Media API — Images, Video, Music

Generate images, video, and music using the same sk-nr-… key and balance as text models. One account, pay as you go, no subscriptions or packages.

• Base URL: https://nordrouter.com (or https://global-api.nordrouter.com for networks outside Russia).
• Authentication: Authorization: Bearer sk-nr-… header (or X-API-Key: sk-nr-…).
• Format: JSON, asynchronous generation (create a job → poll → retrieve the file).

Endpoints

Method	Purpose
POST /media/generate	Create a generation job
GET /media/job/:id	Get job status and result
POST /media/estimate	Get the price for a configuration in advance
GET /media/models	Get a machine-readable list of models and parameters
POST /media/upload	Upload a source media file
GET /media/file/:id	Retrieve the generated file (link from result_url)

Request Lifecycle

1. Create a job

curl https://nordrouter.com/media/generate \
  -H "Authorization: Bearer sk-nr-YOUR-KEY" -H "Content-Type: application/json" \
  -d '{ "model": "video/kling-2.6-i2v",
        "input": { "prompt": "smooth camera push-in", "image": "https://…/photo.png", "duration": "5", "audio": true } }'

{ "id": "b1f2…", "model": "video/kling-2.6-i2v", "status": "processing", "poll": "https://nordrouter.com/media/job/b1f2…" }

2. Poll for status (every 3–5 seconds)

curl https://nordrouter.com/media/job/b1f2…

{ "id": "b1f2…", "status": "done", "result_url": "https://nordrouter.com/media/file/b1f2…", "cost_usd": 0.55, "error": null }

3. Retrieve the file — via result_url. The link is on nordrouter.com (your brand); files are stored for 14 days, then deleted — download them or copy them to your own storage in advance.

Uploading source media files

The image, image2, audio_url, and video_url fields accept a direct public URL. Models with an attachments field accept an array of objects returned by POST /media/upload: send { "data": "data:audio/mpeg;base64,…", "name": "track.mp3" }, then pass the entire upload response in input.attachments. For Suno, you can instead pass a public URL in input.source_url, or two URLs in input.source_urls for mashup.

Billing

Charges are deducted from your USD balance (the same balance used for text models) — after generation completes, for exactly the amount listed in the tables below.

• Reserve → actual charge. The estimated cost is reserved when a job is created; after completion, the amount is recalculated based on the actual result, and any excess is returned to your balance.
• Video — per second. The price increases with duration and resolution; enabling audio costs more. The tables list starting prices for default settings; /sec indicates a per-second rate.
• No charge for failures. If a job ends with status failed, the reserve is returned in full — you only pay for successful generations.
• Exact price in advance — POST /media/estimate with the same model + input returns { "usd": … } before generation starts.
• Top up your balance in rubles (≈ $1 = 89–96 ₽); all subsequent calculations are in USD.

Model catalog

Pricing: for video — rate per second (/sec), total = rate × duration (resolution/audio may also affect the rate); for everything else — per generation with default settings. POST /media/estimate returns the exact cost of any configuration.

There may be more fields in input — GET /media/models always returns a machine-readable list (types, options, defaults, duration ranges in seconds). Required fields are marked with *. Any additional model fields (negative_prompt, seed, cfg_scale…) can be sent directly in input.

Legacy parameter names

On October 19, 2026, parameters were standardized: snake_case names and lowercase values (for example, version: "v6-wild", resolution: "720p", audio: true). Names and values from the previous documentation are still accepted — existing integrations need no changes: the task, result, and price remain the same. GET /media/models, MCP, and this documentation use only the new names and values. If a request contains both the legacy and new name for the same parameter, the new name takes precedence.

