Sacra estimates that Cline hit $5M in annual recurring revenue (ARR) in August 2025.

The company explicitly passes through AI inference at cost with no markup, which means the economically relevant revenue figure is software licensing revenue, enterprise contracts and eventual team seat fees, rather than total payment volume flowing through the platform.

The install base grew from 1M in March 2025 to 2.7M by July 2025, 3.8M by November 2025, and 5M by January 2026, with GitHub contributor growth of over 4,700% year-over-year cited in late 2025. That top-of-funnel expansion creates a larger conversion opportunity once the Teams plan moved off its free promotional period after Q1 2026, at $20 per user per month with the first 10 seats free.

Cline raised a $27M Series A in July 2025 at a post-money valuation of $110M, led by Emergence Capital.

Before the Series A, the company had raised seed funding, bringing total capital raised to $32M across both rounds. Seed participants included Pace Capital, 1984 Ventures, Essence VC, and Cox Exponential, alongside angel investors Jared Friedman, Eric Simons, Logan Kilpatrick, Addy Osmani, and Theo Browne.

The Series A was announced alongside the launch of Cline Teams and the company's move into enterprise, with stated uses of proceeds including expansion beyond VS Code to additional editors and the build-out of enterprise governance features.

Cline is an AI coding agent embedded in a developer's editor or terminal that executes multi-step software engineering tasks from start to finish.

A developer typically opens Cline from the VS Code sidebar, the JetBrains plugin, or the terminal CLI, selects a model and provider, and types a task in plain language, something like "add OAuth login to this app" or "refactor the data layer to use the new schema."

Cline then works through the repository in a pattern similar to a junior engineer, reading files, tracing function definitions, searching for patterns, and building a picture of what needs to change.

After analyzing the codebase, it begins making edits, writing to files directly in the editor, showing diffs, running terminal commands like package installs or test runners, watching the output, and iterating to fix anything that breaks.

In web projects, it can also open a browser, interact with the running app, take screenshots, and read console logs to debug runtime or UI issues rather than limiting itself to static code.

Every file write, terminal command, and browser action can require explicit developer approval before execution. Users can tighten or loosen those controls depending on how much they trust the task at hand, from reviewing every micro-step to enabling a more autonomous mode where Cline runs freely and the developer reviews the result.

Checkpointing is a core feature: Cline automatically saves snapshots of the project in a shadow Git repository every time it makes a change, so a developer can roll back to any prior state without losing the conversation history. That reduces the cost of a wrong turn and makes longer autonomous runs more manageable.

For longer or more complex tasks, Cline offers a Deep Planning mode that investigates the codebase and asks clarifying questions before writing any code, producing an architecture plan first. Memory Bank provides a structured way to persist project context, goals, architecture decisions, and active work, across sessions so it does not lose the thread on multi-day efforts. Focus Chain maintains an explicit to-do list and progress tracker for tasks that span many files.

Cline also supports subagents: the main agent can spin up parallel read-only research agents to inspect different parts of a large codebase simultaneously, then synthesize their findings before acting. This is an experimental feature intended for very large repositories where stuffing everything into one context window is not practical.

The platform is extensible through Model Context Protocol (MCP), which lets teams connect Cline to external tools and data sources, Jira tickets, internal APIs, browser automation, cloud consoles, via a marketplace of one-click-installable MCP servers. Teams can also define Rules (persistent project instructions), Skills (modular instruction packs for recurring domains), Workflows (slash-command-driven runbooks for repeatable processes), and Hooks (deterministic validation or policy checks that fire at specific moments in the agent loop).

On the model side, Cline supports its own hosted provider as well as bring-your-own-key integrations with Anthropic, OpenAI, Google, OpenRouter, AWS Bedrock, Azure, Vertex AI, and local runtimes like Ollama and LM Studio. The developer chooses which model to use for each task, and Cline routes accordingly.

Cline operates as a B2B open-source software company with a three-layer monetization structure built around enterprise governance rather than inference margin.

The open-source core is free for individual developers, who can install the VS Code extension or CLI and connect their own model API keys at cost. This creates a large organic top-of-funnel of engineers who adopt Cline on their own before any procurement conversation.

The middle layer is a Teams plan at $20 per user per month (with the first 10 seats free), which adds centralized billing, admin controls, provider restrictions, usage dashboards, and priority support. The bottom layer is custom-priced Enterprise, which adds SSO, audit logs, VPC and on-premises deployment, global policy enforcement, and dedicated support.

