Media

See Using commands for how flags, JSON input, paging and confirmations work. Run adside media --help to see what your installed version has.

Recommend up to three models for a task described in plain words, with why, ready-to-use parameters, the price and the model family’s prompting guide. Beta: this command may still change.

adside media recommend --task <task> [flags]

Flag	Value
`--task` (required)	`<task>`
`--type`	`image\|video`
`--aspect-ratio`	`<aspect-ratio>`
`--duration`	`<1-30>`
`--generate-audio`	`true\|false`
`--inputs`	`'<json>'`
`--max-price-usd`	`<max 1000>`
`--priority`	`quality\|balanced\|cost`

media avatars

media avatars cancel

Cancel a talking avatar clip job; one a provider already accepted may still finish and is still charged. Beta: this command may still change.

adside media avatars cancel --job-id <id>

media avatars estimate

Price a talking avatar clip request without generating or charging. Beta: this command may still change.

adside media avatars estimate [flags]

Flag	Value	What it does
`--input-image`	`'<json>'`
`--input-audio`	`'<json>'`
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--prompt`	`<prompt>`
`--image`	`<ref>`	The still that speaks: a local file (uploaded), an https URL or an asset ID
`--character`	`<handle>`	A workspace character, instead of —image: its cover image speaks
`--audio`	`<ref>`	The line it says, an MP3 or WAV: a local file (uploaded) or an asset ID

media avatars generate

Generate one talking avatar clip, prepaying its estimate then settling measured cost; the same requestId returns the same job. Beta: this command may still change.

adside media avatars generate [flags]

Flag	Value	What it does
`--input-image`	`'<json>'`
`--input-audio`	`'<json>'`
`--wait-seconds`	`<0-30>=0`
`--request-id`	`<id>`	Submission identity (generated automatically if omitted)
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--prompt`	`<prompt>`
`--image`	`<ref>`	The still that speaks: a local file (uploaded), an https URL or an asset ID
`--character`	`<handle>`	A workspace character, instead of —image: its cover image speaks
`--audio`	`<ref>`	The line it says, an MP3 or WAV: a local file (uploaded) or an asset ID
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media avatars get

Read a talking avatar clip job, waiting up to waitSeconds for it to finish. Beta: this command may still change.

adside media avatars get --job-id <id> [flags]

Flag	Value	What it does
`--job-id` (required)	`<id>`
`--wait-seconds`	`<0-30>=0`
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media avatars models

List the talking avatar clip models, what each is best for, its supported values and prices. Beta: this command may still change.

adside media avatars models

media images

media images cancel

Cancel an image job; one a provider already accepted may still finish and is still charged. Beta: this command may still change.

adside media images cancel --job-id <id>

media images estimate

Price an image request without generating or charging. Beta: this command may still change.

adside media images estimate --prompt <prompt> [flags]

Flag	Value	What it does
`--prompt` (required)	`<prompt>`
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--resolution`	`512\|768\|1K\|1.5K\|2K\|4K`
`--aspect-ratio`	`auto\|1:1\|2:3\|3:2\|3:4\|4:3\|4:5\|5:4\|9:16\|16:9\|21:9\|9:21\|1:2\|2:1\|1:4\|4:1\|1:8\|8:1`
`--input-references`	`'<json>'`
`--input-reference` (repeatable)	`<ref>`	A reference image: a local file (uploaded), an https URL or an asset ID (repeatable)
`--character` (repeatable)	`<handle>`	A workspace character: its reference images, best first, as many as the model takes (repeatable)

media images generate

Generate one image, prepaying its estimate then settling measured cost; the same requestId returns the same job. Beta: this command may still change.

adside media images generate --prompt <prompt> [flags]

Flag	Value	What it does
`--prompt` (required)	`<prompt>`
`--wait-seconds`	`<0-30>=0`
`--request-id`	`<id>`	Submission identity (generated automatically if omitted)
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--resolution`	`512\|768\|1K\|1.5K\|2K\|4K`
`--aspect-ratio`	`auto\|1:1\|2:3\|3:2\|3:4\|4:3\|4:5\|5:4\|9:16\|16:9\|21:9\|9:21\|1:2\|2:1\|1:4\|4:1\|1:8\|8:1`
`--input-references`	`'<json>'`
`--input-reference` (repeatable)	`<ref>`	A reference image: a local file (uploaded), an https URL or an asset ID (repeatable)
`--character` (repeatable)	`<handle>`	A workspace character: its reference images, best first, as many as the model takes (repeatable)
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media images get

Read an image job, waiting up to waitSeconds for it to finish. Beta: this command may still change.

adside media images get --job-id <id> [flags]

Flag	Value	What it does
`--job-id` (required)	`<id>`
`--wait-seconds`	`<0-30>=0`
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media images models

List the image models, what each is best for, its supported values and prices. Beta: this command may still change.

