Can Codex make videos? No, but it can call a video tool
No: Codex's docs list image generation but not video. Add a video tool over MCP, such as Sume's hosted server, and Codex can start clips for you.

No: Codex does not make videos on its own. OpenAI's Codex docs list image generation among its capabilities, but no video generation. Codex can still get you a clip by calling a video tool on an MCP server, such as generate_video on Sume's hosted server.
Codex facts come from OpenAI's Codex image generation, Codex CLI and Codex MCP pages; Sume's side comes from its MCP overview and MCP quickstart. All were read on 2026-09-28. Sume has no official Codex plugin: this is a plain remote MCP connection, and the basics page says hosted MCP still works but is not the primary path today.
What can Codex make without a video tool?
The Codex docs' Capabilities list covers Browser, Computer use, Voice, Plugins, Web search, Image generation, Image inputs, Appshots, Browser extension and Work with files. Here is how that maps to media work:
| Task | Codex on its own | With Sume's hosted MCP |
|---|---|---|
| Still image | Built in, uses gpt-image-2; $imagegen invokes it | generate_image |
| New video clip | Not in the capabilities list | generate_video |
| Music track | Not in the capabilities list | music_create |
| Edit existing video | Runs tools already installed on your machine, on local files | video_trim, timeline_create, on Sume-hosted media only |
How do I add a video tool to Codex?
Codex keeps MCP servers in ~/.codex/config.toml, shared by the Codex CLI, the IDE extension and the ChatGPT desktop app. Add Sume as a streamable HTTP server, then run Codex's OAuth login for it:
Sume's consent page opens in the browser. The default grant is read-only (mcp:read), and paid tools such as generate_video return insufficient_scope until you turn Write on. In the Codex TUI, /mcp lists the active servers. Sume MCP server for Claude Code, Cursor and Codex covers the other clients.
# ~/.codex/config.toml
[mcp_servers.sume]
url = "https://mcp.sume.com/mcp"
# then, in a terminal:
codex mcp login sumeWill Codex time out while the video renders?
Codex's tool_timeout_sec defaults to 60 seconds per tool call. On Sume, generate_video answers with a job id, not a clip, and one jobs_wait call holds at most 55 seconds, so each wait fits inside the default. On wait_slice_expired, the agent should call jobs_wait again with the same id and never resubmit the paid create. Codex MCP tool timeout covers raising the limit.
Can Codex generate music?
Not on its own: music generation is not in the Codex capabilities list either. With Sume's server connected, music_create turns a prompt into a track; Lyria MCP covers the call.
What does it cost?
Codex's page says built-in image generation counts toward your general Codex usage limits. Sume's tools bill the Sume workspace instead: video is reserved on submit at provider list × 1.25, plus a 5.5% agent fee by default, and a music generation is listed at $0.125 per audio on API pricing. Paid calls need an idempotency_key; ask Codex to send dry_run: true first to preview the cost without submitting.
Sources
Related posts
More in Agents
- Can Cursor generate videos? Images yes, video via MCP
Cursor's Agent has a built-in image generation tool but no video tool. To make video from Cursor, connect an MCP server that has one.
- ChatGPT text to speech: make an audio file with MCP
ChatGPT Voice speaks its replies. To get your own script as an audio file, add a text-to-speech tool to ChatGPT over MCP in developer mode.
- HeyGen Video Agent: what it does, its API, and pricing
HeyGen's Video Agent turns a text prompt into a finished avatar video, in the app or via POST /v3/video-agents. Its inputs, outputs, and API price.
- How to write a video brief: parts, template, agent tips
A video brief states the deliverable, audience, one message, must-show items, tone, and what done means. A template, and how an AI agent reads it.
Written by Sume