iCreatAI News
Back to main site
Skip to content
Only First party#Agent
Oct 30Fri
  1. AWS Machine Learning Blog55

    Build a repeatable PHP-to-Java migration engine with Kiro

    This AWS post describes a PHP-to-Java migration engine packaged as versioned Kiro skill files (php-model-discovery and php-to-java-migration) that can be run either interactively in Kiro or agentically at portfolio scale via AWS Transform. The engine uses a dependency-free PHP scanner to emit a JSON manifest of models, fields, relationships and review flags, then follows an eight-phase, parity-first workflow in which a parity harness is built before any Java is written and the PHP code serves as the specification. A sample project migrates four Laravel Eloquent models to Spring Boot 3.4.1 on Java 21 with Maven and H2, and the post notes the parity harness was not built for that slice, calling it "the single biggest gap."

  2. Baseten Blog30

    Baseten and Postman: Building reliable, cost-efficient agents for the enterprise

    Postman's Agent Stack (Passport, Fabric, Astropods, Fern) provides tools to build, secure, and govern enterprise agents, while Baseten serves as its recommended inference partner, offering fast model APIs, multi-cloud capacity management with zero data retention by default, and post-training infrastructure. Baseten cites seven 2026 model launches ranking in the top two on Artificial Analysis for throughput and latency, and notes Harvey post-trained a 27B open model to match closed-source frontier performance on its legal agent benchmark. The post describes post-training via supervised fine-tuning and reinforcement learning, with customers retaining model weights, evaluations, and training scripts.

  3. Codex Releases22

    0.166.0

    Version 0.166.0 adds keyboard-driven transcript selection with word, line, and page motions plus integrated search, renders Markdown in async questions and transcript replies, and adds GPT-6.1 Terra to the bundled, Vertex AI, and Amazon Bedrock model catalogs. It also introduces `codex queue --resume`, clearer safety-block explanations with safer follow-up prompts, and `codex doctor --network-only` for focused network diagnostics. Bug fixes preserve acknowledged permissions on resume/fork, restore spawned descendant threads when unarchiving a parent, and improve TUI reconnection recovery.

  4. 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.

  6. Meta Engineering58

    Catching Bugs Before They Land: JiTTesting’s Human-on-the-Loop Approach at Full Scale

    Meta reports that its Catching JiTTests system, which uses LLMs to generate executable tests that fail on risky code diffs before they land, is now part of its production infrastructure rather than a research prototype. The system combines intent-aware and "dodgy diff" test-generation workflows with an LLM-as-Judge ensemble and a rule-based assessor, and uses a human-on-the-loop model where engineers review outcomes at an aggregate level. Meta says 589 engineers (64% of outreach) engaged with the system, which showed a 60% true-positive rate and helped catch 74 serious privacy bugs at the diff stage in the first half of 2026, lowering privacy SEVs by 20%.

  7. AWS Machine Learning Blog41

    Accelerate agentic AI with NVIDIA Nemotron models on Amazon EKS and SageMaker HyperPod EKS

    NVIDIA Nemotron 3.5 Super VL is a 120B-parameter mixture-of-experts model activating 12B parameters per token that handles text, image, document, and video input, joining other Nemotron models already on AWS. The post describes its hybrid Mamba-2, MoE, and attention architecture with a RADIO vision encoder and up to 1M-token context, then shows deployment on Amazon EKS, EKS Auto Mode, and SageMaker HyperPod EKS using NVIDIA Dynamo and vLLM, with validation and performance results on B200 and H200 instances.

  8. NVIDIA Blog40

    Into the Omniverse: How Developers Turn Ideas Into Simulations With Frontier AI Agents

    NVIDIA describes a series of developer projects in which frontier AI models such as GPT-6 Astra are directed through natural-language instructions to build simulations using Omniverse libraries for GPU-accelerated physics, rendering and sensor simulation. Examples include recreating an OP-1 synthesizer in Houdini, testing robotic disassembly in Isaac Sim, reconstructing buildings from site scans, and a gas-turbine what-if application. Other projects cover SimReady asset scene assembly, warehouse performance scaling, cable routing, CAD-to-SimReady conversion, humanoid simulation, autonomous-driving testing, sensor-based digital twins, Robo Olympics, an ISS browser model and room reconstruction.

