English text is being prepared.

Stealth Bench V2, released by Browser Use, shows that headless Chrome without a proxy passes only 3% of visits, headed Chrome with a US proxy reaches 14%, and Browser Use Cloud reaches 90.5%. Cloudflare identifies automation through navigator.webdriver, TLS and request header fingerprints, IP reputation, and cookie and behavioral signals; stealth plugins alone cannot change TLS and IP. Browser Use Cloud is priced at $0.02 per browser hour plus $5/GB for proxies, and its official documentation st

Matched fieldsBody

Replit announced that Cloud Agents are now available to all Pro and Enterprise customers. Non-engineers can describe changes in natural language, and the agent locates the code, runs the application in a cloud sandbox to verify it, and submits a PR or MR with screenshots and video to GitHub or GitLab. Teams can configure a shared environment once, and review feedback can be turned into skill updates that require engineer approval; the feature is in early access. The Slack integration is now avai

Matched fieldsBody

a16z partners Ben Horowitz and Michael McGuiness wrote the article, using Databricks' four generations of technological evolution—from Spark and Lakehouse to Lakebase and agents—as a thread to discuss how agents are changing enterprises. The article states that Databricks has surpassed $7 billion in annualized revenue, up 80% year over year, at a valuation of $190 billion, with more than 17 million databases launched by agents every day. The authors argue that enterprises need an agent-facing sy

Matched fieldsBody

Google A2UI adds swappable inference formats, among which A2UI Express, as the new default format, can reduce latency by up to 3x and cut the tokens needed to generate comparable UIs by up to 3x; when combined with decision models such as Jev for layout decisions, latency reductions can reach up to 80x. Express is now available in the Python agent SDK (a2ui-agent-sdk 0.9.0), the TypeScript SDK (@a2ui/agent), and the Dart SDK (a2ui agent 0.0.1-wip004); developers can switch formats with a single

Matched fieldsBody

Anthropic has released Claude for Financial Advisors, connecting data from custodians, asset management and wealthtech platforms into Claude, and providing skills tailored to financial advisors' daily work. The suite includes connectors such as Addepar, BlackRock, Charles Schwab, Envestnet, FactSet, Orion and Wealthbox, as well as skills for meeting preparation, portfolio rebalancing review, post-meeting notes and follow-up, and compliance and AI policy checks; regulated steps such as investment

Matched fieldsBody

Based on a recent webinar with Google Cloud, Anthropic presents a three-lever framework—control, track, prove—for managing Claude Code costs on Google Cloud and measuring return on investment. The Claude apps gateway supports setting daily/weekly/monthly spending limits by user, IdP group, or organization, alerts within the CLI when usage reaches 75% and 95%, and returns 429 once the cap is hit. On the Google Cloud side, native capabilities include per-project billing, budget alerts, Spend caps

Matched fieldsBody

Three studies by ElevenLabs of shoppers in the US, UK and other countries show that when shoppers were asked to actually use the ElevenLabs voice Agent first before answering, nearly half chose "speak and listen to the reply," about 30% chose "speak and read text on screen," and voice ranked first across all six shopping task categories. Among respondents who had not tried it, preference for typing led voice by more than 4:1. The more serious the problem, the less willing shoppers were to hand i

Matched fieldsBody

ElevenLabs has released a practical guide on how to build an AI voice assistant by chaining speech-to-text, an LLM, and text-to-speech into a real-time pipeline. The guide recommends keeping round-trip responses within 500 ms and pairing this with VAD, while also connecting to internal knowledge bases and business tools, and putting safety measures such as verbal confirmation and handoff to a human agent in place. Developers can build directly with ElevenAPI, or configure and launch on the Eleve

Matched fieldsBody

Anthropic has published the 335-page prospectus required for a U.S. stock market listing, with the market expecting its listing valuation could reach $2 trillion or more. The document shows the company lost $80.7 billion in the first half of this year, twice its full-year loss for 2025, while revenue over the same period was $16.2 billion; its business is highly dependent on a small number of major customers, with just five enterprise clients contributing a quarter of its business. The document

