Video extract frames

POST /api/video-extract-frames12 quota units per call · cache hits free

Pull several still frames out of an MP4, MOV, WebM, or MKV video — evenly spaced across the clip or at a fixed interval — and download them as a ZIP. Powered by a WebAssembly ffmpeg build.

Decodes a video and returns several frames as a ZIP of images. Two sampling modes: - `count` — that many frames spread evenly across the clip (default 9). Each frame is taken at the midpoint of its slice, so the first/last aren't exactly at 0s / the very end. - `everySeconds` — one frame every N seconds from the start. When set, it overrides `count`. Either way the result is capped at 24 frames to stay inside the Workers CPU budget. `format` (webp/png/jpeg) and `quality` apply to every frame. The ZIP holds `frame-0001.<ext>`, `frame-0002.<ext>`, … and the JSON response also lists each frame's timestamp. For a single frame at a precise time, use video-extract-frame. Accepts MP4, MOV, WebM, and MKV (same containers as video-info). `count` mode needs a video whose duration is known from its header (true for normal files); otherwise pass `everySeconds`.

Inputs

Name	Type	Default	Description
video*	file	—	MP4, MOV, WebM, or MKV file.
count	number (1…24)	9	Number of frames, spread evenly across the clip. Ignored when everySeconds is set.
everySeconds	number (0.1…3600)	—	Sample one frame every N seconds from the start (overrides count). Capped at 24 frames.
format	enum (webp \| png \| jpeg)	"webp"	Output image format for every frame.
quality	number (1…100)	80	Encoder quality for webp / jpeg (1–100). Ignored for png.

Response

Modes: json, binary. Cache: yes (24h TTL).

Every response carries x-cache (HIT / MISS / BYPASS), x-cache-signature (stable across identical inputs), and the rate-limit headers listed below.

Quota & limits

Cost per call	12 units against the 10,000-unit monthly quota. Cache hits are free — only cache misses decrement.
Burst limit	200 requests per rolling minute, shared across all tools.
Max request size	`100 MiB` (base64 inflates file payloads ~33% — the limit applies to the decoded bytes).
Max response size	`50 MiB`
Timeout	120s wall clock.
Rate-limit headers	`X-RateLimit-Limit`, `X-RateLimit-Remaining`, `X-RateLimit-Burst-Limit` on every response; `X-Quota-Warning` once 80% of the monthly quota is spent; `Retry-After` on 429s.

Errors

Errors are JSON with an error string; schema failures add a details object (Zod's flattened field errors).

Status	When	Body
400	Input fails the tool's schema (wrong type, missing required field, malformed base64) or the file bytes can't be parsed as the expected format.	{ "error": "invalid inputs", "details": { "fieldErrors": { … } } }
401	Missing or invalid X-Api-Key / X-Account-Id headers (or no session when called from the browser).	{ "error": "unauthorized" }
404	Unknown tool id, or calling /api/* on the website hostname instead of api.xtract.bot.	{ "error": "not found" }
413	Request body over the tool's max request size, or an image header declaring more megapixels than the tool's pixel budget.	{ "error": "…exceeds maxRequestBytes…" }
415	Content-Type isn't application/json on the json-body transport.	{ "error": "unsupported content-type; expected application/json" }
429	Monthly quota or the 200/min burst limit exhausted. Check Retry-After and the X-RateLimit-* headers.	{ "error": "monthly quota exceeded", "monthRemaining": 0, … }
5xx	Conversion engine failure. Safe to retry; if it persists, the input is hitting a bug — please report it.	{ "error": "…" }

Code samples

Built from the mp4-4-frames example.

# Download or substitute the example input:
#   curl -O https://xtract.bot/examples/video-extract-frames/sample.mp4
VIDEO=$(base64 -w0 < sample.mp4)

curl -X POST https://api.xtract.bot/api/video-extract-frames \
  -H "Content-Type: application/json" \
  -H "Accept: application/json" \
  -H "X-Account-Id: $XTRACT_ACCOUNT_ID" \
  -H "X-Api-Key: $XTRACT_API_KEY" \
  -d '{
  "count": 4,
  "format": "webp",
  "quality": 80,
  "video": "'"$VIDEO"'"
}'