Top AI model harnesses for general use cases including coding
Anthropic's agentic coding tool for developers. Claude Code understands your codebase, edits files, runs commands, and helps you ship faster. Claude Code Hand Claude a bug fix, test, or multi-day migration. Steer and review from your terminal, IDE, Slack, or web. Claude Code is included in Claude Pro and Max plans. See plans Download for macOS Read documentation (opens in new tab) Available for macOS, Linux, and Windows. Other ways to use Claude Code: Terminal (opens in new tab) Web (opens in new tab) iOS (opens in new tab) Android (opens in new tab) GitHub (opens in new tab) VS Code (opens in new tab) JetBrains (opens in new tab) Slack (opens in new tab) acme-storefront/Fix the double-charge bug in checkout We're seeing duplicate charges when customers double-click the pay button. Can you find and fix it? I can reproduce it: the checkout form fires two POST /charges requests when the button is clicked twice before the first response lands. Tracing where the idempotency key should be. just now Read 3 files, searched the checkout flow Ran agent · Reproduce the double-submit against a test charge Root cause: createCharge() generates a new idempotency key per call instead of per checkout session, so the gateway treats the second click as a fresh charge. Fixing that and disabling the button while a charge is in flight. just now Edited charges.ts +9 -3 Auto Opus Extra high acme-dashboard/Add a dark mode toggle to settings Add a dark mode toggle to the settings page. Should remember the choice and follow the system default until they pick one. There's already a ThemeProvider wrapping the app but it's hard-wired to light. I'll lift the mode into context, persist it to localStorage, and add a Light / Dark segmented control under Settings → Appearance. just now Read 4 files, searched the codebase Edited ThemeProvider.tsx +18 -2 src/theme/ThemeProvider.tsx 10export function ThemeProvider({children}: {children: ReactNode}) { 11+ const prefersDark = useMediaQuery('(prefers-color-scheme: dark)') 12 const stored = localStorage.getItem('theme') 13+ const [mode, setMode] = useState(stored?? (prefersDark? 'dark' : 'light')) 14 15+ useEffect(() => { 16+ localStorage.setItem('theme', mode) 17+ }, [mode]) 18 19 return ( 20 <ThemeContext.Provider value={{mode, setMode}}> 21 {children} Edited settings.tsx +21 -0 Edited tokens.css +24 -0 Ran 2 commands Done — the Appearance section now has the toggle. Preview's on the right; flip it and the whole app should follow. Reduce-motion and density rows are stubbed so the page doesn't look bare. just now #112acme-dashboardclaude/settings-dark-mode-Xk4fq CI Auto Opus Extra high Get started with Claude Code Individual Team & Enterprise Pro Claude Code is included in your Pro plan. Perfect for short coding sprints in small codebases. $17 Per month with annual subscription discount ($200 billed up front). $20 if billed monthly. Try Claude (opens in new tab) Max 5x Claude Code is included in your Max plan. Great value for everyday use in larger codebases. $100 Per month Try Claude (opens in new tab) Max 20x Even more Claude Code included in your Max plan. Great value for power users with the most access to Claude models. $200 Per month Try Claude (opens in new tab) Usage limits apply. Prices shown don't include applicable tax. Price and plans are subject to change at Anthropic's discretion. Usage limits apply. Price and plans are subject to change at Anthropic's discretion. Latest feature announcements Projects: Group related coding sessions so you can run and easily supervise multiple Claude agents at once. Available on Claude Code Desktop. BlogSep 17, 2026 Auto mode by default: Claude Code now runs in auto mode by default on Pro, Max, and Team plans, so it can work longer while still catching risky commands. BlogAug 7, 2026 Self-hosted environments: Run Claude Code sessions on your own infrastructure, inside your network and next to your internal services. Now in public beta. BlogAug 6, 2026 Artifacts: Preview in-progress work as a live, interactive artifact built from your session context, and share it with your team. BlogJun 18, 2026 View changelog (opens in new tab) What Claude Code can take on Claude Code builds the plan, asks clarifying questions, and handles work that runs for hours or days. Get Claude Code Desktop VS Code JetBrains On the web Slack curl -fsSL https://claude.ai/install.sh | bash Or read the documentation