A distinctive structural choice is that Cline explicitly passes through AI inference at cost with no markup, and supports bring-your-own-key arrangements where customers connect their existing Anthropic, OpenAI, Bedrock, or other contracts directly. This differs from much of the AI tooling market, where vendors often embed margin into token usage. Cline's model is that transparent inference economics can make procurement easier for sophisticated buyers, while the company's revenue comes from software controls and governance packaging rather than model resale.

That approach also leaves Cline relatively asset-light on infrastructure. Because much of the compute flows directly from the developer's environment to their chosen model provider, Cline does not need to finance or subsidize large GPU workloads.

Its go-to-market motion is bottoms-up developer adoption that converts to top-down enterprise expansion. Engineers install the free product, use it in their daily workflow, and eventually create internal demand for the organizational controls available only in the paid tiers, SSO, audit trails, centralized billing, and policy enforcement. The free Teams promotional period through Q1 2026 was designed to accelerate that internal spread before the billing switch flipped.

The extensibility features, MCP integrations, Rules, Hooks, Workflows, and Skills, function as switching-cost deepeners. A team that has built custom MCP connectors to its internal systems, encoded its engineering standards into Rules, and automated its review processes through Hooks has embedded Cline into its engineering operating system in a way that makes replacement expensive regardless of what competitors offer.

The flywheel is: open-source distribution builds community and installs, community builds integrations and ecosystem trust, ecosystem trust makes enterprise standardization more attractive, enterprise revenue funds product development that keeps the open-source core competitive.

The AI coding agent market in 2025 and 2026 has split into two camps: app generation and prototyping tools like Lovable, Bolt.new, and Vercel's v0, and in-flow coding agents that work inside existing codebases. Cline competes in the second camp, where rivals are better funded, more deeply integrated into developer infrastructure, and increasingly moving toward the same agentic capabilities Cline introduced earlier.

Cursor is the most direct commercial threat. It rebuilt the editor experience around AI, and its enterprise page cites adoption across more than 53% of Fortune 1000 companies and over 50,000 enterprises. Cursor's background agents let developers offload tasks to remote environments asynchronously, and its Bugbot and GitHub app integration extend it into code review and issue remediation, making it a workflow platform rather than just a coding UI. Cursor raised a $105M Series B at a $2.5B valuation in early 2025, giving it resources to deepen that integration.

Windsurf competes on similar ground with its Cascade agentic core, which handles code and chat modes, planning, tool use, and linter integration across VS Code and JetBrains. Windsurf claims over 4,000 enterprises in production and markets a clean upgrade path from free to Pro, Teams, and Enterprise with RBAC, SSO/SCIM, and centralized billing. Both Cursor and Windsurf compete with Cline by offering a more opinionated, lower-configuration experience. Buyers who want a single bundled editor rather than an open, composable stack tend to choose those products.

Continue is Cline's closest structural analogue in the open-source space. It emphasizes model choice, on-premises data planes, and configurable governance, with a Teams plan at $10 per developer per month and enterprise adding SSO and separation of control and data planes. Continue has roughly 32,000 GitHub stars and is pushing toward CI enforcement and source-controlled AI checks, which could pressure Cline from below on openness while differentiating on pipeline-native governance.

Aider competes among terminal-native power users with a git-first, highly scriptable approach that supports cloud and local LLMs. With roughly 41,000 GitHub stars, 5.3M installs, and 15B tokens per week of usage, Aider shows that a minimalist open-source CLI agent can accumulate substantial adoption. For developers who want lightweight git repo edits without a heavier GUI workflow, Aider is a credible substitute that caps Cline's pricing power in that segment.

OpenHands takes a broader platform approach, spanning CLI, GUI, SDK, cloud, and enterprise self-hosted deployments, with integrations into Slack, Jira, and Linear. It describes itself as an open agent platform for orchestrated software engineering work rather than an in-editor assistant, which makes it a threat if enterprise buying criteria shift from best-in-editor copilot to best controllable agent runtime.

The most consequential competitive pressure comes from model providers building their own coding agents. Anthropic's Claude Code is a terminal-native agent that overlaps directly with Cline CLI, with transparent cost disclosures and tight alignment between model behavior and tool UX. OpenAI's Codex now spans app, CLI, IDE, and cloud, and is bundled into existing ChatGPT subscriptions, lowering activation friction. Google's Gemini CLI is open-source, free-tiered, and backed by a Vertex AI enterprise path, with roughly 98,800 GitHub stars, indicating the distribution advantage hyperscalers bring to this category.

GitHub Copilot competes from the strongest distribution position, embedded directly in VS Code, GitHub repositories, pull requests, and CI workflows. Its coding agent operates in GitHub-hosted environments as part of the PR workflow, and Copilot CLI now supports ACP in public preview. GitHub's edge is not feature parity with Cline but control of the repository, issue, PR, and developer identity surface. That means it can absorb more of the surrounding workflow by default.

