Guides
Sume guides: the API, Formats, agents, models, media tools and pricing, each answer dated and sourced.
AI B-roll generator: make cutaway clips from a prompt
An AI B-roll generator makes cutaway clips from a text prompt or a still. Match your edit's shape, pick a length the model allows, and skip the sound.
How to make AI ASMR videos: prompt the picture and the sound
Use a video model that generates sound with the picture, and describe the close-up action and its sounds in the prompt. Models, prompts, and length.
AI lofi video generator: a looping scene plus a lo-fi track
An AI lofi video is a short looping scene under a lo-fi track. Generate the loop and the music, then repeat both for up to 30 minutes in one render.
AI Spotify Canvas generator: make a 3–8 second loop
Spotify asks for a 3–8 second, 9:16 Canvas, 720–1080 px tall. Generate a looping 9:16 clip with AI, then render it at an exact size and length.
AI birthday video from photos: animate, add music and text
Make an AI birthday video by animating a few photos into short clips, joining them over an original instrumental, and burning your message on screen.
AI motion poster generator: animate the art, keep the title
A motion poster moves the artwork while the title stays readable. Animate a text-free version of the art, then put the title back on top of the clip.
Can ChatGPT make videos? Sora's shutdown and what works now
OpenAI discontinued Sora in 2026. ChatGPT can still start a video by calling a video tool on a remote MCP server in developer mode.
Can Claude make videos? Only by calling a video tool
Claude's output is text, not video. It can still get you a clip by calling a video tool, such as Sume's hosted MCP server added as a connector.
Can Claude generate images? No, but it can call a tool
No: Anthropic says Claude understands images but cannot create or edit them. Connect an image tool over MCP, such as Sume's generate_image.
Can Claude edit videos? Only by calling editing tools
Claude takes text and images, not video, and replies in text. It can still cut, crop, or assemble a clip by calling editing tools over MCP.
How to add an MCP server to ChatGPT with developer mode
Turn on ChatGPT developer mode, create an app for the server's URL, and sign in with OAuth. The steps, with Sume's hosted MCP server as the example.
Video generation MCP server: how Sume's generate_video works
Sume's hosted MCP server has a paid generate_video tool: a prompt or image in, a job id back, then jobs_wait and jobs_result for the clip.
MCP tool annotations: readOnlyHint and how clients use them
MCP tool annotations are optional hints such as readOnlyHint. What each one means, its default, and how ChatGPT, VS Code, and Copilot use them.
Agent skills vs MCP: what each does and when to use both
An agent skill is know-how the agent reads when a task matches; an MCP tool is an action it calls on a server. How they differ, and when to use both.
Human in the loop AI agents: where a person approves
Human in the loop means an AI agent stops for a person before steps that matter. How Sume's agent does it in chat, and what replaces it in API runs.
MCP server needs authentication in Claude Code: how to fix
Claude Code flags an MCP server as needing authentication after a 401 or 403 it can't clear. How to sign in again, and when an API key fits better.
A Claude skill to generate video: what works today
Anthropic's pre-built skills cover documents, not video. A Sume CLI skill lets Claude Code make avatar videos; other clips need MCP or the API.
MCP server for OpenCode: add Sume in opencode.json
Add Sume's hosted MCP server to OpenCode as a remote entry in opencode.json, then sign in with OAuth or send an API key, and keep paid tools in check.
MCP server for Kiro: add Sume's hosted MCP in mcp.json
Add Sume's MCP server to Kiro's mcp.json, sign in through OAuth with the right scope or send an API key, and keep paid tools out of autoApprove.
MCP server for Antigravity: add Sume in mcp_config.json
Add Sume's hosted MCP server to Google Antigravity with serverUrl in mcp_config.json, sign in with OAuth or an API key, and keep paid tools on Ask.
MCP server for LM Studio: add Sume's hosted MCP
LM Studio 0.3.17 and later can use remote MCP servers. Add Sume's hosted MCP to mcp.json with an API-key header, then confirm each paid call.
GitHub Copilot coding agent MCP: add Sume's hosted server
Copilot's coding agent reads MCP servers from repository settings. Add Sume with an API key from a COPILOT_MCP_ secret and a tools allowlist.
Visual Studio MCP server: add Sume's hosted MCP in .mcp.json
Add Sume's hosted MCP server to Copilot agent mode in Visual Studio with one .mcp.json entry, sign in through CodeLens, and confirm tool calls.