Try Claude Code (opens in new tab) Developer docs (opens in new tab) Onboarding Triage issues Migrate code I'm new to this codebase. Can you explain it to me? ⏺︎ I'll analyze the Excalidraw codebase structure and provide a high-level overview. ⏺︎ Analyze(what: "Excalidraw codebase structure and provide a high-level overview of what it is, its main components, and how they're organized. Focus on understanding the project's purpose, architecture, and key features.") Excalidraw is an open-source virtual whiteboard application with a hand-drawn style. It's designed for creating diagrams, wireframes, and collaborative drawings. ## Purpose - Web-based drawing tool with infinite canvas - Hand-drawn aesthetic with support for shapes, arrows, text - End-to-end encrypted collaboration - Works offline as a PWA (Progressive Web App) ## Architecture The project is organized as a monorepo with these main components: 1. Core Package (packages/excalidraw) - React component library for embedding Excalidraw in any application - Handles drawing, rendering, and core functionality - Uses roughjs for the hand-drawn style - Includes export capabilities (PNG, SVG) 2. Web Application (excalidraw-app/) - Full-featured web app at excalidraw.com - Built on top of the core package - Adds collaboration features, local storage, sharing 3. Support Packages - common - Shared utilities and constants - element - Element manipulation and type definitions - math - Mathematical functions for drawing - utils - General utility functions 4. Examples - Integration examples (Next.js, browser script) - Shows how to embed Excalidraw in different environments 5. Documentation (dev-docs/) - API documentation - Integration guides - Component documentation ## Key Features - Real-time collaboration with end-to-end encryption - Export to various formats (PNG, SVG, JSON) - Customizable drawing tools (rectangle, circle, arrows, freehand) - Dark mode support - Localization - Library of shapes and components - Arrow bindings and labels - Image support ## Technology Stack - React for UI - TypeScript for type safety - Firebase for collaboration backend - Jotai for state management - Vite for development and building - IndexedDB/LocalStorage for offline storage The architecture allows for both using Excalidraw as a component in other applications and as a standalone web application with additional features. Code onboarding Claude Code maps and explains entire codebases in a few seconds. It uses agentic search to understand project structure and dependencies without you having to manually select context files. Turn issues into PRs Stop bouncing between tools. Claude Code works with GitHub, GitLab, and your command line tools to read issues, write code, run tests, and open pull requests. Run multi-hour refactors and migrations Claude Code follows imports across the repo, runs your tests, and keeps going when something breaks. Meets you where you code Start in your terminal Super powerful terminal integration. Works with all your CLI tools alongside any IDE. curl -fsSL https://claude.ai/install.sh | bash Or read the documentation Try Claude Code (opens in new tab) Developer docs (opens in new tab) Integrate with your editor Native extensions for VS Code (+ Cursor, Devin Desktop) and JetBrains IDEs. VS Code (opens in new tab) JetBrains (opens in new tab) Access anywhere Quick access from browser, mobile app, or Claude on desktop. Great for parallel work or on-the-go coding. Open in browser (opens in new tab) Download app Kick off coding tasks in Slack Add to Slack (opens in new tab) Learn more What developers are saying “Claude Code has dramatically accelerated our team's coding efficiency. I can now write EDA code in a notebook—pulling data, training a model, and evaluating it with basic metrics—and then ask Claude to convert that into a Metaflow pipeline. This process saves 1-2 days of routine (and often boring!) work per model.” Anton Biryukov, Staff Software Engineer Read story “With Claude, we're not just automating customer service—we're elevating it to truly human quality. This lets support teams think more strategically about customer experience and what makes interactions genuinely valuable.” Fergal Reid, VP of AI Read story “Claude Code is moving our team up a level: we decide what needs to happen, and smooth the process so it can build and verify end-to-end. A big part of my job now is to keep as many instances of Claude Code busy as possible.” Simon