adside media images models

media jobs

media jobs list

List the workspace’s image, video, audio, avatar and lip-sync jobs, newest first. Beta: this command may still change.

adside media jobs list [flags]

Flag	Value	What it does
`--type`	`image\|video\|speech\|sound\|music\|avatar\|lipsync`
`--status`	`pending\|in_progress\|completed\|failed\|cancelled\|expired`
`--cursor`	`<next>`	The next page, from the previous output.
`--limit`	`<1-50>=20`

media lipsync

media lipsync cancel

Cancel a lip-sync clip job; one a provider already accepted may still finish and is still charged. Beta: this command may still change.

adside media lipsync cancel --job-id <id>

media lipsync estimate

Price a lip-sync clip request without generating or charging. Beta: this command may still change.

adside media lipsync estimate [flags]

Flag	Value	What it does
`--input-video`	`'<json>'`
`--input-audio`	`'<json>'`
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--video`	`<ref>`	The MP4 clip to re-voice: a local file (uploaded) or an asset ID
`--audio`	`<ref>`	The line it says, an MP3 or WAV: a local file (uploaded) or an asset ID

media lipsync generate

Generate one lip-sync clip, prepaying its estimate then settling measured cost; the same requestId returns the same job. Beta: this command may still change.

adside media lipsync generate [flags]

Flag	Value	What it does
`--input-video`	`'<json>'`
`--input-audio`	`'<json>'`
`--wait-seconds`	`<0-30>=0`
`--request-id`	`<id>`	Submission identity (generated automatically if omitted)
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--video`	`<ref>`	The MP4 clip to re-voice: a local file (uploaded) or an asset ID
`--audio`	`<ref>`	The line it says, an MP3 or WAV: a local file (uploaded) or an asset ID
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media lipsync get

Read a lip-sync clip job, waiting up to waitSeconds for it to finish. Beta: this command may still change.

adside media lipsync get --job-id <id> [flags]

Flag	Value	What it does
`--job-id` (required)	`<id>`
`--wait-seconds`	`<0-30>=0`
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media lipsync models

List the lip-sync clip models, what each is best for, its supported values and prices. Beta: this command may still change.

adside media lipsync models

media music

media music cancel

Cancel a music track job; one a provider already accepted may still finish and is still charged. Beta: this command may still change.

adside media music cancel --job-id <id>

media music estimate

Price a music track request without generating or charging. Beta: this command may still change.

adside media music estimate --prompt <prompt> [flags]

Flag	Value
`--prompt` (required)	`<prompt>`
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--duration`	`<3-600>`
`--instrumental`	`true\|false`

media music generate

Generate one music track, prepaying its estimate then settling measured cost; the same requestId returns the same job. Beta: this command may still change.

adside media music generate --prompt <prompt> [flags]

Flag	Value	What it does
`--prompt` (required)	`<prompt>`
`--wait-seconds`	`<0-30>=0`
`--request-id`	`<id>`	Submission identity (generated automatically if omitted)
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--duration`	`<3-600>`
`--instrumental`	`true\|false`
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media music get

Read a music track job, waiting up to waitSeconds for it to finish. Beta: this command may still change.

adside media music get --job-id <id> [flags]

Flag	Value	What it does
`--job-id` (required)	`<id>`
`--wait-seconds`	`<0-30>=0`
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media music models

List the music track models, what each is best for, its supported values and prices. Beta: this command may still change.

adside media music models

media sounds

media sounds cancel

Cancel a sound effect job; one a provider already accepted may still finish and is still charged. Beta: this command may still change.

adside media sounds cancel --job-id <id>

media sounds estimate

Price a sound effect request without generating or charging. Beta: this command may still change.

adside media sounds estimate --prompt <prompt> [flags]

Flag	Value
`--prompt` (required)	`<prompt>`
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--duration`	`<1-30>`
`--loop`	`true\|false`

media sounds generate

Generate one sound effect, prepaying its estimate then settling measured cost; the same requestId returns the same job. Beta: this command may still change.

adside media sounds generate --prompt <prompt> [flags]

Flag	Value	What it does
`--prompt` (required)	`<prompt>`
`--wait-seconds`	`<0-30>=0`
`--request-id`	`<id>`	Submission identity (generated automatically if omitted)
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--duration`	`<1-30>`
`--loop`	`true\|false`
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media sounds get

Read a sound effect job, waiting up to waitSeconds for it to finish. Beta: this command may still change.

adside media sounds get --job-id <id> [flags]

Flag	Value	What it does
`--job-id` (required)	`<id>`
`--wait-seconds`	`<0-30>=0`
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media sounds models

List the sound effect models, what each is best for, its supported values and prices. Beta: this command may still change.

adside media sounds models

media speech

media speech cancel

Cancel a speech clip job; one a provider already accepted may still finish and is still charged. Beta: this command may still change.

adside media speech cancel --job-id <id>

media speech estimate

Price a speech clip request without generating or charging. Beta: this command may still change.

