Introducing ClinePass: a low-cost monthly subscription to access the best open weights models across Cline. - $9.99/month subscription to give you 2-5x discounted access - Models included: GLM 5.2, Kimi K2.7-Code, Kimi K2.6, Deepseek-v4-pro, Deepseek-v4-flash, Minimax-M3, Mimo-v2.5, MiMo-V2.5-Pro, Qwen3.7-max, Qwen3.7-plus (more to come!) - Use it across Cline’s CLI and IDE Extension Run `npm i -g cline` and get started for just $1.99 Thank you to collaboration from Z.ai, DeepSeek AI, Kimi (Moonshot AI), MiniMax, Alibaba Cloud, @Xiaomi MiMo Learn more: https://lnkd.in/grx-93gD
About us
With over 5M+ installs and 57k+ GitHub stars, a community that grew 4,704% year over year. Cline is the leading open-source AI coding agent. Trusted by the world's largest engineering organizations and individual developers alike, Cline is model agnostic, provider agnostic, and IDE agnostic, meeting engineers everywhere they work while keeping code on your infrastructure. Teams choose Cline because it's the only coding agent that combines cutting-edge capability with enterprise-grade security, using existing provider contracts directly with zero markup. Cline has raised $32M in combined Series A and Seed funding, led by Emergence Capital and Pace Capital, with participation from 1984 Ventures, Essence VC, and Cox Exponential.
- Website
-
https://cline.bot/
External link for Cline
- Industry
- Software Development
- Company size
- 11-50 employees
- Headquarters
- San Francisco, California
- Type
- Privately Held
Locations
-
Primary
Get directions
San Francisco, California 94111, US
Employees at Cline
Updates
-
Cline reposted this
Excited to power GLM-5.2 on Cline! How to use it in about 10 seconds:
-
Today we're introducing Cline Spec Driven, an enterprise AI engineering platform we co-developed with LG CNS. Most AI coding tools are assistants. They autocomplete, they suggest, and they leave the hard enterprise problems to you. Spec Driven is different. It's a team of specialized agents that takes work end to end, from analysis to test, with every step running on your codebase, your business context, and your operational standards. We built it for the environment enterprises actually live in: large brownfield codebases where working code cannot be casually rewritten, where compliance and data isolation are non-negotiable, and where unpredictable token costs are a real budgeting problem. Three principles guide the product: Knowledge Foundation. It learns your codebase and your business so every agent decides and builds with real context instead of guessing. The more tasks it completes, the smarter it gets. Spec-driven Execution. Every spec is written through a fixed pipeline, not ad hoc prompts, so the output is consistent and shaped to fit your existing codebase. Enterprise Deployment. Run an open-weight model on your own infrastructure for complete data isolation, or connect a cloud LLM like Claude or GPT. Bundle it through us or bring your own key. The result is an engineering partner that targets the exact changes you need and touches nothing else. No hallucination. No rewrites. No bill-shock. Huge thanks to the LG CNS team for partnering with us to build this. This is what enterprise-grade agentic engineering looks like. Talk to our team about a proof of concept scoped to your codebase: https://lnkd.in/ergHTe-P
-
🔐 Exciting news — Cline is heading to Marco Island! We’re exhibiting at the Innovate Cybersecurity Summit this week, and we’d love to see you there. 📍 Find us at Booth #718 | 📅 April 19–21 Stop by to: ✅ See how Cline is transforming the way developers write and ship secure code 🎮 Play a fun game at the booth 🎁 Win amazing raffle prizes + grab some swag Security isn’t an afterthought — it’s built into every line. Come chat with our team about how AI-powered development can help you Secure Your Code without slowing you down. 👉 Book your session with us: https://lnkd.in/gN6bysW4 See you on the floor! 👋 #CyberSecurity #InnovateSummit #MarcoIsland #AI #DeveloperTools #Cline #SecureCode #TechEvents
-
-
Spotify are contributing 3 real-world challenges to the Cline AI-Assisted Enterprise Coding Hackathon at HiQ Sweden HQ - organized by 2Hero!🔥 Thanks a million to our idea submitter: Mavina Kandhari, Head of Engineering (Product Area Lead), Strategic Partnerships and Experience. You are force of nature! 👇 Swipe through the challenges below to explore. Hack Day: May 9 - Saturday Grand Finale: May 11 Are you a developer? Apply here: https://lnkd.in/gjQpApXC Everyone eager to learn and explore is welcome to apply! This hackathon is all about sharing knowledge, sparking inspiration, and fostering a supportive community for all. ❤️ PS. Read about the other challenges contributed by Carlsberg Group, Atlas Copco, Saab, Ericsson, Getinge, Electrolux Group, Husqvarna Group, WWF, Bufab and CONSAT: https://lnkd.in/g-yEFDGh
