before being applied, and they connect to a company's existing software systems. The first Accelerators are available today in Agent Studio, with user-built and shareable Accelerators planned.