adside media speech estimate --text <text> [flags]

Flag	Value
`--text` (required)	`<text>`
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--voice`	`<voice>`
`--language-code`	`<language-code>`
`--stability`	`<0-1>`
`--speed`	`<0.7-1.2>`

media speech generate

Generate one speech clip, prepaying its estimate then settling measured cost; the same requestId returns the same job. Beta: this command may still change.

adside media speech generate --text <text> [flags]

Flag	Value	What it does
`--text` (required)	`<text>`
`--wait-seconds`	`<0-30>=0`
`--request-id`	`<id>`	Submission identity (generated automatically if omitted)
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--voice`	`<voice>`
`--language-code`	`<language-code>`
`--stability`	`<0-1>`
`--speed`	`<0.7-1.2>`
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media speech get

Read a speech clip job, waiting up to waitSeconds for it to finish. Beta: this command may still change.

adside media speech get --job-id <id> [flags]

Flag	Value	What it does
`--job-id` (required)	`<id>`
`--wait-seconds`	`<0-30>=0`
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media speech models

List the speech clip models, what each is best for, its supported values and prices. Beta: this command may still change.

adside media speech models

media videos

media videos cancel

Cancel a video job; one a provider already accepted may still finish and is still charged. Beta: this command may still change.

adside media videos cancel --job-id <id>

media videos estimate

Price a video request without generating or charging. Beta: this command may still change.

adside media videos estimate --prompt <prompt> [flags]

Flag	Value	What it does
`--prompt` (required)	`<prompt>`
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--duration`	`<1-30>`
`--resolution`	`480p\|720p\|768p\|1080p\|1K\|2K\|4K`
`--aspect-ratio`	`9:16\|16:9\|1:1\|4:3\|3:4\|3:2\|2:3\|21:9\|9:21`
`--frame-images`	`'<json>'`
`--generate-audio`	`true\|false`
`--first-frame`	`<ref>`	The clip’s first frame: a local file (uploaded), an https URL or an asset ID
`--last-frame`	`<ref>`	The clip’s last frame, with —first-frame, for a model that takes one: a local file (uploaded), an https URL or an asset ID

media videos generate

Generate one video clip, prepaying its estimate then settling measured cost; the same requestId returns the same job. Beta: this command may still change.

adside media videos generate --prompt <prompt> [flags]

Flag	Value	What it does
`--prompt` (required)	`<prompt>`
`--wait-seconds`	`<0-30>=0`
`--request-id`	`<id>`	Submission identity (generated automatically if omitted)
`--model`	`<model>`
`--max-price-usd`	`<max 1000>`
`--duration`	`<1-30>`
`--resolution`	`480p\|720p\|768p\|1080p\|1K\|2K\|4K`
`--aspect-ratio`	`9:16\|16:9\|1:1\|4:3\|3:4\|3:2\|2:3\|21:9\|9:21`
`--frame-images`	`'<json>'`
`--generate-audio`	`true\|false`
`--first-frame`	`<ref>`	The clip’s first frame: a local file (uploaded), an https URL or an asset ID
`--last-frame`	`<ref>`	The clip’s last frame, with —first-frame, for a model that takes one: a local file (uploaded), an https URL or an asset ID
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media videos get

Read a video job, waiting up to waitSeconds for it to finish. Beta: this command may still change.

adside media videos get --job-id <id> [flags]

Flag	Value	What it does
`--job-id` (required)	`<id>`
`--wait-seconds`	`<0-30>=0`
`--wait`		Wait for the job to finish, 30 s per request, up to —wait-timeout
`--wait-timeout`	`<duration>`	How long —wait may take: 90s, 14m or seconds (default 15m)
`--out`	`<file>`	Download the finished output to a new file (never overwrites one)

media videos models

List the video models, what each is best for, its supported values and prices. Beta: this command may still change.

adside media videos models

media voices

media voices list

List the stock voices a speech request may name, with what each suits and the models that offer it. Beta: this command may still change.

adside media voices list [flags]

Flag	Value
`--model`	`<model>`

​media

​media recommend

​media avatars

​media avatars cancel

​media avatars estimate

​media avatars generate

​media avatars get

​media avatars models

​media images

​media images cancel

​media images estimate

​media images generate

​media images get

​media images models

​media jobs

​media jobs list

​media lipsync

​media lipsync cancel

​media lipsync estimate

​media lipsync generate

​media lipsync get

​media lipsync models

​media music

​media music cancel

​media music estimate

​media music generate

​media music get

​media music models

​media sounds

​media sounds cancel

​media sounds estimate

​media sounds generate

​media sounds get

​media sounds models

​media speech

​media speech cancel

​media speech estimate

​media speech generate

​media speech get

​media speech models

​media videos

​media videos cancel

​media videos estimate

​media videos generate

​media videos get

​media videos models

​media voices

​media voices list

media

media recommend