搜索“Agent”

Amp has a new agent and it specializes in code review. In the VS Code extension, you can use this agent by going to the review panel. Start by dragging the selection of changes you want to review. Sometimes, you want to review a single commit; other times you want to review all outstanding changes on your branch: Amp will pre-scan the diff and recommend an order in which to review the files. It will also provide a summary of the changes in each file and the changeset overall: This addresses a ke

匹配字段正文

The future of AI agents is one where models work seamlessly across hundreds or thousands of tools. An IDE assistant that integrates git operations, file manipulation, package managers, testing frameworks, and deployment pipelines. An operations coordinator that connects Slack, GitHub, Google Drive, Jira, company databases, and dozens of MCP servers simultaneously. To build effective agents , they need to work with unlimited tool libraries without stuffing every definition into context upfront. O

匹配字段正文

rovement in overall performance compared to version 2507. Behind every improvement in experience, there is a new technological breakthrough. Qwen DeepResearch 2511 employs a multi-agent collaborative mechanism , with joint optimization in areas such as user need analysis, overall strategy planning, search verification strategies, parallel tool calls, web page reading analysis, and research report writing. This is combined with an overall Memory management mechanism and research state tracking to

匹配字段正文

ost, you pay it every time you switch context and codebases. The faster and better you understand your codebase, the faster and better you’ll be able to fix it yourself, or prompt agents to do it. Until today, the standard approach by Copilot, Claude Code, Codex, and even Windsurf Cascade, was to have you ask questions of a generalist agent with access to your code in a typical chat experience. But those solutions don’t solve focused onboarding and strongly grounded navigation to onboard, debug,

匹配字段正文

October 28, 2025 MiniMax M2 is now available on Ollama’s cloud. It’s a model built for coding and agentic workflows. Highlights Superior Intelligence. According to benchmarks from Artificial Analysis, MiniMax-M2 demonstrates highly competitive general intelligence across mathematics, science, instruction following, coding, and agentic tool use. Its composite score ranks 1 among open-source models globally. Advanced Coding. Engineered for end-to-end developer workflows, MiniMax-M2 excels at multi

匹配字段正文

We heard you! Amp Free no longer requires you to opt into training. Now more programmers can use Amp Free as a free coding agent for both sensitive work code and on personal side projects. How is this possible? It turns out that the Internet's best business model (advertising) works, and we're able to make Amp Free without the training requirement. Try it: install or update Amp, then use /mode free in the Amp CLI, or toggle smart/free mode in the editor extension. Thank you to our fantastic laun

匹配字段正文

Agentic AI systems, capable of reasoning, planning, and taking autonomous actions, are powering the next leap in developer applications. To build these systems, developers need tools that are open, efficient, and ready to scale. And, as demand for agents grows, open, performant models are the key as they provide transparency, adaptability, and cost-control. NVIDIA Nemotron is a family of open models, datasets, and technologies that unlock developers to build highly efficient and accurate models

匹配字段正文

sensitive information or downloading malware. It is worth noting that effective sandboxing requires both filesystem and network isolation. Without network isolation, a compromised agent could exfiltrate sensitive files like SSH keys; without filesystem isolation, a compromised agent could easily escape the sandbox and gain network access. It’s by using both techniques that we can provide a safer and faster agentic experience for Claude Code users. Two new sandboxing features in Claude Code Sandb

匹配字段正文

ng Huang for their invaluable guidance throughout this work. Special thanks as well to Nicolo Lucchesi, Alexander Matveev, Akshat Tripathi, and Saheli Bhattacharjee, for being an integral part of the V1 integration and the push for Cloud Next. Related Posts Announcing vllm-metal: Concurrent Serving on Apple Silicon vllm-metal brings vLLM's paged, continuously batched serving stack to Apple Silicon, with flatter TTFT under concurrent agent load, batched MTP, and automatic M5 prefill acceleration.