🖼 Images

model	Mode	input parameters	File	Price
image/nano-banana	text→image	prompt *, aspect_ratio=1:1|16:9|9:16|3:4|4:3|3:2|2:3|21:9|auto, safety_filter=bool	PNG	$0.025
image/nano-banana-2	text→image	prompt *, aspect_ratio=auto|1:1|16:9|9:16|3:4|4:3|21:9, resolution=1k|2k|4k, image=url, safety_filter=bool	PNG	$0.050
image/nano-banana-2-lite	text→image	prompt *, aspect_ratio=auto|1:1|16:9|9:16|3:4|4:3, image=url, safety_filter=bool	PNG	$0.025
image/nano-banana-pro	text→image	prompt *, aspect_ratio=1:1|16:9|9:16|3:4|4:3|21:9|auto, resolution=1k|2k|4k, image=url, safety_filter=bool	PNG	$0.11
image/imagen4-fast	text→image	prompt *, aspect_ratio=1:1|16:9|9:16|3:4|4:3, safety_filter=bool	PNG	$0.025
image/imagen4	text→image	prompt *, aspect_ratio=1:1|16:9|9:16|3:4|4:3, safety_filter=bool	PNG	$0.050
image/imagen4-ultra	text→image	prompt *, aspect_ratio=1:1|16:9|9:16|3:4|4:3, safety_filter=bool	PNG	$0.075
image/gpt-image-1.5	text→image	prompt *, aspect_ratio=1:1|2:3|3:2, quality=medium|high, safety_filter=bool	PNG	$0.025
image/gpt-image-1.5-edit	image editing	prompt *, image=url *, aspect_ratio=3:2|1:1|2:3, quality=medium|high, safety_filter=bool	PNG	$0.025
image/gpt-image-2	text→image	prompt *, aspect_ratio=1:1|3:2|2:3|16:9|9:16|4:3|3:4, resolution=1k|2k|4k	PNG	$0.006
image/gpt-image-2-edit	image editing	prompt *, image=url *, aspect_ratio=auto|1:1|3:2|2:3|16:9|9:16, resolution=1k|2k|4k, safety_filter=bool	PNG	$0.037
image/flux2-pro	text→image	prompt *, aspect_ratio=1:1|4:3|3:4|16:9|9:16|3:2|2:3, resolution=1k|2k, safety_filter=bool	PNG	$0.031
image/flux2-pro-edit	image editing	prompt *, image=url *, aspect_ratio=1:1|4:3|3:4|16:9|9:16|3:2|2:3|auto, resolution=1k|2k, safety_filter=bool	PNG	$0.031
image/flux2-flex	text→image	prompt *, aspect_ratio=1:1|4:3|3:4|16:9|9:16|3:2|2:3, resolution=1k|2k, safety_filter=bool	PNG	$0.087
image/flux2-flex-edit	image editing	prompt *, image=url *, aspect_ratio=1:1|4:3|3:4|16:9|9:16|3:2|2:3|auto, resolution=1k|2k, safety_filter=bool	PNG	$0.087
image/seedream-edit	image editing	prompt *, image=url *, aspect_ratio=1:1-hd|1:1|3:4|9:16|4:3|16:9, resolution=1k|2k|4k, safety_filter=bool	PNG	$0.037
image/seedream-4.5-edit	image editing	prompt *, image=url *, aspect_ratio=1:1|4:3|3:4|16:9|9:16|2:3|3:2, quality=basic|high, safety_filter=bool	PNG	$0.044
image/seedream-5-lite-edit	image editing	prompt *, image=url *, aspect_ratio=1:1|4:3|3:4|16:9|9:16|2:3|3:2, quality=basic|high, safety_filter=bool	PNG	$0.037
image/seedream-5.0-pro-edit	image editing	prompt *, image=url *, aspect_ratio=1:1|4:3|3:4|16:9|9:16|2:3|3:2|21:9, quality=basic|high, safety_filter=bool	PNG	$0.044
image/ideogram-v3	text→image	prompt *, render_mode=fast|balanced|quality, aspect_ratio=1:1-hd|1:1|3:4|9:16|4:3|16:9, style=auto|general|realistic|design, safety_filter=bool	PNG	$0.044
image/ideogram-edit	image editing	prompt *, image=url *, render_mode=fast|balanced|quality, aspect_ratio=1:1-hd|1:1|3:4|9:16|4:3|16:9, safety_filter=bool	PNG	$0.044
image/ideogram-character	image editing	prompt *, image=url *, render_mode=fast|balanced|quality, style=auto|realistic|fiction, safety_filter=bool	PNG	$0.11
image/qwen	text→image	prompt *, aspect_ratio=1:1-hd|1:1|3:4|9:16|4:3|16:9, safety_filter=bool	PNG	$0.031
image/qwen-edit	image editing	prompt *, image=url *, aspect_ratio=1:1-hd|1:1|3:4|9:16|4:3|16:9, safety_filter=bool	PNG	$0.037
image/qwen-i2i	image editing	prompt *, image=url *, safety_filter=bool	PNG	$0.031
image/qwen2-edit	image editing	prompt *, image=url *, aspect_ratio=16:9|1:1|3:4|4:3|9:16|3:2|2:3, safety_filter=bool	PNG	$0.037
image/grok	text→image	prompt *, aspect_ratio=1:1|2:3|3:2|16:9|9:16, pro=bool, safety_filter=bool	PNG	$0.025
image/grok-edit	image editing	prompt *, image=url *, safety_filter=bool	PNG	$0.025
image/z-image	text→image	prompt *, aspect_ratio=1:1|16:9|9:16|3:4|4:3, safety_filter=bool	PNG	$0.006
image/wan-2.7	text→image	prompt *, aspect_ratio=1:1|16:9|9:16|4:3|3:4|21:9, resolution=1k|2k|4k, image=url, safety_filter=bool	PNG	$0.031
image/wan-2.7-pro	text→image	prompt *, aspect_ratio=1:1|16:9|9:16|4:3|3:4|21:9, resolution=1k|2k|4k, image=url, safety_filter=bool	PNG	$0.075
image/seedream-5.0-pro	text→image	prompt *	JPG	$0.013
image/gemini-3.1-flash	text→image	prompt *, aspect_ratio=1:1|16:9|9:16|4:3|3:4, resolution=1k|2k|4k	PNG	$0.013
image/gemini-3-pro	text→image	prompt *, aspect_ratio=1:1|16:9|9:16|4:3|3:4, resolution=1k|2k|4k	PNG	$0.031