Cognition's Devin, which has raised $196M, and Magic AI, which raised $320M in 2024, represent the fully autonomous end of the spectrum, where the agent operates with minimal human supervision. These products compete less on in-editor UX and more on delegating entire engineering tasks end-to-end.

As more code is AI-authored, a new ecosystem is forming around validating, scanning, and remediating agent-written code. Companies like Momentic, Semgrep, Endor Labs, and DryRun are building adjacent to the coding agent layer. This creates both a competitive dynamic, if testing and security tooling becomes a buying criterion bundled with the agent, and a partnership opportunity for Cline to integrate these tools via MCP rather than build them natively.

Cline's most immediate TAM expansion is the CLI becoming a first-class product rather than a secondary interface. CLI 2.0, launched in February 2026, added interactive TUI sessions, headless operation, stdin/stdout piping, JSON output, CI/CD integration, and parallel isolated agent instances.

That shifts Cline from a developer seat product to a programmable agent runtime that can be embedded in shell scripts, pre-merge checks, internal developer portals, and automated pipelines, budget pools owned by DevOps, platform engineering, and release engineering teams rather than individual developers.

The subagents feature points to a multi-agent orchestration model where Cline decomposes large tasks across parallel research agents before acting. If that matures into a full orchestration layer, Cline can capture more of the software lifecycle: architecture discovery, migration planning, code review, test generation, and incident triage.

Cline's enterprise launch in late 2025 opened a new buyer set: engineering leaders, CIOs, security teams, and procurement organizations rather than engineers with credit cards.

The highest-upside segment within enterprise is regulated and zero-trust environments, automotive, defense, finance, healthcare, and large industrials, where closed SaaS coding tools are often rejected by security teams before they reach developers. Cline's architecture, which keeps code processing local and routes inference to customer-controlled providers, is designed to clear those procurement gates.

Adjacent internal users beyond software engineers also represent seat expansion. AppSec teams using Cline for vulnerability remediation, SRE teams using it for incident triage, QA teams using it for test generation, and technical writers using it for documentation maintenance represent incremental seats within existing enterprise accounts.

The MCP Marketplace is a TAM expansion mechanism that lets Cline absorb adjacent categories without building every integration natively.

With hundreds of MCP servers covering Jira, cloud consoles, browser automation, observability tools, and internal APIs, Cline can expand from code editing into developer operations more broadly. The more workflows a team runs through Cline, not just writing code but also triaging issues, managing deployments, and querying internal systems, the more Cline becomes SDLC infrastructure rather than a narrow coding assistant.

For enterprise buyers, centralized MCP governance, approved tool catalogs, policy controls, auditability around which external tools agents may invoke, creates a natural upsell from the base agent product into a managed agent tooling control plane. That moves Cline up the stack from assistant vendor to infrastructure vendor, which carries higher ACVs and stronger retention.

Cline's VPC, on-premises, and air-gapped deployment options are the primary geographic expansion mechanism, enabling rollouts in countries and sectors with strict cross-border data rules or internal restrictions on external SaaS.

The automotive vertical, where modern vehicles are large software systems subject to cybersecurity and safety requirements, is an explicit target. The same template extends to aerospace, defense, medtech, banking, telecom, and energy, segments where private inference, audit trails, and policy controls are prerequisites for adoption rather than optional add-ons.

Platform absorption: Major IDE, source control, and model providers are converging on agentic coding inside the tools developers already use. GitHub Copilot owns the repository and PR surface, Cursor and Windsurf own the editor experience for a large share of the market, and Anthropic, OpenAI, and Google are shipping first-party CLI agents that overlap directly with Cline's terminal story. If these platforms make similar functionality native, Cline could be compressed into a thin orchestration layer with limited pricing power.

Open-source monetization lag: Cline's free individual tier, no-markup inference positioning, and Teams plan that was free through Q1 2026 mean adoption has outpaced recognized revenue. Conversion from 5M+ installs to paying enterprise and team accounts depends on Cline's governance and compliance features being strong enough to justify procurement, and on the Teams plan activating at scale after the promotional period, neither of which is guaranteed given the competitive alternatives available at lower friction.

Supply-chain trust: In February 2026, an unauthorized publish of Cline CLI version 2.3.0 via a compromised npm token created a supply-chain incident, even though the company reported no malicious payload was delivered and the exposure was limited to the CLI package. For a product that asks enterprise security teams to trust an agent with code editing and terminal command execution, any release-security incident raises the bar for procurement in high-compliance accounts and gives competitors a concrete narrative around the risks of open plugin ecosystems.