Can I run Cody on my own servers?

Yes. The sheet lists a self-hosted Enterprise option that keeps source code on customer infrastructure. It also warns that self-hosted setup and indexing are non-trivial.

Does Codex work with non-OpenAI models like Claude or Gemini?

No. The sheet says Codex is locked to OpenAI's GPT models, with no Claude or Gemini out of the box.

What does Codex charge for?

The sheet lists Free $0/month, Go $8/month, Plus $20/month, Pro 5x $100/month, and Pro 20x $200/month. Business and Enterprise are Not published. An API Key option is listed where you pay only for the tokens Codex uses, based on API pricing.

The starting price is Not recorded and the verified plan list is none on file. The sheet notes that pricing shifts from per-seat to an enterprise contract at scale.

Which platforms does each tool support?

Codex is on web, mac, Windows, Linux, CLI, and VS Code extension. Cody is on VS Code extension, JetBrains plugin, and web.

Codex vs Sourcegraph Cody: How They Actually Differ

Codex is the pick for a solo founder or small team that wants one agent across terminal, IDE, and cloud with clear monthly pricing from $0. Sourcegraph Cody is the pick for teams with large monorepos that need search-grounded answers and a self-hosted option.

Compiled 2026-08-28 from our catalogue entries for Codex and Sourcegraph Cody.

The short answer

Pick Codex if: You want a single OpenAI GPT-5 agent that follows you across CLI, IDE, desktop, and cloud, reviews GitHub pull requests in the cloud, and ships with a published plan ladder starting at Free $0/month and going to Pro 20x $200/month.
Pick Sourcegraph Cody if: You sit on a very large monorepo, need semantic, regex, and structural search over your own code, want answers grounded in that index rather than the model's training data, and may need a self-hosted Enterprise deployment.
Cheapest way in: Codex Free at $0/month is on file. For Cody, the starting price is Not recorded and the verified plan list is none on file.
What we could not compare: No Cody plans or starting price are on file, so a direct pricing comparison is not possible. Codex's Business and Enterprise prices are Not published. We did not compute any yearly totals.

Codex is OpenAI's agentic coding tool. One agent runs in the terminal CLI, an IDE extension, a desktop app, and a cloud environment, so work can move between surfaces. It reads the shared AGENTS.md instructions file, speaks the Model Context Protocol (MCP), and can spin up a cloud code review on a GitHub pull request that comments inline and proposes fixes. It runs on OpenAI's GPT models — the sheet notes GPT-5.5 as the recommended frontier model — and is included with ChatGPT plans from the free tier up through Pro. The model choice is the tradeoff: it is locked to GPT, with no Claude or Gemini out of the box.

Sourcegraph Cody takes a different starting point: code search first, AI assistant second. It indexes very large monorepos for semantic, regex, and structural search, then feeds that context to the model so answers about internal APIs are grounded in your own code rather than guesses. A self-hosted Enterprise option keeps source on customer infrastructure, which matters for teams with that constraint. The caveats on file are real: self-hosted setup and indexing are non-trivial, and pricing moves from per-seat to an enterprise contract as you scale.

Side by side

Every cell is a field on our catalogue entry for that tool, copied as recorded. Rows where the two differ come first; blank fields say so rather than being filled in.

	Codex	Sourcegraph Cody
Starts at	Free — $0 /month	Not recorded
Runs on	web, mac, win, linux, cli, vscode-ext	vscode-ext, jetbrains-plugin, web
Category	AI Coding Agents	AI Coding Agents
Pricing model	Freemium	Freemium
Public source code	No public repo listed	No public repo listed
Public docs	Not listed	Not listed

Public source code: From the repository link on each catalogue entry, not a licence audit.

Where Codex is the better fit

One agent runs across CLI, IDE, desktop, and cloud, so work started in one surface can continue in another.
Cloud GitHub integration that performs code review on a pull request, comments inline, and proposes fixes.
Reads the shared AGENTS.md standard and supports MCP tools, and is included with ChatGPT plans from the free tier up through Pro.

Full Codex entry · Codex pricing, verified

Where Sourcegraph Cody is the better fit

Indexes very large monorepos for semantic, regex, and structural search.
Self-hosted Enterprise option keeps source code on customer infrastructure.
Combines code-search context with LLM answers, which the sheet says reduces hallucinations on internal APIs.

Full Sourcegraph Cody entry

Which one should you pick?

Solo founder or small team this week

Codex

Published plan ladder starts at Free $0/month with paid tiers up to Pro 20x $200/month, runs on mac, Windows, Linux, web, CLI, and a VS Code extension, and the agent moves between those surfaces without rework.