🎬 Video

model	Mode	input parameters	File	Price
video/veo-3.1-fast	text→video	prompt *, aspect_ratio=16:9|9:16, image=url	MP4	$0.38
video/veo-3.1-quality	text→video	prompt *, aspect_ratio=16:9|9:16, image=url	MP4	$1.56
video/veo-3.1-lite	text→video	prompt *, aspect_ratio=16:9|9:16, image=url	MP4	$0.19
video/gemini-omni	text→video	prompt *, duration=4|6|8|10 *, aspect_ratio=16:9|9:16, resolution=720p|1080p|4k, image=url, safety_filter=bool	MP4	$0.66
video/gemini-omni-flash	text→video	prompt *, attachments=upload[]	MP4	$0.13
video/kling-3.0	text→video	prompt *, duration=3–15s, aspect_ratio=16:9|9:16|1:1, mode=standard|pro|4k, audio=bool, image=url, image2=url, safety_filter=bool	MP4	$0.11/sec
video/kling-t2v	text→video	prompt *, duration=3–15s, aspect_ratio=16:9|9:16|1:1, resolution=720p|1080p, safety_filter=bool	MP4	$0.11/sec
video/kling-turbo-i2v	photo→video	prompt *, image=url *, duration=3–15s, resolution=720p|1080p, safety_filter=bool	MP4	$0.11/sec
video/kling-2.6-t2v	text→video	prompt *, duration=5|10, aspect_ratio=16:9|9:16|1:1, audio=bool, safety_filter=bool	MP4	$0.34
video/kling-2.6-i2v	photo→video	prompt *, image=url *, duration=5|10, audio=bool, safety_filter=bool	MP4	$0.34
video/kling-2.5-t2v	text→video	prompt *, duration=5|10, aspect_ratio=16:9|9:16|1:1, safety_filter=bool	MP4	$0.26
video/kling-2.5-i2v	photo→video	prompt *, image=url *, image2=url, duration=5|10, safety_filter=bool	MP4	$0.26
video/kling-i2v	photo→video	prompt *, image=url *, image2=url, duration=5|10, safety_filter=bool	MP4	$0.31
video/kling-2.1-standard	photo→video	prompt *, image=url *, duration=5|10, safety_filter=bool	MP4	$0.16
video/kling-2.1-master-t2v	text→video	prompt *, duration=5|10, aspect_ratio=16:9|9:16|1:1, safety_filter=bool	MP4	$1.00
video/kling-2.1-master-i2v	photo→video	prompt *, image=url *, duration=5|10, safety_filter=bool	MP4	$1.00
video/kling-motion	video→video	image=url *, video_url *, prompt, mode=standard|pro	MP4	$1.25
video/kling-avatar	avatar (photo+audio)	image=url *, audio_url *, prompt *	MP4	$0.75
video/kling-avatar-pro	avatar (photo+audio)	image=url *, audio_url *, prompt *	MP4	$1.50
video/wan-2.7-t2v	text→video	prompt *, duration=2–15s, aspect_ratio=16:9|9:16|1:1|4:3|3:4, resolution=720p|1080p, audio_url, safety_filter=bool	MP4	$0.15/sec
video/wan-2.7-i2v	photo→video	prompt *, image=url *, image2=url, duration=2–15s, resolution=720p|1080p, audio_url, safety_filter=bool	MP4	$0.15/sec
video/wan-2.7-r2v	photo→video	prompt *, image=url *, duration=2–10s, aspect_ratio=16:9|9:16|1:1, resolution=720p|1080p, safety_filter=bool	MP4	$0.15/sec
video/wan-2.7-edit	video→video	prompt *, video_url *, resolution=720p|1080p, audio_source=auto|original	MP4	$1.50
video/wan-t2v	text→video	prompt *, duration=5|10|15, resolution=720p|1080p, safety_filter=bool	MP4	$0.44
video/wan-i2v	photo→video	prompt *, image=url *, duration=5|10|15, resolution=720p|1080p, safety_filter=bool	MP4	$0.44
video/wan-v2v	video→video	prompt *, video_url *, duration=5|10, resolution=720p|1080p	MP4	$0.44
video/wan-2.5-t2v	text→video	prompt *, duration=5|10 *, aspect_ratio=16:9|9:16|1:1, resolution=720p|1080p, safety_filter=bool	MP4	$0.38
video/wan-2.5-i2v	photo→video	prompt *, image=url *, duration=5|10 *, resolution=720p|1080p, safety_filter=bool	MP4	$0.38
video/wan-2.2-t2v	text→video	prompt *, aspect_ratio=16:9|9:16, resolution=480p|720p, safety_filter=bool	MP4	$0.50