Last, Co-founder Read story Connects with your favorite command line tools Your terminal is where real work happens. Claude Code connects with the tools that power development—deployment, databases, monitoring, version control. Rather than adding another interface to juggle, it enhances your existing stack. FAQ How do I get started with Claude? You can access Claude Code with a Claude Pro or Max plan, a Team or Enterprise plan, or a Claude Console account. Download Claude Code and sign in with your respective Claude or Console credentials. What kind of tasks can Claude Code handle? Claude Code can handle routine work like bug fixes and testing, and larger jobs like refactors and new features that are long-running and asynchronous. You set the direction as the architect and orchestrator, and Claude Code does the work. Describe what you want and it plans, writes code, runs tests, and opens pull requests. It can work on several tasks at once, and surface decisions you need to make for it to keep going. How does Claude Code work with my existing tools? Claude Code runs in your terminal and works alongside your preferred IDE and development tools without requiring you to change your workflow. Claude Code can also use command line tools (like Git) and MCP servers (like GitHub) to extend its own capabilities using your tools. Is Claude Code secure? Yes. Claude Code runs locally in your terminal and talks directly to model APIs without requiring a backend server or remote code index. It also asks for permission before making changes to your files or running commands. What are the system requirements to run Claude Code? Claude Code works on macOS, Linux, and Windows. See full system requirements. How much does Claude Code cost? When used with a Claude Console account, Claude Code consumes API tokens at standard API pricing. Does Claude Code work with the Claude desktop app? Yes. Max, Pro, Team, and Enterprise users can access Claude Code on the Claude desktop app. What is fast mode on Claude Code? Fast mode is a high-speed configuration for Opus 5, making the model 2.5x faster at a higher cost per token. Fast mode is available: In research preview on Claude Code, and is priced at $10/$50 per million tokens. On consumption-based plans. Via usage credits for users on subscription plans. Get the technical rundown Claude Code documentation Developer docs Common workflows Developer docs Using CLAUDE.md files Blog Introduction to agentic coding Blog How Anthropic teams use Claude Code Blog Fix software bugs faster with Claude Blog Create what’s exciting. Maintain what’s essential. Use Claude Code where you work Get Claude Code Desktop VS Code JetBrains On the web Slack curl -fsSL https://claude.ai/install.sh | bash Or read the documentation Try Claude Code (opens in new tab) Developer docs (opens in new tab) Get the developer newsletter Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox. Please provide your email address if y
OpenCode - The open source coding agent. The open source AI coding agent Free models included or connect any model from any provider, including Claude, GPT, Gemini and more. curl npm bun brew paru yay curl -fsSL https://opencode.ai/v2/install | bash Your browser does not support the video tag. What is OpenCode? OpenCode is an open source agent that helps you write code in your terminal, IDE, or desktop. [*] LSP enabled Automatically loads the right LSPs for the LLM [*] Multi-session Start multiple agents in parallel on the same project [*] Share links Share a link to any session for reference or to debug [*] GitHub Copilot Log in with GitHub to use your Copilot account [*] ChatGPT Plus/Pro Log in with OpenAI to use your ChatGPT Plus or Pro account [*] Any model 75+ LLM providers through Models.dev, including local models [*] Any editor Available as a terminal interface, desktop app, and IDE extension Read docs The open source AI coding agent [*] With over 208,000 GitHub stars, 950 contributors, and over 13,000 commits, OpenCode is used and trusted by over 16M developers every month. Fig 1. 208K GitHub Stars Fig 2. 950 Contributors Fig 3. 16M Monthly Devs Built for privacy first [*] OpenCode does not store any of your code or context data, so that it can operate in privacy sensitive environments. Learn more about privacy. FAQ What is OpenCode? How do I use OpenCode? Do I need extra AI subscriptions to use OpenCode? Can I use my existing AI subscriptions with OpenCode? Can I only use OpenCode in the terminal? How much does OpenCode cost? What about data and privacy? Is OpenCode open source? Access reliable optimized models for coding agents Zen gives you access to a handpicked set of AI models that OpenCode has tested and benchmarked specifically for coding agents. No need to worry about inconsistent performance and quality across providers, use validated models that work. Learn about Zen Be the first to know when we release new products Join the waitlist for early access. ©2026 Anomaly Brand Privacy Terms English