  9. Suno Blog40

    Meet Chat: Our In-house Studio Assistant

    Suno Studio introduces Chat, an in-house assistant that can create stems and plugins, rename tracks based on clip names, lower all faders by a set amount such as 3dB, add Compressor and EQ effects to every track, and offer creative prompts for writer's block. The post notes that Suno Studio is one of a few DAWs with built-in assistants, alongside FL Studio, Fender Studio Pro and Logic Pro, and is available with every Premier subscription.

Oct 29Thu
  1. Wiz Research60

    The Rise of Agentic Attacks: What They Look Like, and Where They're Headed

    Wiz reports a growing trend of intrusions in which AI agents, run through harnesses that write custom tooling, inspect results, and iterate at machine speed, appear to handle much of the hands-on work and potentially orchestrate the attack. The post reviews public cases involving frameworks such as ARTEX and Hermes plus custom agentic workflows, and distills seven fingerprints of agentic activity, including machine-paced sustained iteration, natural-language reasoning embedded in tooling, self-correction, breadth-first TTP coverage, verbose parallel lateral movement, probe-before-payload discipline, and self-labeled disposable artifacts. It argues that attribution will get harder as tooling becomes easier to share and adapt, that attacks can scale beyond human operator limits, and that agentic attacks tend to be noisy, creating detection opportunities for defenders with the right visibility.

    Why it matters: It offers a detailed, case-grounded framework for recognizing agentic intrusions and reflects on how the trend affects attribution, attack scale, and defensive readiness.

  2. Mistral AI82

    Le Chonk: Built for Coders, Trusted by Defenders.

    Mistral released the open weights of Mistral Large 4 (Le Chonk), a 1-trillion-parameter natively multimodal model trained from scratch in Europe, reporting leading open-weight results on coding benchmarks such as FrontierCode 2.0 (19.7%) and Terminal-Bench 4.0 (41%), plus 51% on the Artificial Analysis Cyber Index. It offers three reasoning modes (Low, Medium, High) and is priced at $0.70 input / $0.07 cached input / $2.10 output per million tokens under a modified MIT license that requires a separate commercial license for Model-as-a-Service offerings.

    Why it matters: The post details benchmark scores, pricing, licensing terms and deployment options for a flagship open-weight model, useful for developers weighing coding performance against cost and control.

  3. ElevenLabs Blog25

    How we use ElevenAgents to save 1,200 hours a month in daily operations

    ElevenLabs says its internal agents save teams over 1,200 hours a month, answering more than 5,000 questions monthly and taking actions such as building customer demos and opening over 150 tickets in the last month. The agents span sales (GTM and FDE bots), support and engineering triage, and product data analysis, with lessons including schema curation over prompt engineering, injecting Salesforce picklist values into tool type definitions, pruning overlapping procedures, and verifying tickets after creation.

  4. Amp Chronicle58

    Visualize This

    Amp can now visualize or build things directly in a thread, including charts, timelines, diagrams, and clickable mockups. The post shares examples from real threads, such as Quinn's request for ways to fit a diff stat into a small button across desktop, mobile, and tablet platforms.

  5. n8n Blog22

    Why we're moving n8n's automation team beyond the service desk model

    Romina da Costa, n8n's head of internal AI and automation, argues that a centralized automation team that only builds workflows shifts the bottleneck downstream rather than fixing it, and that with over 12,000 internal workflows no team can build and maintain them all. She describes reframing the team's goal from throughput to building an automation culture, guided by four principles: build the builders, require a baseline before building, choose the tool for the problem, and make what's built resilient. She notes the team still operates like a service desk more often than she'd like, and that shifting culture requires leadership buy-in and cross-functional change.

  6. Amp Chronicle58

    macOS Satellites

    Amp now offers macOS Satellites: a cloud Mac with Xcode and simulators, booted in about 30 seconds, attached to the thread's Linux orb so Amp can build, run, and test iOS or Mac apps while code stays synced on the orb. Macs start fresh with nothing saved, stay running for 20 minutes after Amp finishes so follow-ups can reuse the same Mac and build cache, and are then deleted. Pricing is $2/hour for mac.medium (6 CPUs, 14 GB) and $4/hour for mac.large (12 CPUs, 28 GB), drawn from included orb usage and then Amp credits; the feature is in beta with a waitlist in Settings → Experimental Features.

