Sourcegraph Cody
Enterprise AI coding assistant with repo-wide context
An enterprise AI coding assistant built on Sourcegraph's code search engine, offering chat, autocomplete and inline edits with context drawn from an entire codebase rather than just open files.
Picked for AI Tools for Regulated Industries.
Cody’s differentiator has always been context: because it’s built on top of Sourcegraph’s code search and code graph, it can pull relevant snippets from across an entire codebase rather than just whatever files happen to be open in the editor. For a small project that barely matters, but in a sprawling monorepo it’s the difference between an assistant that guesses and one that actually knows how a function is used three services away.
That focus makes it a fit for large, established engineering organizations — the kind already running Sourcegraph for code search — that need AI assistance to reason about a genuinely large codebase, and that can also justify an enterprise contract and, often, a self-hosted or air-gapped deployment for compliance reasons.
The significant change worth flagging is that Cody is no longer a product an individual developer or small team can just sign up for. Sourcegraph discontinued the free and Pro tiers in mid-2025, and Cody now ships exclusively as part of Sourcegraph’s Enterprise plan, with contracts reported from roughly $49–59 per user per month on an annual commitment, layered on top of a platform that itself starts in five figures. It’s also worth knowing that Cody’s agentic, multi-file editing ability hasn’t kept pace with agent-first competitors — its edge is retrieval and context, not autonomous task execution.
Features
Codebase-wide context
Pulls context from Sourcegraph's code search and code graph across an entire monorepo, not just the files open in the editor.
Multi-IDE chat and autocomplete
Chat, inline edit, autocomplete and explain-code across VS Code, JetBrains IDEs and Neovim.
Self-hosted and air-gapped deployment
Enterprise customers can run Sourcegraph fully on-premises with a no-training guarantee over their code.
Cross-repository code search
Search every repository at once, which is the underlying capability the AI answers are built on and useful on its own.
Prompt library
Shared, reusable prompts so an organisation's common tasks are standardised rather than reinvented per developer.
Model choice
Swap the model behind chat and completion, including for teams whose procurement has already settled on one vendor.
Use cases
- Large organizations with monorepos where context across the whole codebase matters
- Regulated enterprises that need self-hosted or air-gapped AI coding tools
- Teams already using Sourcegraph code search who want AI chat layered on top
- Answering where is this used across hundreds of repositories
- Standardising common AI prompts across an engineering organisation
- Assessing the blast radius of a change before making it
Compare Sourcegraph Cody head to head
Side-by-side comparisons, on pricing, platforms and where each one wins.
Looking for something similar?
Other code tools worth comparing against Sourcegraph Cody.