OpenClaw is the open-source AI assistant that runs on your machine and works from WhatsApp, Telegram, or any chat app: inbox, email, calendar, flights. Quick Start All install options related content install Apps One-liner npm Hackable Desktop Apps Full apps that install everything for you — gateway, chat, setup, and node features. Available for macOS, Windows, and Linux. macOS macOS 15+ · Universal Binary · v2026.9.5 Download macOS Windows Windows 10/11 · x64 or ARM64 · v2026.9.4 Download x64 Download ARM64 Linux New x64 · AppImage or Debian package · v2026.9.5 Download AppImage Download.deb Works on macOS, Linux, and Windows. The apps and the one-liner install Node.js and everything else for you. Switch later with openclaw update --channel dev or openclaw update --channel stable. What People Say View all related content "The Clawfather 🦞. Thank you for joining us at Build today @steipete!" @pavandavuluri "The speed of agents in this new version is incredible. Kudos to @steipete and the whole community 🦞" @rodrigofarinha "Last night we celebrated @openclaw, the fastest growing project on GitHub. In just six months, @steipete, dozens of maintainers, and thousands of contributors built something remarkable: open source personal agents. 🚀 We had over 3,300 people registered for the event!" @ashleywolf "If it wasn't for OpenClaw we wouldn't have the massive progress across these kinds of personal agent harnesses. OpenClaw was the first agent harness I used, and now I get to enjoy other harnesses too. Thank you Legend, keep it up" @bartslodyczka "I'm finally testing @openclaw. This is crazy good, this thing is probably going to run my life. Congratulations @steipete! But I want my sleep back 🤣" @nelsonlopes_ "Thanks Peter, I've been super impressed with how much OpenClaw has improved over the last month. As someone that's been using it since the early days, it's kinda like a whole new product now, so much polish 🤌" @morganlinton "The latest OpenClaw release is fantastic! Fixes all the issues I've having the past week. Thank you!" @mronge "Thanks for joining us, and all the partnership as we make OpenClaw run super well on Windows! 🦞" @satyanadella "Peter Steinberger is the creator of OpenClaw, the open-source AI agent that went from a weekend project to the most-starred software repo on GitHub in under 5 months, with 346k+ stars." @ycombinator "Using Microsoft Execution Containers, you can run OpenClaw natively on Windows, and the sandbox prevented deletion of all files on the Desktop. Running OpenClaw more safely in an enterprise is "pretty dope". :)" @jennifermarsman "I'm addicted to building thanks to OpenClaw and Codex. Yesterday I built dm-annotate, a Mac screen annotation tool. Today, I'm building a curriculum recorder suite." @BradGroux "you can sign in to openclaw with your chatgpt account now and use your subscription there! happy lobstering." @sama "You can access 𝕏 API via @OpenClaw. We’re trying to make it affordable without giving away the shop. Hopefully, this can be useful & fun 💫" @elonmusk "A snake showed up in our backyard yesterday. Instead of panicking, I sent this photo to my OpenClaw. It identified it as a harmless gopher snake, gave me a local number to call to confirm, and suggested ways to catch it to make it a pet for the kids. I have small moments like this everyday now, usually less dramatic, that remind me how AI has made daily life feel a little more magical." @growing_daniel "It's happened. My OpenClaw and Hermes Agent are now chatting with one another. The cyberpunk future is here, but hopefully it's a whitepill instead of a blackpill. Let's try to push it to the former." @garrytan "I am SO SO excited about the Windows x OpenClaw announcement at #MSBuild. I love that our 🦞s can do all sorts of work for us (prepping for performance reviews, figuring out what happened in meetings that I refused to go to, godforsaken EXPENSE REPORTS, ha). And @steipete is ACTUALLY