Katto vs MCP Server
A side-by-side look at the Katto and MCP Server MCP servers: transport, auth, tools, quality signals, and the config to install either one in Claude, Cursor, Windsurf or VS Code.
Verdict: choose Katto for multimedia process work as a local process; choose MCP Server for multimedia process as a local process. You can also run both side by side in one mcpServers block.
Which MCP server should you choose?
Choose Katto when
- You need dedicated capabilities in the Multimedia Process domain.
- You prefer a local stdio subprocess transport.
- You have the keys it needs:
KATTO_API_KEY, KATTO_API_URL. - You want tools like
katto_create_clip_job, katto_get_job, katto_list_jobs.
Choose MCP Server when
- You need dedicated capabilities in the Multimedia Process domain.
- You prefer a local stdio subprocess transport.
- Its auth model fits you: API key required.
- You have the keys it needs:
WAVESPEED_API_KEY. - You want tools like
list_models, get_model_schema, run_model.
Feature & specification comparison
Rows where the two servers differ are marked.
| Spec | ||
|---|---|---|
| Category | Multimedia Process | Multimedia Process |
| Quality signal | 63/100 · Good | 67/100 · Great |
| Install command | Confirmed | Confirmed |
| Transport | Local subprocess (stdio) | Local subprocess (stdio) |
| Auth | Not declared | API key required |
| Pricing | Freemium | BYOK (pay the provider) |
| Required env vars | KATTO_API_KEYKATTO_API_URL | WAVESPEED_API_KEY |
| Verified owner | Community listing | Community listing |
| GitHub stars | 3 | 34 |
| Directory activity | 1 views 0 installs 0 upvotes | 1 views 0 installs 0 upvotes |
| Tools | 29 | 11 |
| Full listing | View Katto | View MCP Server |
Tools & capabilities
Katto tools 29
katto_create_clip_jobSubmit a long video for clipping, either by public link in url (YouTube, Twitch, Kick, Vimeo, Zoom, Google Drive, Dropbox, Loom, LinkedIn, Facebook, Rumble, Dailymotion, Medal.tv, or a direct https link to a video file; Google Drive and Dropbox links must be shared publicly) or, for a local file, by upload_id after katto_upload_video and the PUT of the file. Pass exactly one of url or upload_id. Consumes 1 video from your monthly quota — check katto_get_usage first. Clips are produced asynchronously; returns a job id, then poll katto_get_job every 30-60 s until status is 'completed', or pass webhook_url to be notified (url jobs only). Pass idempotency_key when you might retry a url job, so a retry never spends a second video (an upload can only be started once anyway). Fails if the URL is unsupported, the uploaded file is missing, or the video is longer than your plan allows.katto_get_jobRead-only. Get the status, progress and clips of a job by id. Clips come ranked by virality score (0-100, highest first; ties broken by clip_index). Each clip is { url, captions_url, title, score, clip_index, hd }: when status is 'completed' every url is a finished, downloadable clip; hd is true once the 1080p high-quality render is ready. hd=false does NOT mean broken — the url already works, the 1080p is just still rendering in the background (and stays the standard render on the free plan). The job-level hd_ready flag turns true once every clip is hd, i.e. the URLs are final and won't be swapped again (status='completed' fires earlier). Returns 404 for an unknown id.katto_list_jobsRead-only. List your recent jobs, newest first. Paginate with 'cursor' (pass the previous next_cursor); optional 'status' filter. Returns { jobs: [{id, status, source, created_at, completed_at}], next_cursor }.katto_get_clipsRead-only convenience: the clips of a job as { url, captions_url, title, score, clip_index, hd }, ranked by score (highest first). hd=true means the 1080p render is ready; hd=false means the url is the standard render (HD still finishing, or the free plan). Empty until the first clip is ready.katto_get_usageRead-only. Your current plan and monthly video quota: { plan, videos_used, videos_limit, videos_remaining }. Call before katto_create_clip_job to confirm remaining quota.katto_get_transcriptRead-only. The transcript of a completed job as timestamped segments [{ start, end, text }]. Returns 404 while the job is still processing.katto_cancel_jobCancel a still-running job (queued/processing) and refund the video back to your monthly quota. Safe to retry (idempotent). Returns an error if the job already finished, failed, or was cancelled.katto_get_accountRead-only. The account behind this key: plan, this key's scopes (read/write), and monthly quota { videos_used, videos_limit, videos_remaining }.katto_list_sourcesRead-only. The link sources Katto can clip from (YouTube, Twitch, Kick, Vimeo, Zoom, Google Drive, Dropbox, Loom, LinkedIn, Facebook, Rumble, Dailymotion, Medal.tv, or a direct https link to a video file), each with an example URL. Use it to confirm a URL is supported before calling katto_create_clip_job; for a local file use katto_upload_video instead.katto_list_clip_lengthsRead-only. The valid values for the optional config.clipLength on katto_create_clip_job (target clip-duration buckets). Note: clipLength is fixed at creation and cannot be changed by re-render.katto_list_caption_stylesRead-only. The valid caption_style preset names for katto_rerender_clip (id + label), e.g. 