video/wan-2.2-i2v	photo→video	prompt *, image=url *, resolution=480p|720p, safety_filter=bool	MP4	$0.50
video/wan-s2v	avatar (photo+audio)	prompt *, image=url *, audio_url *, resolution=480p|580p|720p	MP4	$0.38
video/wan-animate-move	video→video	image=url *, video_url *, resolution=480p|580p|720p	MP4	$0.38
video/wan-animate-replace	video→video	image=url *, video_url *, resolution=480p|580p|720p	MP4	$0.38
video/seedance-2	text→video	prompt *, duration=4–15s, aspect_ratio=16:9|9:16|1:1|4:3|21:9, resolution=480p|720p|1080p|4k, audio=bool, image=url, image2=url, safety_filter=bool	MP4	$0.26/sec
video/seedance-2-fast	text→video	prompt *, duration=4–15s, aspect_ratio=16:9|9:16|1:1, resolution=480p|720p, audio=bool, image=url, image2=url, safety_filter=bool	MP4	$0.21/sec
video/seedance-2-mini	text→video	prompt *, duration=4–15s, aspect_ratio=16:9|9:16|1:1, resolution=480p|720p, audio=bool, image=url, image2=url, safety_filter=bool	MP4	$0.13/sec
video/seedance-1.5-pro	text→video	prompt *, duration=4–12s, aspect_ratio=16:9|9:16|1:1|4:3, resolution=480p|720p|1080p, audio=bool, image=url, safety_filter=bool	MP4	$0.025/sec
video/hailuo-02-t2v	text→video	prompt *, duration=6|10, safety_filter=bool	MP4	$0.19
video/hailuo-02-t2v-pro	text→video	prompt *, safety_filter=bool	MP4	$0.36
video/hailuo-02-i2v	photo→video	prompt *, image=url *, image2=url, duration=6|10, resolution=768p|512p, safety_filter=bool	MP4	$0.19
video/hailuo-02-i2v-pro	photo→video	prompt *, image=url *, image2=url, safety_filter=bool	MP4	$0.36
video/hailuo-2.3-i2v	photo→video	prompt *, image=url *, duration=6|10, resolution=768p|1080p, safety_filter=bool	MP4	$0.19
video/hailuo-2.3-i2v-pro	photo→video	prompt *, image=url *, duration=6|10, resolution=768p|1080p, safety_filter=bool	MP4	$0.28
video/happyhorse-1.1-t2v	text→video	prompt *, duration=3–15s, aspect_ratio=16:9|9:16|1:1|4:3|3:4, resolution=720p|1080p, safety_filter=bool	MP4	$0.18/sec
video/happyhorse-1.1-i2v	photo→video	prompt, image=url *, duration=3–15s, resolution=720p|1080p, safety_filter=bool	MP4	$0.18/sec
video/happyhorse-1.1-r2v	photo→video	prompt *, image=url *, duration=3–15s, aspect_ratio=16:9|9:16|1:1, resolution=720p|1080p, safety_filter=bool	MP4	$0.18/sec
video/happyhorse-t2v	text→video	prompt *, duration=3–15s, aspect_ratio=16:9|9:16|1:1|4:3|3:4, resolution=720p|1080p, safety_filter=bool	MP4	$0.30/sec
video/happyhorse-i2v	photo→video	prompt, image=url *, duration=3–15s, resolution=720p|1080p, safety_filter=bool	MP4	$0.30/sec
video/happyhorse-edit	video→video	prompt *, video_url *, resolution=720p|1080p, audio_source=auto|original	MP4	$1.50
video/grok-t2v	text→video	prompt *, duration=6–30s, aspect_ratio=16:9|9:16|1:1|3:2|2:3, resolution=480p|720p, mode=normal|fun|spicy, safety_filter=bool	MP4	$0.019/sec
video/grok-i2v	photo→video	prompt *, image=url *, duration=6–30s, resolution=480p|720p, mode=normal|fun|spicy, safety_filter=bool	MP4	$0.019/sec
video/grok-1.5	text→video	prompt *, duration=1–15s, aspect_ratio=auto|16:9|9:16|1:1, resolution=480p|720p, image=url, safety_filter=bool	MP4	$0.019/sec
video/omnihuman	avatar (photo+audio)	image=url *, audio_url *, prompt, resolution=720p|1080p	MP4	$1.69
video/infinitalk	avatar (photo+audio)	prompt *, image=url *, audio_url *, resolution=480p|720p	MP4	$0.19
video/lipsync	video→video	video_url *, audio_url *, mode=lite|basic	MP4	$0.50
video/topaz-upscale	upscaling	video_url *, scale=2|4	MP4	$0.50