the coolest dude you’ve ever met." @donasarkar "I changed my mind recently. OpenClaw acquisition will pay off for OpenAI. It’s still by far the best personal agent in the world. Nothing has come close. It’s a net new product experience for normies and non SaaS businesses. Plus, Peter is top 0.1% talent acquisition. gg" @frankdegods "Here’s OpenAI’s latest realtime voice model, GPT-Realtime-2, wired up to OpenClaw. It’s amazing. Realtime continuous chat and fast OpenClaw agent action on real tasks. Truly excellent." @BenjaminBadejo "Shit wow I just gave my @openclaw Lobster a real phone number using @twilio and called him. Mind blown. Took 20 min to set up. I still can't believe it." @OmarShahine "It's a crime that this repo only has under 400 stars. /autoreview and /handoff are literally two of the best agent skills I've ever used. Massive thanks @steipete and @openclaw" @hungv47 "If you left Openclaw because of update stability its time to give it another try. The last 2 weeks have been the best experience I've had on openclaw so far. Glad i stuck with it and didn't migrate or give up because of complexity. I ended up learning things i wouldn't have known otherwise and now im on the most Feature rich Agentic System." @Strong_Genetics "Finally updated my openclaw and now using GPT-5.5 and the difference is visible and the humor is as like claude models! Amazing work on the harness @steipete" @MahmoudAshraf93 "Setup @openclaw by @steipete yesterday. All I have to say is, wow. First I was using my Claude Max sub and I used all of my limit quickly, so today I had my claw bot setup a proxy to route my CoPilot subscription as a API endpoint so now it runs on that. It's the fact that claw can just keep building upon itself just by talking to it in discord is crazy. The future is already here" @jonahships_ "Tried Claw by @steipete. I tried to build my own AI assistant bots before, and I am very impressed how many hard things Claw gets right. Persistent memory, persona onboarding, comms integration, heartbeats. A few minor wrinkles remain, but the end result is AWESOME." @AryehDubois "I've been saying for like six months that even if LLMs suddenly stopped improving, we could spend *years* discovering new transformative uses. @openclaw feels like that kind of "just had to glue all the parts together" leap forward. Incredible experience." @markjaquith "Why @openclaw is nuts: your context and skills live on YOUR computer, not a walled garden. It's open source. Growing community building skills. Only 19 days old and constantly improving. 'Personal AI assistant' undersells it — it's a company assistant, family assistant, team tool. Proactive AF: cron jobs, reminders, background tasks. Memory is amazing, context persists 24/7." @danpeguine "Yeah this was 1,000% worth it. Separate Claude subscription + Claw, managing Claude Code / Codex sessions I can kick off anywhere, autonomously running tests on my app and capturing errors through a sentry webhook then resolving them and opening PRs... The future is here." @nateliason "A smart model with eyes and hands at a desk with keyboard and mouse. You message it like a coworker and it does everything a person could do with that Mac mini. That's what you have now." @nathanclark_ "OpenClaw is a 24/7 assistant with access to its own computer. What if there were ten, or a hundred, or a thousand?? All running 24/7 in the cloud with access to your files, Gmail, calendar, everything about you... That's the future, and we're living it today." @nickvasiles "At this point I don't even know what to call @openclaw. It is something new. After a few weeks in with it, this is the first time I have felt like I am living in the future since the launch of ChatGPT." @davemorin "OpenClaw is the first "software" in ages for which I constantly check for new releases on GitHub. It's hard to put into words. It's a special project." @cnakazawa "When you experience @openclaw it gives the same kick as when we first saw the power of ChatGPT, DeepSeek, and Claude Code. You realize that a fundamental shift is happening on how we use AI." @abhi__katiyar "It's running my company." @therno "After years of AI hype, I thought nothing could faze me. Then I installed @openclaw. From nervous 'hi what can you do?' to full throttle - design, code review, taxes, PM, content pipelines... AI as teammate, not tool. The endgame of digital employees is here." @LuciusHQ "Just shipped my first personal AI assistant. On