'hormozi' for bold word-by-word highlight. Call this before re-rendering so you pass a real preset instead of guessing.katto_rerender_clipRe-render one already-finished clip with a new reframe layout and/or caption style (get valid caption_style values from katto_list_caption_styles). Free — does NOT use video quota. Each call starts a new render (not idempotent); the original clip is kept. Returns a rerender_id; poll katto_get_rerender for the new clip url.- +17 more on the listing page
MCP Server tools 11
list_modelsSearch the live WaveSpeed model catalog (image, video, audio, 3D). Returns model IDs usable with run_model. Do not invent model IDs — always pick one returned by this tool.get_model_schemaGet a model's real input schema (required fields, properties, defaults). Call this before run_model so inputs match what the model actually accepts.run_modelRun any WaveSpeed model. input keys come from get_model_schema; unknown keys are rejected before submission. Local files: pass "@./path" string values — they are uploaded and replaced with hosted URLs (bare paths are NOT uploaded). Returns the prediction id right away (done=false); call get_prediction with that id and wait_seconds until done=true to get the output URLs. The task keeps running on the server. Pass wait_seconds here to wait for a quick model in the same call.get_priceEstimate the cost of a run before executing it (no charge). Provide the same input you would pass to run_model — pricing often depends on inputs like duration or resolution, and a quote made without them can be lower than the real charge.get_balanceShow the WaveSpeed account credit balance, with the link to top up.list_predictionsList past predictions (run history), newest window first. Defaults to the last 24 hours; pass since/until (ISO or YYYY-MM-DD) for other ranges. Use get_prediction with an id for one task in full.list_billingsList itemized charges and refunds (billing records). Requires an API key with Billing access.get_usageUsage statistics: spend, request counts, daily series and per-model breakdown. Defaults to the last 7 days. Requires an API key with Billing access.delete_predictionsDelete finished predictions from this account's run history, up to 100 at a time. They stop appearing in list_predictions and get_prediction, and this tool cannot bring them back. Runs that are still going are left alone, and only the key's own organization's records are deleted. Ask the user first, and pass confirm: true only after they agreed to delete these ids.upload_fileUpload a local file to WaveSpeed and get its hosted URL (identical bytes reuse the same upload for 24h). Usually unnecessary — run_model handles "@./path" inputs itself.get_predictionFetch the status and outputs of a past or in-flight prediction by id — use to pick up a run that returned done=false. Pass wait_seconds to keep waiting for it to finish.
Ready-to-paste client configs
Paste either block (or both) into your client config, e.g. claude_desktop_config.json or ~/.cursor/mcp.json. We only show a config when the install command is confirmed.
Katto
{
"mcpServers": {
"katto": {
"command": "npx",
"args": [
"-y",
"katto-mcp"
],
"env": {
"KATTO_API_KEY": "YOUR_KATTO_API_KEY_HERE",
"KATTO_API_URL": "YOUR_KATTO_API_URL_HERE"
}
}
}
}
MCP Server
{
"mcpServers": {
"mcp-215": {
"command": "npx",
"args": [
"-y",
"fetch"
],
"env": {
"WAVESPEED_API_KEY": "YOUR_WAVESPEED_API_KEY_HERE"
}
}
}
}
Frequently asked questions
Katto is categorized under Multimedia Process and uses a local stdio subprocess. MCP Server belongs to Multimedia Process and uses a local stdio subprocess. Pick Katto for multimedia process work and MCP Server for multimedia process.
Copy the mcpServers block from the config section on this page into your client's config (for example claude_desktop_config.json or ~/.cursor/mcp.json), then fully restart the client.
Katto is listed as freemium — a free tier with paid plans. MCP Server is bring-your-own-key: usage is billed by the API provider you supply a key for. Auth: api key required. If a server needs environment variables such as API keys, define them under the env key in your client's MCP config.
Yes. MCP clients support multiple servers: add katto and mcp-215 as separate keys in the same mcpServers object, and your assistant will call tools from either one.
On AllMCPs, Katto has a quality signal of 63/100 (Good) with 1 views, 0 installs and 3 GitHub stars. MCP Server has 67/100 (Great) with 1 views, 0 installs and 34 GitHub stars.