  7. Sakana AI Blog38

    Sakana API Platform、エンタープライズ向けにアーリーアクセスプログラムを開始

    Sakana AI announced the Sakana API Platform for enterprise customers and opened its Early Access program, offering its LLMs via an OpenAI-compatible API with management features for handling usage costs and members by organization and workspace. Sakana Fugu and Sakana Fugu Ultra are available first, with Sakana Namazu and Sakana Fugu Max to follow, and features include role-based access control, audit logs with CSV download, a commitment not to use inputs for training, budget limits, invoice payment, and MSA and DPA support.

  8. OpenRouter Announcements65

    How to Prevent Prompt Injection: 13 Attacks, Tested

    The article describes testing 13 self-written and 12 model-written prompt injection attacks against four current models in a document-summarizing assistant with an email tool, finding that text-only attacks rarely worked but that one model (GLM 5.3 Flash) sent 58 of 60 attack emails without defenses. It recommends labeling untrusted content as data (cutting attack emails from 60 to 13 of 720), using a guardrail with a model allowlist and budget limit, checking every tool call with Armor (which blocked all 73 attempted attack emails at about 0.3 seconds added latency), and keeping secrets out of the prompt. Armor blocked real work in one case: a user approving a document's request without naming the address.

    Why it matters: It offers a concrete, measured comparison of layered prompt-injection defenses, including where a tool-call guardrail blocks legitimate user requests.

  9. OpenRouter Announcements62

    How to Extract Tables from a PDF with an LLM

    The guide shows how to send a PDF to a file-reading model together with a strict JSON schema describing one row, so the model returns typed rows with cleanup such as parenthesized negatives and footnote markers handled by field descriptions. In tests over five PDFs (two scans, 644 cells), every model that read the PDF directly got all cells right, while failures came from the parser route and output token limits, including two silent failures. It recommends rejecting empty results and any response whose finish_reason is "length", and comparing routes on your own PDFs before choosing one.

    Why it matters: Offers a concrete, tested recipe with code for schema-driven PDF table extraction, plus specific guidance on silent failure modes worth knowing before building a pipeline.

  10. ElevenLabs Blog23

    Conversational AI in banking: Use cases, benefits, examples

    This guide explains how banks use conversational AI—NLP-based chat and voice agents that understand natural language, pull account context, and escalate to humans when needed—for 24/7 support, onboarding, disputes, and fraud alerts. It cites CFPB findings on early bank chatbots, McKinsey figures on reduced handling time and call volume, and ElevenLabs customer examples from Customers Bank and Klarna, while detailing security standards such as SOC 2 Type II, ISO 27001, PCI DSS, and Zero Retention Mode.

  12. Sierra Blog42

    Introducing Compliance Console

    Sierra announced Compliance Console, built into its Horizon platform, which gives companies configurable guardrails so their agents follow outreach rules across channels, use cases, states, and countries. The controls cover disclosures, contact windows based on recipient local time, holiday exclusions, frequency limits, and opt-outs, with Insights dashboards showing how many outbound calls and messages were allowed, deferred, blocked, or held for review.

  13. Replit Blog60

    ViBench 2.0: A Benchmark for Long-Horizon Software Building

    Replit releases ViBench 2.0, a long-horizon software development benchmark in which an agent builds a product from an empty project and then adds up to eighteen features on top of its own code, spanning seventeen products, 241 feature requests and 544 test plans. In Fireworks' official runs, GPT-6.1 Sol leads with a pass@1 of 65%, followed by GPT-6 Astra at 60% and Opus 5.5 at 49%, and no open-weight model passes a full build. Eight of the products are public on GitHub in the Harbor format, and results can be rerun from release v2.0.6 on the OpenHands-based harness.

    Why it matters: It offers concrete pass rates, costs and failure modes for frontier models on multi-feature product building, with a public, rerunnable harness.