-
Cline gave an AI Agent Access to Blender In this latest installment of Agentic Exploration, we connect the Cline CLI to the Blender MCP. Watch and see what Rayyan Zahid and Tony L. build with this agentically supercharged stack
Agentic Exploration: Cline gave an AI Agent Access to Blender
www.linkedin.com
-
Three Cline agents. Same 120B model. One rule: terminate your opponents' processes before they terminate yours. We tested raw inference across an @NVIDIA_AI_PC DGX Spark, an RTX 4090, and cloud. Time-to-first-token decides the race. Throughput decides who ships. DGX Spark: 42.9 tok/s. RTX 4090: 8.7. Same weights. 4.9x gap. #AIonRTX https://lnkd.in/gVH8dUxd
-
The AI tooling industry talks a lot about productivity gains. It talks much less about what happens when the model underneath your workflow quietly changes. This is the monitoring gap the industry isn't naming clearly enough. Standard observability tells you when your inference provider is down. It tells you nothing about when it's degraded. With more production code being generated by AI, those are not the same risk. We built Cline around the conviction that engineers deserve to have more control over their infrastructure. Saoud covers what that looks like architecturally, and what teams building on AI infrastructure should be demanding from their tooling.
What happens when the model changes and you don't know about it? Most engineering teams treat their inference provider like a stable API. It isn't. Model versions rotate. Output formats shift. Context window handling changes. And you find out through production failures, not changelogs. The lock-in that accumulates isn't dramatic. It's incremental. Every prompt optimized against one model's quirks. Every eval pipeline calibrated to one model's output characteristics. Every parsing layer that assumes a specific response structure. None of it announces itself as a model-specific dependency. It looks like application logic, until you try to switch providers and it breaks quietly, with subtly malformed data flowing into systems that have no reason to reject it. The fix is the same principle engineers already apply to every other critical dependency: abstract the interface, stay agnostic about the implementation. Build your prompt logic, evaluation pipelines, and output parsing against a normalized layer. Make provider-switching a configuration change. We go deep on what that actually looks like in practice. Link in comments.
-
Come build MCP servers with us in Sweden, apply today!
Welcome to Cline AI-Assisted Enterprise Coding Hackathon at HiQ Sweden HQ - organized by 2Hero. Free of charge! Food and beverages included. The challenge: Build an MCP server that enhances enterprises’ capability to leverage coding agents to reliably generate enterprise-grade software solutions faster. You will be trained and challenged to build MCP servers for enterprise coding agents. Your MCP server will enhance the coding agents’ ability to build a certain type of solution or navigate a specific requirement that an enterprise has. Real problems to solve: Ericsson, Carlsberg Group, Atlas Copco, Getinge and other leading companies have contributed real-world challenges for the teams to work on. The game changing tech we are exploring: We are exploring AI coding agents powered by the Model Context Protocol (MCP). MCP is a standard that allows AI models to securely connect directly to our private codebases, internal tools, and company documents. Instead of using generic AI chatbots, this technology creates active, context-aware coding assistants that understand our specific systems. They can automate complex tasks like writing, testing, and debugging code within our exact environment. This combination of autonomous AI agents and secure data access is the specific technology that makes it possible to increase developer productivity 10 to 100 times — fundamentally changing how we build enterprise-grade software faster and better. 🇺🇸🇸🇪 Executives from Cline are flying in from Silicon Valley to meet the Nordic developer community. A great chance to meet some of the world's leading figures in AI-assisted enterprise coding in person, ask them anything and discuss the future of coding agents. Hack Day: 9 May 08:30 – 17:00 (Saturday!) Grand Finale: 11 May 16:30 – 20:30 HiQ Sweden HQ, Stockholm Spots are filling fast, so apply today: https://lnkd.in/gjQpApXC We would ❤️ for you to join!
-