🎵 Music

model	Mode	input parameters	File	Price
music/suno-v5	music	title *, style *, version=v6|v6-wild|v6-mini, instrumental=bool, prompt	MP3	$0.075
music/suno-cover	music	prompt *, attachments=upload[] *, version=v6|v6-wild|v6-mini, instrumental=bool, advanced=bool, title, style, duration=10–360s	MP3	$0.075
music/suno-extend-upload	music	prompt *, attachments=upload[] *, version=v6|v6-wild|v6-mini, continue_from=1–120, instrumental=bool	MP3	$0.075
music/suno-add-vocals	music	prompt *, attachments=upload[] *, version=v6|v6-wild|v6-mini, style, title, exclude_styles, vocal=female|male	MP3	$0.075
music/suno-add-instrumental	music	prompt *, attachments=upload[] *, version=v6|v6-wild|v6-mini, title, exclude_styles	MP3	$0.075
music/suno-mashup	music	prompt *, attachments=upload[] *, version=v6|v6-wild|v6-mini, instrumental=bool, advanced=bool, title, style, duration=10–360s	MP3	$0.075
music/suno-sounds	music	prompt *, version=v6|v6-wild|v6-mini, loop=bool, bpm=auto|60|80|100|120|140|160, tonality=auto|C|Dm|Em|F|G|Am|D#m	MP3	$0.016