  14. AWS Machine Learning Blog23

    Accelerating software delivery with agentic QA automation using Amazon Nova Act – Part 3

    This third post in the series describes two QA Studio capabilities for scaling agentic QA across an organization: applications, which group use cases and test suites under an owner and report health via pass rate, a flaky-adjusted 0–100 score, trend charts, and failure triage built on Amazon CloudWatch custom metrics; and team-based user management, which assigns five cumulative roles (Admin, Team Lead, Editor, Executor, Viewer) through Amazon Cognito groups, with cross-team sharing capped by the recipient's role. The backend enforces access through a Lambda authorizer resolving Cognito groups into scopes plus a resource check on application ownership or sharing, and the reference solution is deployed with the AWS CDK.

  15. AWS Machine Learning Blog42

    Building an Agentic AI Platform on Amazon Bedrock AgentCore

    This AWS post describes an enterprise Agentic AI Platform blueprint on Amazon Bedrock AgentCore that separates a centrally governed Platform control plane from repeatable, team-owned Workstream cells, with separate inference and tool gateways and a federated operating model. It details multi-account topology, service control policies, a CI/CD pipeline with deployed-runtime evaluation gates and adversarial denial tests, and dependency-ordered retirement, using the eu-west-1 (Ireland) Region. The companion GitHub blueprint lets teams deploy a representative Workstream cell and validate its security and lifecycle contracts end to end.

  16. Google Developers Blog58

    Hill-Climbing Agent Quality with Agents CLI Evaluation

    A walkthrough of the Agents CLI eval loop using the LangChain template in Agents CLI v1.8.0, covering a baseline run (mean 4.0), a system-prompt regression that dropped the mean to 2.75, per-case diffs via eval compare, and a gemini-3.8-flash to gemini-2.5-flash swap whose mean drop traced to a single off-topic case. It also shows scheduling eval runs with cron or Cloud Scheduler, a script that exits non-zero when scores fall past a threshold, and eval optimize for automated prompt hill-climbing on ADK projects.

  17. Pydantic AI Articles74

    Code mode: make your agents 2x faster and 2x cheaper on long tasks

    A benchmark of 5,276 graded runs across eight models on five small Pydantic AI apps found that code mode, where the model writes a program that calls tools inside a sandbox, cut cost by 58% and wall time by 51% on long tasks while raising success by 16 points. Gains grew with task length: at 200 items code mode was 47% to 75% cheaper and 42% to 67% faster, whereas at 5 items it cost 21% to 38% more on three of five tasks. The author attributes the win not to fewer round trips, since models already batch calls, but to less content per request, and runs the model's code in Monty, a Rust Python interpreter with no host access by default and 30-second, 256 MiB limits.

    Why it matters: Offers a detailed, methodology-heavy benchmark with per-model cost, latency and correctness numbers, plus an honest list of where code mode costs more.

  18. Claude Blog34

    Smart reports, surveys, and pooled group budgets for Claude Enterprise

    Anthropic announced that smart reports, in-app surveys, and pooled group budgets are now generally available for Claude Enterprise. Smart reports analyze a sample of a team's transcripts to group work into workstreams with spend, deliverables, friction points, and recommendations, while in-app surveys collect team feedback and pooled group budgets give a group one shared monthly budget with member limits as a guardrail. Availability varies by configuration, including HIPAA configurations and customer-managed encryption keys.

  19. Databricks Blog36

    Voice Agents are here. Context turns them into real customer value.

    Databricks argues that voice agents are now practical because synthetic voices sound natural and can respond within roughly 200 milliseconds, but that most enterprise AI investment has focused inward on employee assistants rather than customer conversations. It describes a prototype in which LiveKit carries calls, Deepgram handles speech-to-text, ElevenLabs adds voices, and Databricks supplies the "brain" that looks up caller context, applies business rules, and records each decision, built on context, ontology, and governance. The post outlines three pillars—grounding answers in business data, fast and slow AI decision-making including the ai_decide capability, and a learning loop from recorded calls—and points to the open-source DIVA blueprint on GitHub and a suggested 90-day pilot.