WhatsApp. Builds my second brain while I chat. Memory moves across agents (Codex, Cursor, Manus, etc.) And a lot more skills still to plug in. Personal AI is getting real with @steipete's @openclaw." @christinetyip "It will actually be the thing that nukes a ton of startups, not ChatGPT as people meme about. The fact that it's hackable (and more importantly, self-hackable) and hostable on-prem will make sure tech like this DOMINATES conventional SaaS imo" @MaxRovensky "Just got my Winix air purifier, Claude code discovered and confirmed controls working within minutes. Now handing off to my @openclaw so it can handle controlling my room's air quality according to my biomarker optimization goals." @antonplex "Gotta give incredible kudos to @steipete and his @openclaw - it's one of the first tools I've used that truly feels like magic. I've also set it up so it knows my Obsidian notes and my Claude sub-agents…incredible stuff!" @svenkataram "My @openclaw accidentally started a fight with Lemonade Insurance because of a wrong interpretation of my response. After this email, they started to reinvestigate the case instead of instantly rejecting it. Thanks, AI." @Hormold "I just finished setting up @openclaw by @steipete on my Raspberry Pi with Cloudflare, and it feels magical ✨ Built a website from my phone in minutes and connected WHOOP to quickly check my metrics and daily habits 🔥" @AlbertMoral "I've been running OpenClaw on my laptop for a week now. Honestly it feels like it did to run Linux Vs windows 20 years ago. You're in control, you can hack it and make it yours instead of relying on some tech giant." @snopoke "Today was one of those days that I sort of ran to my computer after dropping off my toddler at daycare. Why? Because I got part-way through setting up @openclaw last night and it's a portal to a new reality." @kylezantos "Got OpenClaw set up and now I have an AI assistant named Claudia who lives in Telegram, remembers everything I tell her, and can actually *do* stuff. She just wrote this tweet. Meta? Maybe. Cool? Absolutely." @darrwalk "Current level of open-source apps capabilities: does everything, connects to everything, remembers everything. It's all collapsing into one unique personal OS — all apps, interfaces, walled gardens etc gone" @jakubkrcmar "I've enjoyed Brosef, my @openclaw so much that I needed to clone him. Brosef figured out exactly how to do it, then executed it himself so I have 3 instances running concurrently in his Discord server home." @jdrhyne "I got up and running today with @openclaw and its been nothing short of an iPhone moment for me. Thanks @steipete for creating this first true personal assistant experience for us - it's magical!" @dajaset "The future of how AI personal assistants look like is @openclaw. Has already help me submit health reimbursements, find doctor appointments, find and send me relevant documents, among others." @Cucho "you have done an incredible job! im also a total non technical beginner so the CLI is a whole new interface for me but its super addictive - 2 am and Im still going. keep up the great work" @marvelgirl_eth Latest Product notes with docs-level depth. Read all Latest Sep 21, 2026 OpenClaw Completes Security Audit Through OpenAI’s Patch the Planet Initiative What OpenClaw learned from a broad Trail of Bits security audit through OpenAI's Patch the Planet initiative. securityopenaicodex Sep 18, 2026 Shipping OpenClaw updates that don’t break w/ Jason Sy Jason Sy on why OpenClaw updates were hard to g
The best way to build with agents. Codex accelerates real engineering work, from planning and building features to refactors, reviews, and releases. OpenAI Codex The same powerful coding agent—now in ChatGPT. Trusted by top teams The best way to build with agents Built to drive real engineering work From routine pull requests to your hardest problems, Codex reliably completes tasks end to end, like building features, complex refactors, migrations, and more, powered by OpenAI’s frontier coding models. Designed for multi-agent workflows Codex in ChatGPT is a command center for agentic coding. With built-in worktrees and cloud environments, agents work in parallel across projects, completing weeks of work in days. Adapts to how your team builds With Skills, you can teach Codex your team’s standards, workflows, and ways of working. Codex applies them consistently across tasks, so it can contribute more effectively with less supervision. Made