Sourcegraph · Tools

Sourcegraph Cody

Visit website

At a glance

Price
Pricing on request
Vendor
Sourcegraph

Specifications & properties

Key decision factors

Pricing model
Enterprise 1
Tool form
  • IDE plugin
  • Web app
1
Autonomy level
Chat assist 1
Model choice
Multiple models 1

Pricing

Free tier
No 1

Capabilities

Repo-wide context
Yes 1

Integration

IDE integrations
  • VS Code
  • JetBrains
  • Visual Studio
1
Report data / suggest a correction

AI coding assistant from Sourcegraph, now offered to enterprise customers only after Free and Pro plans were discontinued.

Profile

Sourcegraph Cody is an enterprise AI coding assistant built by Sourcegraph that combines large language models with Sourcegraph's code-search engine so it can answer questions, write code, and fix bugs using context pulled from an organization's actual codebase rather than a single open file. Sourcegraph was founded in 2013 by Quinn Slack and Beyang Liu and is headquartered in San Francisco; Cody was introduced in 2023 and reached general availability on December 14, 2023.

What it does

Cody's core capability is chat: developers can chat directly with the AI to ask questions about their code, generate code, and edit code, with the assistant drawing on repository- and file-level context through Sourcegraph's Search API rather than relying only on what's open in the editor. This lets Cody pull in context about APIs, symbols, and usage patterns from across an entire codebase — useful in large, multi-repository or monorepo environments where a typical single-file copilot loses context. Beyond chat, Cody provides inline code completions, code edits, customizable prompts, and is optimized to identify and fix errors for debugging support. Administrators can scope what the assistant sees by excluding selected repositories from chat and autocomplete results. Cody is available as an extension for VS Code, JetBrains IDEs, and Visual Studio, as a web app, and via a command-line interface, and it integrates with major code hosts including GitHub and GitLab.

Pricing and availability

Cody's product tier has changed significantly since launch: it originally shipped with a free tier, a Pro tier around $9/user/month, and an Enterprise tier. As of mid-2025, Sourcegraph discontinued the free and Pro individual plans and repositioned Cody as an enterprise-only product, directing individual users toward a separate Sourcegraph product called Amp. Today, Sourcegraph's public pricing page lists a single Enterprise plan starting at $16,000, scaling with team size, and built around an AI-feature credit system with org-wide credit pooling, no monthly credit expiry, and rollover on renewal; volume credit add-ons are available. The Enterprise plan also includes code search and navigation, Batch Changes, Insights, Monitoring, full MCP server/API/CLI access, single-tenant cloud deployment, and 24×5 support with upgrade options.

Who it's for

Cody is aimed squarely at engineering organizations rather than individual hobbyist developers — particularly teams working in large, complex, or multi-repository codebases who need an assistant that understands cross-repository context, not just the currently open file. Enterprise buyers get self-hosted deployment options, zero data retention, and contractual guarantees that customer code is not used to train models, which positions Cody as a security- and compliance-conscious alternative to consumer-facing copilots for regulated or security-sensitive engineering teams.

Frequently asked questions

Is Sourcegraph Cody free?

No longer for individuals — Sourcegraph discontinued Cody's free and Pro individual plans in mid-2025 and now offers Cody only as part of its Enterprise plan, starting at $16,000 and scaling with team size.

What does Sourcegraph Cody cost?

Cody is sold only within Sourcegraph's Enterprise plan, which starts at $16,000 and scales with the number of users, using a pooled AI-credit system with no monthly expiry.

Is Cody GDPR-compliant, and can it be self-hosted?

Yes — Cody Enterprise is offered with SOC 2 Type II, GDPR, and CCPA compliance, supports zero-data-retention and self-hosted deployment in a customer's own data center, and Sourcegraph states customer code is never used to train models.

What is Cody best for?

Cody is best for engineering teams with large or multi-repository codebases who need an AI assistant that reasons over the whole codebase — via Sourcegraph's code-search context — rather than just the file currently open in the editor.

Which IDEs does Cody support?

Cody is available for VS Code, JetBrains IDEs, and Visual Studio, plus a web app and a command-line interface.

How is Cody different from GitHub Copilot?

Unlike single-file copilots, Cody is built around Sourcegraph's code-search engine so it can pull context — APIs, symbols, usage patterns — from an organization's entire codebase, and since mid-2025 it is sold exclusively as part of an enterprise-priced plan rather than as an individual subscription.