Enterprise team with a large internal monorepo

Sourcegraph Cody

The sheet calls out indexing of very large monorepos for semantic, regex, and structural search, plus a self-hosted Enterprise option for source code staying on customer infrastructure.

Team that needs AI grounded in internal APIs

Sourcegraph Cody

The fact sheet frames Cody as combining code-search context with LLM answers specifically to cut hallucinations on internal APIs.

Frequently asked questions

Can I run Cody on my own servers?: Yes. The sheet lists a self-hosted Enterprise option that keeps source code on customer infrastructure. It also warns that self-hosted setup and indexing are non-trivial.
Does Codex work with non-OpenAI models like Claude or Gemini?: No. The sheet says Codex is locked to OpenAI's GPT models, with no Claude or Gemini out of the box.
What does Codex charge for?: The sheet lists Free $0/month, Go $8/month, Plus $20/month, Pro 5x $100/month, and Pro 20x $200/month. Business and Enterprise are Not published. An API Key option is listed where you pay only for the tokens Codex uses, based on API pricing.
What does Cody cost?: The starting price is Not recorded and the verified plan list is none on file. The sheet notes that pricing shifts from per-seat to an enterprise contract at scale.
Which platforms does each tool support?: Codex is on web, mac, Windows, Linux, CLI, and VS Code extension. Cody is on VS Code extension, JetBrains plugin, and web.

What we compared

This page states nothing that is not in the two catalogue entries below. The exact text used to write it is reproduced here so you can check our reading, and both tools link through to their own pages and to the vendor.

/tools/codex — checked 2026-07-02 · official site
/tools/sourcegraph-cody — checked 2026-05-25 · official site

Show the source text

## Codex (/tools/codex)
Tagline: OpenAI's coding agent — works in the terminal, IDE, and cloud, reviews GitHub pull requests, and runs on GPT-5 models.
Category: AI Coding Agents
Pricing model: freemium
Starts at: Free — $0 /month
Platforms: web, mac, win, linux, cli, vscode-ext
Public repo: none listed
Docs: none listed
Verified plans (read 2026-08-21): Free $0 /month | Go $8 /month | Plus $20 /month | Pro 5x $100 /month | Pro 20x $200 /month | Business / Enterprise Not published | API Key Pay only for the tokens Codex uses, based on API pricing
Strengths on file: Runs one agent across CLI, IDE, desktop, and cloud — work moves between surfaces.; Strong GitHub integration: a cloud code review on a pull request that catches real bugs and proposes fixes.; Reads the shared AGENTS.md standard and supports MCP tools.; Included with ChatGPT plans, from the free tier up through Pro.
Limits on file: Locked to OpenAI's GPT models — no Claude or Gemini out of the box.; Fewer built-in customization primitives (hooks, sub-agents) than some rivals.
Codex is OpenAI's agentic coding tool. It runs the same agent across a terminal CLI, an IDE extension, a desktop app, and a cloud environment, so work started in one surface can continue in another. It reads the shared AGENTS.md instructions file, speaks the Model Context Protocol (MCP), and can spin up a cloud code review on a GitHub pull request that comments inline and proposes fixes. Codex runs on OpenAI's GPT models — currently GPT-5.5 as the recommended frontier model — and is included with ChatGPT plans, from the free tier up through Pro.

## Sourcegraph Cody (/tools/sourcegraph-cody)
Tagline: Code-search and AI assistant by Sourcegraph — semantic search across enterprise codebases plus inline coding help.
Category: AI Coding Agents
Pricing model: freemium
Starts at: Not recorded
Platforms: vscode-ext, jetbrains-plugin, web
Public repo: none listed
Docs: none listed
Verified plans: none on file
Strengths on file: Indexes very large monorepos for semantic, regex, and structural search.; Self-hosted Enterprise option keeps source code on customer infrastructure.; Combines code-search context with LLM answers, reducing hallucinations on internal APIs.
Limits on file: Setup and indexing for self-hosted deployment is non-trivial.; Pricing shifts from per-seat to enterprise contract at scale.
Sourcegraph Cody is a code-search and AI assistant from Sourcegraph — semantic search across large codebases paired with inline coding help.
Its differentiator is search-grounded answers: it indexes very large monorepos for semantic, regex, and structural search and feeds that context to the model, which cuts hallucinations on internal APIs. A self-hosted Enterprise option keeps source on your own infrastructure.
It fits enterprises with big codebases that want AI grounded in their own code and control over where it runs. Caveats: self-hosted setup and indexing are non-trivial, and pricing moves from per-seat to an enterprise contract at scale.
It's the code-search-first assistant, where Copilot and Cursor lead with completion and editing.

Change log

2026-08-28 — First published, compiled from the catalogue entries for Codex and Sourcegraph Cody.