  20. Google DeepMind82

    Gemini 4 Carbon: Our next frontier model

    Google announced Gemini 4 Carbon, its strongest model with state-of-the-art cybersecurity and long-horizon capabilities, ranking #1 on the Artificial Analysis Intelligence Index v5.0 at 59 and scoring 88.2% on SEC-bench Pro, 54.8% on ExploitGym, and 71.0% on CWE-bench. It launches at Argon's introductory price of $2 per million input tokens and $10 per million output tokens, initially available to a subset of cyber defenders and trusted partners through the Fairwind Program before wider release. Google reports reduced misalignment alerts, a 0.0% prompt-injection attack success rate on Gray Swan's IPI Benchmark, and 0.0% successful cheating on Terminal-Bench 4.0.

    Why it matters: It documents a major frontier model release with concrete benchmark numbers and a staged, safety-focused rollout, useful for tracking competitive and cybersecurity trends.

  21. Mastra Blog36

    Introducing Span Queries

    Mastra now offers span queries that filter and group individual spans across traces, letting you return only matching spans rather than whole traces. Span queries are available in Mastra platform and Studio, the Mastra CLI ([email protected] or later), the Mastra Client SDK, and HTTP observability endpoints, and work on self-hosted projects where storage.observability uses DuckDB, PostgreSQL, or ClickHouse. Each query can combine up to 100 conditions, returns up to 1,000 spans per page, and span groups can group by up to 2 dimensions over a time range of up to 365 days.

  22. Replit Blog42

    Replit’s Architecture Behind Secure Agent Work

    Replit outlines four principles for securing AI agents in its Agent Habitat: least-privilege access with a credential broker that keeps secrets outside the agent's context, isolation via microVMs and separate development/production databases, engineering review of agent-proposed production changes, and audit logging with encryption and BYOC options. It says its existing microVM architecture prevented exploitation of recently disclosed Linux kernel vulnerabilities and that it patched affected Node.js, Python, and PHP tooling and deployed firewall rules centrally. Replit also describes round-the-clock AI red teaming and plans for finer-grained per-task permissions and eBPF-based sandbox monitoring.

  23. OpenRouter Announcements58

    Hit Your Claude Code Usage Limit? How to Keep Coding

    Claude plans have a five-hour limit and a weekly limit, and when you hit one you can keep coding by running Claude Code on an OpenRouter API key, which has no such windows and bills per token instead. The setup uses three environment variables (ANTHROPIC_BASE_URL, ANTHROPIC_AUTH_TOKEN, and an empty ANTHROPIC_API_KEY) placed in a shell function so plain `claude` still uses your plan, and you switch back by typing `claude` when the plan resets. The article reports median session costs of $0.305 on Opus 5.5, $0.155 on Sonnet 5.5, and $0.007 on Haiku 5.5, and warns that Claude Code's printed cost is unreliable because it prices OpenRouter models with Anthropic's table.

  24. Artificial Analysis Articles72

    GitHub Copilot Fusion orchestrates Anthropic and OpenAI models to balance performance and cost

    GitHub has released Copilot Fusion in GitHub Copilot, a runtime orchestration system that selects and coordinates models from Anthropic and OpenAI, including Claude Opus 5.5, Claude Sonnet 5.5, GPT-6 Astra and GPT-6 Sol, to optimize performance, cost and latency per task. In Artificial Analysis' Coding Agent Index v2.0, its max setting in GitHub Copilot CLI scored 2 points below Opus 5.5 in Claude Code at half the cost per task ($5.59 versus $11.42), leading Terminal-Bench 4.0 at 71 while ranking third on FrontierCode at 51 and scoring 57 on SWE-Atlas-QnA-AA. It used 7.2M tokens per task, roughly half of Claude Opus 5.5 in Claude Code and 30% less than Devin Fusion with Opus 5.5 (max) + GPT-6 Luna (high).

    Why it matters: Offers concrete benchmark and cost comparisons for a multi-model orchestration approach, useful for readers weighing coding-agent performance against token spend.