🎙 Audio

model	Mode	input parameters	File	Price
audio/gemini-3.1-flash-tts	text-to-speech	text *, voice=Achernar|Achird|Algenib|Algieba|Alnilam|Aoede|Autonoe|Callirrhoe|Charon|Despina|Enceladus|Erinome|Fenrir|Gacrux|Iapetus|Kore|Laomedeia|Leda|Orus|Puck|Pulcherrima|Rasalgethi|Sadachbia|Sadaltager|Schedar|Sulafat|Umbriel|Vindemiatrix|Zephyr|Zubenelgenubi, style=vocal-smile|newscaster|whisper|empathetic|promo-hype|deadpan, pace=natural|fast|slow|staccato, accent=neutral|american|american-valley|american-south|british|british-brixton|transatlantic|australian, voice_description, scene, temperature=0–2	WAV	$0.006
audio/gemini-2.5-pro-tts	text-to-speech	text *, voice=Achernar|Achird|Algenib|Algieba|Alnilam|Aoede|Autonoe|Callirrhoe|Charon|Despina|Enceladus|Erinome|Fenrir|Gacrux|Iapetus|Kore|Laomedeia|Leda|Orus|Puck|Pulcherrima|Rasalgethi|Sadachbia|Sadaltager|Schedar|Sulafat|Umbriel|Vindemiatrix|Zephyr|Zubenelgenubi, style=vocal-smile|newscaster|whisper|empathetic|promo-hype|deadpan, pace=natural|fast|slow|staccato, accent=neutral|american|american-valley|american-south|british|british-brixton|transatlantic|australian, voice_description, scene, temperature=0–2	WAV	$0.006
audio/elevenlabs-tts	text-to-speech	text *, voice=james|addison|adeline|allison|ana-rita|anika|arabella|aria|bella|benjamin|bradford|brian|britney|brittney|brock|burt|callum|cassidy|ed|edward|emma|eve|felix|finn|hank|hope-poetic|hope-bubbly|horatius|jamahal|jean|jeff|jessica-eloquent|jessica-chatty|john-doe|john-morgan|johnny-dynamite|johnny-kid|kuon|laura|leon|liam|lucy|lutz|mark-casual|mark-natural|marshal|monika|nathan|nathaniel|phil|rachel|rex-thunder|sam|spuds|sven|taksh|theodore|tiffany|tom|viking-bjorn|viraj-smooth|viraj-rich|wyatt|xavier, voice_id, speed=0.8|1.0|1.2	MP3	$0.037
audio/elevenlabs-multilingual	text-to-speech	text *, voice=james|addison|adeline|allison|ana-rita|anika|arabella|aria|bella|benjamin|bradford|brian|britney|brittney|brock|burt|callum|cassidy|ed|edward|emma|eve|felix|finn|hank|hope-poetic|hope-bubbly|horatius|jamahal|jean|jeff|jessica-eloquent|jessica-chatty|john-doe|john-morgan|johnny-dynamite|johnny-kid|kuon|laura|leon|liam|lucy|lutz|mark-casual|mark-natural|marshal|monika|nathan|nathaniel|phil|rachel|rex-thunder|sam|spuds|sven|taksh|theodore|tiffany|tom|viking-bjorn|viraj-smooth|viraj-rich|wyatt|xavier, voice_id, speed=0.8|1.0|1.2	MP3	$0.075

🧊 3D models

The output is GLB (some models also offer OBJ, FBX, STL, USDZ, USD, 3MF) plus a preview image: the task returns multiple files, each with its own extension. The file list is in results; each file has an ext field.