匹配字段正文

ropdown → and choose the model (e.g. glm-4.6) Usage with Zed First pull the coding models so they can be accessed via Zed: Then, open Zed (now available for Windows!) Click on the agent panel button (glittering stars); Click on the model dropdown → Configure; Select LLM providers → Ollama; Confirm the Host URL is http://localhost:11434, then click Connect; Select a model under Ollama Usage with Droid First, install Droid: Add the following configuration to /.factory/config.json: Then run Droid a

匹配字段正文

Engineering Matching state-of-the-art accuracy is table stakes. We built BU 1.0 to be the fastest browser agent while maintaining the same performance. Co-founder, CTO BU 1.0 achieves 65.7% accuracy on OnlineMind2Web — matching Gemini 2.5 Computer Use and outperforming all the other models. We recently released our LLM gateway that makes this possible. But accuracy is table stakes. People want their agents to work at least as fast as humans. Evaluation Challenges Before diving into performance,

匹配字段正文

Product We built a special LLM gateway that reduces the latency by 6x while keeping the same performance. The agents can now take 20 steps per minute. How many can a human do? Co-founder, CTO Browser Use 0.8.0 introduces ChatBrowserUse gateway which reduces the latency like crazy (6x faster than computer use models) while keeping the same performance. Our agents can now take 20 steps per minute. How many can a human do? BU 1.0 achieves the same accuracy ( 65%) as other state-of-the-art computer

匹配字段正文

ure is built to bypass essentially anything. Whether you're dealing with anti-bot protections, CAPTCHAs, or restrictive environments, Browser Use Cloud handles it seamlessly. Your agents browse like real users—undetectable, unrestricted, and unstoppable. One API Key. That's It. Getting started with cloud browsers used to mean juggling multiple services, configuring CDPs, and managing complex infrastructure. Not anymore. With Browser Use Cloud, you simply need a BROWSER USE API KEY and you're rea

匹配字段正文

e to make contributions, provide feedback, or offer feature requests, please join our community Slack. What is Docent? Docent is our tool for understanding complex behaviors in AI agent transcripts. It enables users to explore large collections of transcripts for interesting or unwanted behaviors, and precisely measure how often particular behaviors occur. For instance, a user might start with a vague question like “is my model reward hacking.” Docent first helps the user explore potential insta

匹配字段正文

and run Ollama’s Python library Then make a request using ollama.web search: Example output JavaScript Install and run Ollama’s JavaScript library Example output Building a search agent Use Ollama’s web search as a tool to build a mini search agent. The example uses Alibaba’s Qwen 3 model with 4B parameters. Result Recommended models: These models have great tool-use capabilities and are able to have multi-turn interactions with the user and tools to get to a final result. qwen3; gpt-oss Recomme

匹配字段正文

mini to summarize and clarify any page, with Workspace support coming soon Zoom launches AI Companion 3.0 with cross-platform meeting intelligence Notion 3.0 introduces autonomous Agents that work inside your docs and databases Workday ships new HR and Finance agents focused on real business tasks Zoom’s AI Companion 3.0 pushes agentic AI across meetings, chat, and support workflows Gemini arrives in Chrome for U.S. desktop users Summary: Google is rolling out Gemini in Chrome on Mac and Windows

匹配字段正文

At OpenHands, our number one mission is to build a highly autonomous and accessible coding agent that can be used for any software development task, any time. One of our flagship ways to provide this is through the OpenHands Cloud where users can interact with agents through a GUI, API, or integrations into Slack, GitHub, GitLab, and BitBucket. We are constantly listening to feedback from the cloud users, and this month worked hard to fix two major requests that we got from users: Cost of the Ag

匹配字段正文

latency, and sustainability. One-Sentence Summary GPT-5: enterprise routing for smarter inference; vLLM Semantic Router: technical-first routing for open-source LLMs; Edge future: context-aware, minimal-compute inference that works seamlessly Related Posts Announcing vllm-metal: Concurrent Serving on Apple Silicon vllm-metal brings vLLM's paged, continuously batched serving stack to Apple Silicon, with flatter TTFT under concurrent agent load, batched MTP, and automatic M5 prefill acceleration.

匹配字段正文

across more TerraTorch models and modalities and beyond, making installation seamless. Longer-term, we envision IO Processors powered vision-language systems, structured reasoning agents, and multimodal pipelines, all served from the same vLLM stack. We're also excited to see how the community uses IO Processors to push the boundaries of what’s possible with vLLM. We also plan to continue working with and contributing to the vLLM community to enable more multimodal models and end-to-end use case

匹配字段正文

RL environments are the playgrounds where agents learn. Until now, they’ve been fragmented, closed, and hard to share. We are launching the Environments Hub to change that: an open, community-powered platform that gives environments a true home. Environments define the world, rules and feedback loop of state, action and reward. From games to coding tasks to dialogue, they’re the contexts where AI learns, without them, RL is just an algorithm with nothing to act on. Environments sit at the center

匹配字段正文