Kilo CLI 1.0: Unleashing Open Source AI Coding in Your Terminal, Powered by 500+ Models

Tired of being locked into a single AI model or a specific development environment? The remote-first AI coding startup Kilo feels your pain. This week, Kilo, backed by GitLab co-founder Sid Sijbrandij, dropped a bombshell: Kilo CLI 1.0. This isn’t just an update; it’s a complete rebuild of their command-line tool, boasting support for over 500 different AI models, from industry giants to open-source powerhouses like Alibaba’s Qwen.

The Future of AI Development Isn’t a Single Interface

This release signals a significant strategic shift. While many are still focused on IDE-centric “sidebar” models (think Cursor, GitHub Copilot) or dedicated apps (like OpenAI Codex), Kilo is taking a different path. Just weeks after launching a Slackbot that allows developers to ship code directly from Salesforce’s popular messaging service (powered by Chinese AI startup MiniMax, no less!), Kilo is betting on embedding AI capabilities into every fragment of the professional software workflow.

Kilo CEO and co-founder Scott Breitenother articulated this vision perfectly in a recent interview: “This experience just feels a little bit too fragmented right now… as an engineer, sometimes I’m going to use the CLI, sometimes I’m going to be in VS Code, and sometimes I’m going to be kicking off an agent from Slack, and folks shouldn’t have to be jumping around.” Kilo CLI 1.0 is specifically “built for this world… for the developer who moves between their local IDE, a remote server via SSH, and a terminal session at 2 a.m. to fix a production bug.”

Rebuilding for ‘Kilo Speed’ with an Open Source Superpower

At its heart, Kilo CLI 1.0 represents a fundamental architectural shift. While 2025 saw senior engineers embracing AI vibe coding, Kilo believes 2026 will be defined by the adoption of agents that can manage end-to-end tasks independently. The new CLI is built on an MIT-licensed, open-source foundation, specifically designed for those critical terminal sessions during production incidents or deep infrastructure work.

For Breitenother, open-source isn’t just a choice; it’s a philosophy: “When you build in the open, you build better products. You get this great flywheel of contributors… Honestly, some people might say open source is a weakness, but I think it’s our superpower.”

Beyond Autocompletion: Agentic Modes for Every Task

The core of Kilo’s “agentic” experience moves far beyond simple autocompletion. The CLI supports multiple operational modes, each tailored for different engineering needs:

• Code Mode: For high-speed generation and multi-file refactors.
• Architect Mode: For high-level planning and technical strategy.
• Debug Mode: For systematic problem diagnosis and resolution.

Solving Multi-Session Memory: No More AI Amnesia

One of the persistent frustrations with AI agents is their tendency to lose context between sessions — “AI amnesia.” Kilo tackles this head-on with its innovative “Memory Bank” feature. This system maintains state by storing context in structured Markdown files directly within your repository. This ensures that an agent working in the CLI has the exact same understanding of your codebase as the one in a VS Code sidebar or a Slack thread.

This synergy between the new CLI and “Kilo for Slack” is central to the company’s “Agentic Anywhere” strategy. Launched in January, the Slack integration already allows teams to fix bugs and push pull requests directly from a conversation, and now it shares context seamlessly with your terminal.

“Engineering teams don’t make decisions in IDE sidebars. They make them in Slack,” Breitenother emphasized, highlighting Kilo’s ability to ingest context from across multiple repositories simultaneously, unlike some competitors limited by single-repo configurations.

Extensibility and the Power of Model Agnosticism

A critical technical advancement in Kilo CLI 1.0 is its support for the Model Context Protocol (MCP). This open standard enables Kilo to communicate with external servers, extending its capabilities beyond local file manipulation. Through MCP, Kilo agents can integrate with custom tools and resources — imagine connecting to internal documentation servers or third-party monitoring tools — effectively turning the agent into a specialized, highly integrated member of your engineering team.

This extensibility underscores Kilo’s unwavering commitment to model agnosticism. While MiniMax powers their Slack integration by default, the CLI and extension support a massive array of over 500 models, including top-tier options from Anthropic, OpenAI, and Google Gemini. You pick the best tool for the job.

Disrupting the Economy of ‘AI Output Per Dollar’ with Kilo Pass

Kilo isn’t just innovating on the technical front; they’re also tackling the economics of AI development with “Kilo Pass.” This subscription service is designed for transparency, charging exact provider API rates with zero commission. Essentially, $1 of Kilo credits equals $1 of provider costs.

Breitenother is critical of the opaque “black box” subscription models prevalent elsewhere: “We’re selling infrastructure here… you hit some sort of arbitrary, unclear line, and then you start to get throttled. That’s not how the world’s going to work.”

Kilo Pass tiers even offer “momentum rewards” with bonus credits for active subscribers:

• Starter ($19/mo): Up to $26.60 in credits.
• Pro ($49/mo): Up to $68.60 in credits.
• Expert ($199/mo): Up to $278.60 in credits.

And for early birds, Kilo is offering a “Double Welcome Bonus” until February 6th, giving users 50% free credits for their first two months. This flexibility is a major draw for power users like Sylvain: “Kilo Pass is exactly what I’ve been waiting for. I can use my credits when I need them and save them when I don’t—it finally fits how I actually use AI.”

Kilo vs. The Competition: A Production-Ready, Open Middle Path

Kilo CLI 1.0 enters a competitive landscape, going head-to-head with terminal-native heavyweights like Anthropic’s Claude Code and Block’s Goose. In the broader IDE space, OpenAI recently launched a new Codex desktop app for macOS.

• Claude Code offers a polished experience but comes with vendor lock-in and high costs, with limits often exhausted quickly on large codebases.
• OpenAI’s new Codex app similarly favors a platform-locked approach, aiming to defend OpenAI’s ecosystem.
• Goose provides an open-source alternative that runs locally for free but appears more localized and experimental.

Kilo positions itself as the intelligent middle path: a production-hardened tool built on the MIT-licensed OpenCode foundation, offering a Terminal User Interface (TUI) that allows engineers to swap between over 500 models. This portability empowers teams to select the best cost-to-performance ratio for any task — a lightweight model for docs, a frontier model for complex debugging.

Security is also paramount, with Kilo ensuring models are hosted on U.S.-compliant infrastructure like AWS Bedrock, keeping proprietary code secure while leveraging efficient intelligence. Kilo’s open-core nature allows for a “superpower” level of community auditing and contribution, contrasting with the dual-use concerns often associated with more closed AI platforms.

The Future: A ‘Mech Suit’ for the Mind

With $8 million in seed funding and a “Right of First Refusal” agreement with GitLab extending to August 2026, Kilo is strategically positioning itself as a cornerstone of the next-generation developer stack. Breitenother views these tools not as replacements, but as “exoskeletons” or “mech suits” for the human mind.

“We’ve actually moved our engineers to be product owners,” Breitenother reveals. “The time they freed up from writing code, they’re actually doing much more thinking. They’re setting the strategy for the product.” By unbundling the engineering stack — separating the agentic interface from the model, and the model from the IDE — Kilo is laying the groundwork for a future where developers think architecturally while machines handle the structural heavy lifting. “It’s the closest thing to magic that I think we can encounter in our life,” Breitenother concludes.

For those chasing “Kilo Speed,” the IDE sidebar is truly just the beginning. The future of AI coding is open, fluid, and ready for your terminal.

Source: Original Article