Matched fieldsBody

Cognition launches the FrontierCode 2.0 coding benchmark to assess whether the code produced by models can actually be merged; even the highest-scoring Opus 5.5 scores only 45%, showing the benchmark is far from saturated. Compared with version 1.1, reference solutions modify on average 67% more files and 89% more lines of code, the codebase size is 25% larger, task-covered repositories increase by 55%, and blocking scoring criteria increase by 43%. DeepSeek V4.1 Flash drops from 43.9% to 19.3%,

Matched fieldsBody

OpenAI Codex 0.165.0 adds a built-in Google Cloud Vertex AI provider (configured via model provider = "gcp") and adds GPT-6.1 Terra to the model catalog. This version also enables memory v2 by default and supports importing external agent memory, while allowing reasoning effort to be adjusted during conversations by default. The session selector adds repository filtering, and the agent command center supports grouping by directory.

Matched fieldsBody

NVIDIA has released the Nemotron-Personas France and Belgium datasets and provided a complete post-training workflow for fine-tuning Nemotron 3.5 Lightning with them. The workflow includes defining behavior contracts, generating synthetic conversations with personas, fine-tuning with LoRA on NVIDIA NeMo, merging adapters, and then comparing and evaluating item by item against the base model on a frozen test set. The France dataset was built in collaboration with Pleias, and the Belgium dataset i

Matched fieldsBody

Liquid AI has released the LFM3 series of open-weight models, all branched from LFM3-1.7B-Base, covering text, vision, audio, Omni, and the d1 decision model. LFM3-1.7B-Base adopts a three-hybrid-layer architecture, is pretrained on 22T tokens, and in the mid-training stage over the final 3.5T tokens incorporates more agent trajectories and tool-calling data, with the context window extended to 256K tokens. The company says LFM3-1.7B is about 25% faster than LFM2-1.2B at an 8K context, and score

Matched fieldsBody

In a Latent Space podcast interview, Anthropic engineering lead Katelyn Lesse and product lead Angela Jiang discussed the Opus 5.5 release, covering adjustments to the model's writing style, pricing based on intelligence per dollar, and API changes such as requiring thinking blocks to be retained across turns to prevent distillation. The interview also described Claude Managed Agents' approach of separating sessions and the harness from the sandbox that executes code, with credentials mounted se

Matched fieldsBody

Google DeepMind has released the world model Genie 4, which can generate interactive worlds in real time and, for the first time, natively generate ambient sound and character dialogue. The model builds on capabilities from the Gemini series; Gemini 3.8 Flash breaks down goals in a loop and checks frames to maintain causal coherence across multi-step events, and Gemini 4 Argon has also been used for joint simulation tasks with Gemini Robotics. The company says it remains an early experimental mo

Matched fieldsBody

Matched fieldsBody

AWS has launched SOCI index images for Amazon SageMaker AI training and processing jobs, enabling containers to start within seconds through lazy loading, with no changes to job definitions required. Official benchmarks show that image download times for 1.8–35.4 GB images dropped from 48 seconds to 7 minutes 17 seconds down to 3–11 seconds, a speedup of 8x–46x, and GPU idle time reduced by up to 90%. If the index is unreadable or lazy loading fails, jobs automatically fall back to standard down

Matched fieldsBody

Matched fieldsBody

The Meta engineering team has introduced PrivacyLib, a general-purpose privacy library that runs within clients across hundreds of data and processing systems and performs billions of checks per second. A system only needs to integrate it once, and new privacy requirements can reuse that integration as modules, without the need to reimplement them in each system. The newly added Agent Observability module can observe data access events of internal AI agents, and systems that have already integra

Matched fieldsBody

Databricks shared its approach to improving document extraction accuracy using ai extract and MLflow Prompt Registry: in a project that extracted about 40 structured fields from dozens of PDF layouts, merely modifying the schema halved the errors in the revision field, but the total extraction errors across all fields actually increased by 9%. The article recommends describing field-level rules with the schema, using global instructions for document-level logic, and using precision mode when the

Matched fieldsBody