Oct 28Wed
  1. NVIDIA Newsroom58

    NVIDIA RTX Spark Windows Laptops Gain New Models, Agent Tools and Games

    NVIDIA announced RTX Spark laptops from ASUS, Dell, HP, Lenovo, Microsoft and MSI, combining a Blackwell RTX GPU and Grace CPU with up to 128GB unified memory and up to 1 petaflop of FP4 AI performance, starting at $2,599. Updates include the open-weight Nemotron 3.5 Transcribe speech model, Liquid AI's LFM3 family with day-zero support, and resource management controls in NVIDIA App for sharing GPU resources between foreground apps and background agents. Phantom Blade Zero launches Oct. 28 with ray tracing and DLSS 4.5, and RTX Remix 1.6 brings path tracing and DLSS to Minecraft Java Edition 1.21.1.

  2. Claude Blog56

    Introducing artifacts for legal work

    Anthropic introduced ready-to-use artifacts for legal professionals, including a gallery with tools for cite checking, jurisdiction surveys, matter boards, playbook deviation tracking, regulatory horizon scanning, and tabular contract review, each opening with an example and linking outputs to sources. The company said it has added more than 60 legal connectors since launching Claude for Legal in May, and it partnered with legal tech companies and law firms such as Goodwin, Harvey, Foley & Lardner, and PwC on artifacts. The gallery also includes nine access-to-justice artifacts, five built by supported nonprofit and public interest organizations, and Anthropic is launching a Legal Ambassadors Program with a waitlist.

  3. ElevenLabs Blog24

    Lettuce cuts creative ad production from over a week to under a minute with ElevenLabs

    Lettuce's growth team uses Ads Engine in ElevenCreative to pull its best-performing Meta ads, generate new versions in about 45 seconds, and push them back to market, with the creative director and a performance marketer reviewing each ad before it runs. The company reports roughly 75% lower creative costs and 40% to 60% lower cost per lead compared with other creative running in similar periods, and its current top-performing Meta ad came from the 15th to 20th iteration of an original asset.

  4. Anthropic Research71

    AI agents shape experts' judgment calls

    In a randomized study, 50 professional analysts using Claude Code on a farming-cooperative analysis task matched their agent's randomly assigned lean 86% of the time across four analytical choices, and 77% later repeated the agent's arguments, often verbatim, while only 15% credited the agent. Analysts who followed the agent's lean were about as confident in the recommendation and as likely to call it their own, and the four leans alone predicted 82% of final recommendations. The authors caution the effect was deliberately strengthened and studied in one task with one model, but note that if experts defer without realizing it, agents' arbitrary or misaligned preferences could shape policy and research decisions.

  5. Sarvam AI Blog38

    Sarvam and HQ IDS establish joint framework for AI adoption across the Army, Navy and Air Force

    Sarvam signed a Memorandum of Understanding with Headquarters Integrated Defence Staff (HQ IDS) under the Ministry of Defence, creating an umbrella framework for AI projects across the Indian Army, Navy and Air Force and HQ IDS joint organisations. Under the MoU, Sarvam will design, develop and deploy AI applications for operational, training and decision-support needs and provide follow-on technical support, with work carried out by its firewalled division Chanakya using its AI stack including Sarvam 105B. Focus areas include decision support, target acquisition and data fusion across the sensor-to-shooter loop, with human oversight retained on operational decisions and deployment on sovereign infrastructure.

  6. Browser Use Blog58

    Odysseys2: a long-horizon web agent benchmark

    Carnegie Mellon University and Browser Use released Odysseys2, a benchmark of 100 long-horizon tasks on the live web, each graded on 8 to 12 pass/fail checks. The best agent, GPT-6 Astra, fully completes 47 of 100 tasks with a mean rubric score of 84.6, while no other agent reaches 20 and five complete none. The authors note the LLM judge has not been validated against a human reference, reward-hacking flags are model judgments, and the live web changes.

  7. Artificial Analysis Articles60

    Announcing Artificial Analysis Intelligence Index v5

    Artificial Analysis released Intelligence Index v5, its biggest upgrade since v4, adding FrontierCode 2.0 and Terminal-Bench-Science and upgrading HLE-Diamond, AA-Omniscience v1.1 and CritPt v2, with 60% of the Index now based on private test sets or held-out answers. Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55), while the open weights gap widens from 3 points to 12. It also launched the Progress Index v1, which tracks model capability over time on one continuous scale.

    Why it matters: It details a major benchmark revision and new leaderboard results, useful for readers tracking how frontier and open-weight models are measured.