MCP server for OpenClaw: add Sume's image and video tools
Save Sume's hosted MCP server in OpenClaw with openclaw mcp add, sign in with OAuth or an API-key header, and set requestTimeoutMs above 55,000.
Open WebUI MCP server: connect Sume's hosted MCP
Add Sume's hosted MCP to Open WebUI as an admin: type MCP (Streamable HTTP), then OAuth 2.1 per user or one shared Bearer key.
MCP server for image editing: restyle, cut out, upscale
Sume's hosted MCP server gives agents three image edits: prompt edits from reference images, background removal to a PNG with alpha, and upscales.
MCP prompt injection: limit what a hijacked agent can do
MCP prompt injection gets an agent to call tools nobody asked for. Bound the session instead: fewer tools, read-only by default, confirmed writes.
Zed MCP server: add Sume's hosted MCP in context_servers
Add a remote MCP server to Zed under context_servers: Sume's hosted MCP with OAuth or an API key, and tool_permissions so paid tools ask first.
Warp MCP server: add Sume's hosted MCP to Warp's agent
Add a remote MCP server to Warp: Sume's hosted MCP URL under Settings > Agents > MCP servers, OAuth or a Bearer header, and approval for its calls.
Roo Code MCP server: add Sume with type streamable-http
Add a remote MCP server to Roo Code: a streamable-http entry for Sume's hosted MCP, an API-key header, and paid tools kept out of alwaysAllow.
Amazon Q MCP server: add Sume's hosted MCP in the IDE
Amazon Q Developer in the IDE takes HTTP MCP servers. Add Sume's hosted MCP with an API-key header or OAuth, then set its paid tools to Ask.
Copilot Studio MCP server: connect Sume's hosted MCP
Connect a Copilot Studio agent to an MCP server: add Sume's hosted MCP in the onboarding wizard with an x-api-key header, then turn off unneeded tools.
goose MCP server: add Sume's hosted MCP as an extension
Add a remote MCP server to goose: Sume's hosted MCP as a Streamable HTTP extension, OAuth by client registration, and approval for paid tools.
Difference between an MCP server and an AI agent
An MCP server exposes tools; an agent is the program where a model decides which tools to call. How the work splits, with Sume's server and agent.
Text to speech in Korean: send Hangul text with language ko
For Korean text to speech, send the script in Hangul with the language set to ko. How Sume's TTS API reads Korean, checks the voice, and bills it.
Multilingual text to speech API: one request per language
Make speech in several languages by sending one TTS request per language, each with the translated script and its language code. How Sume handles it.
Text to speech for long text and audiobooks: split and join
Split long text or a book under the TTS request limits, voice every chunk with the same settings, then join the audio into files of up to 30 minutes.
Text to speech with multiple voices: one voice per line
A TTS request speaks with one voice. For a dialogue, synthesize each speaker's lines in that speaker's voice, then join the clips in script order.
AI podcast generator from text: script, voices, one file
An AI podcast generator is three steps: a speaker-tagged script, one voice per host, and a join into one audio file. How to do each with Sume's API.
IVR text to speech: phone menu prompts in telephony formats
Make IVR prompts and voicemail greetings with text to speech: one request per prompt, returned as 8 kHz μ-law or A-law WAV, or as MP3 or other PCM.
Text to speech for language learning: slow sentence audio
Make language-learning audio with text to speech: set the target language, slow the voice to 0.6×, and get one clip per sentence with word timings.
Text to speech recording for e-learning: one file per slide
Record e-learning narration with text to speech: one audio file per slide, the same voice each time, MP3 or WAV, and redo only the slide that changed.
Slideshow with voiceover: time each slide to its sentence
Make a slideshow with voiceover by holding each slide for its narration line. With Sume, TTS sentence timings become the slide slots in one render.
Text to speech for websites: one audio file per article
Add text to speech to a website: generate each article's audio once on your server, store the file, and play it on the page with an HTML audio player.
Text to speech with highlighted words: sync word timings
Highlight words as text to speech plays: get word timestamps with the audio, then mark the word whose time range holds the player's current time.
AI voiceover with my own voice: clone once, reuse in TTS
Make an AI voiceover in your own voice: clone it once from a short clip in Sume's Voices library, then use its id as the voice for text to speech.
AI voice generator with a prompt: describe the speaker
An AI voice generator with a prompt designs a new voice from a written description. In Sume, describe a person in Assets → Voices, then use it in TTS.
Text to speech API in Python: request, poll, save the MP3
Call a text to speech API from Python with Requests: POST the text and a voice, poll the job, then save the MP3. A complete script for Sume TTS 1.0.