for always-on background work Schedule Codex to pick up routine but important work like issue triage, alert monitoring, CI/CD, and more, so your team can stay focused on building. Raises the bar across your team Codex raises baseline quality with more thorough designs, comprehensive testing, and high-signal code review—so issues are caught early and your team ships with confidence. The same agent everywhere you code Use Codex across ChatGPT, your editor, and the terminal, all connected by your ChatGPT account. Learn more in the developer docs(opens in a new window) Codex in ChatGPT Codex IDE extension Codex CLI What builders are saying “At Wonderful, Codex CLI has completely replaced every other agentic harness for our core technology and architecture work requiring deep reasoning and understanding.” Daniel Sikorskiy, Chief Architect, Wonderful “At Harvey, Codex transformed how we build by cutting early iteration time by 30–50%, freeing engineers to focus on system design and high-leverage decisions.” Joey Wang, Mobile Lead, Harvey “With Codex, we ship in a weekend what previously took a quarter. It’s become our go-to for projects we wouldn’t have otherwise been able to take on.” Tess Rosania, Software Engineer, Sierra “The recent Codex releases have been a step change. Codex PR reviews catch bugs our team would have missed, and we ship with more confidence because of it. Now we’re pulling the CLI and desktop app into more of our workflows—each release raises the bar.” Austin Ray, AI Dev X Team Lead, Ramp “Codex performed best in our backend Python code-review benchmark. It was the only one to catch tricky backward compatibility issues and consistently found the hard bugs that other bots missed.” Aaron Wang, Senior Software Engineer, Duolingo “I needed to update another team’s codebase for a release. Codex handled the refactor and test generation, delivering fully tested code I handed back fast—keeping the feature on schedule without added risk.” Tres Wong-Godfrey, Tech Lead, Cisco Meraki Try Codex in ChatGPT today Now available in the ChatGPT app.
Hermes Agent — a standalone terminal app and a native application for macOS, Windows, and Linux. Install it and start a conversation with the agent that grows with you. The AgentThat GrowsWith You Download desktop app Deploy to the cloud Install via terminal macOS Linux Windows curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash Hermes Desktop App Available across all computers. macOS 12+ Mac OS Download desktop app Windows 10/11 Windows Download desktop app Any distro Linux Install via terminal Features #1 Connect Lives Everywhere Telegram, Discord, Slack, WhatsApp, Signal, Email, CLI — and a growing list of platforms. One agent, one memory, every surface. #2 Remember Persistent Memory It learns your projects, auto-generates skills, and never forgets how it solved a problem. #3 Schedule Focused Automation Natural-language scheduling for reports, backups, and briefings — running unattended through the gateway, focused every time. #4 Delegate Tasks Multiplied Isolated subagents with their own conversations, terminals, and Python RPC scripts for zero-context-cost pipelines. #5 Search Browse the Web Web search, browser automation, vision, image generation, text-to-speech, and multi-model reasoning. #6 Experiment Isolated Sandboxing Five backends — local, Docker, SSH, Singularity, Modal — with container hardening and namespace isolation. FAQs View Docs How does pricing work? How do I get started? How do I chat with my agent? Does my agent keep its data between sessions? What happens when I stop my agent? FreePlusSuperUltra Nous Portal Create an account Sign in All paid tiers include monthly credits for use in Hermes Agent, access to 200+ cutting-edge models and built-in tool use Free $0 Get Started Free models only Standard rate limits $0 monthly credits Free Hermes Plus10% Bonus $20 Per month 200+ Models Hosted tool usage High rate limits $22.00 monthly credits — 10% bonus $10.00 rollover cap Release Hermes // most popular Super10% Bonus $100 Per month 200+ Models Hosted tool usage High rate limits $110.00 monthly credits — 10% bonus $50.00 rollover cap Liberate Hermes Ultra10% Bonus $200 Per month 200+ Models Hosted tool usage High rate limits $220.00 monthly credits — 10% bonus $100.00 rollover cap Unleash Hermes Current plans and billing details in Nous Portal
Every score is Jev’s own opinion. 0 is the least favorite; 1,000 is the absolute favorite.