The price is known only after generation

3D is billed based on the resources actually used. We reserve the upper limit from the table, then charge the actual cost once it is known and refund the difference. The final amount is always in cost_usd.

model	Mode	input parameters	File	Price
3d/tripo-text	text→3D	prompt *, negative_prompt, texture_detail=standard|high, geometry_detail=standard|high, pbr=bool, quad_mesh=bool	GLB/FBX	$0.31–0.79
3d/tripo-image	photo→3D	image=url *, texture_detail=standard|high, geometry_detail=standard|high, pbr=bool, quad_mesh=bool	GLB/FBX	$0.31–0.79
3d/hunyuan-rapid-text	text→3D	prompt *, file_format=glb|obj|fbx|stl|usdz, pbr=bool, extra_mesh=bool	GLB and others	$0.43–0.71
3d/hunyuan-rapid-image	photo→3D	image=url *, file_format=glb|obj|fbx|stl|usdz, pbr=bool, extra_mesh=bool	GLB and others	$0.43–0.71
3d/hunyuan-pro-text	text→3D	prompt *, output=textured|mesh, polygons=auto|40000|100000|300000|800000|1500000, pbr=bool	GLB	$0.71–1.00
3d/hunyuan-pro-image	photo→3D	image=url *, output=textured|mesh, polygons=auto|40000|100000|300000|800000|1500000, pbr=bool	GLB	$0.71–1.00
3d/seed3d-image	photo→3D	image=url *, mesh_detail=low|medium|high, file_format=glb|obj|usd|usdz	GLB/OBJ/USD	$0.50
3d/meshy-text	text→3D	prompt *, texture_size=2k|4k|8k, pose=none|a-pose|t-pose, ultra=bool, pbr=bool, texture_text, texture_photo=url	GLB	$0.94–1.26
3d/meshy-image	photo→3D	image=url *, texture_size=2k|4k|8k, ultra=bool, texture=bool, pbr=bool, texture_text, texture_photo=url	GLB	$0.94–1.26
3d/meshy-multi-image	multiple views→3D	images=[url] * (1–4), texture_size=2k|4k|8k, ultra=bool, texture=bool, pbr=bool, texture_text	GLB	$0.94–1.26

Photo as texture. For 3d/meshy-*, the texture is specified separately from the geometry: with an image in texture_photo or words in texture_text.

Takes time. 3D generation takes from half a minute to several minutes — longer than an image or a short clip. You do not need to keep the connection open: the status is saved, and webhooks are available.

🛠 Utilities

model	Mode	input parameters	File	Price
image/topaz-upscale	upscaling	image=url *, scale=2|4|8	PNG	$0.063
image/recraft-remove-bg	editing	image=url *	PNG	$0.006
image/recraft-upscale	upscaling	image=url *	PNG	$0.006

Examples by Mode

Text to image

{ "model": "image/flux2-pro", "input": { "prompt": "neon fox in snow", "aspect_ratio": "16:9", "resolution": "2k" } }

Image editing

{ "model": "image/seedream-edit", "input": { "prompt": "remove the background, add neon", "image": "https://…/photo.png" } }

Photo→video (first/last frame, duration, audio)

{ "model": "video/wan-2.7-i2v", "input": {
    "prompt": "smooth camera push-in", "image": "https://…/first.png",
    "image2": "https://…/last.png", "duration": 8, "resolution": "1080p" } }

Music

{ "model": "music/suno-v5", "input": { "title": "Northern Lights", "style": "epic orchestral", "instrumental": true, "version": "v6" } }

Text to speech

{ "model": "audio/gemini-3.1-flash-tts", "input": {
    "text": "Hello, world", "voice": "Zephyr", "style": "empathetic",
    "pace": "natural", "accent": "neutral" } }

Two-voice dialogue — instead of the text field, put lines in dialogue and voices in voices (voice 1 is the first in the list):

{ "model": "audio/gemini-3.1-flash-tts", "input": {
    "voices": [ { "voice": "Kore", "style": "newscaster" }, { "voice": "Puck", "pace": "fast" } ],
    "dialogue": [ { "speaker": 1, "text": "Good evening!" }, { "speaker": 2, "text": "Glad to hear it." } ] } }

Avatar (talking video from a photo + audio)

{ "model": "video/omnihuman", "input": { "image": "https://…/face.png", "audio_url": "https://…/voice.mp3" } }

Statuses and Errors

GET /media/job/:id → status field:

Status	Meaning
processing	In progress — continue polling
done	Complete — file in result_url, charge in cost_usd
failed	Failed — reason in error, funds refunded

Response codes for POST /media/generate: 400 — unknown model or missing required field; 401 — missing/invalid key; 402 — insufficient balance; 403 — model unavailable for the account.

The error body always has the same format — { "error": "текст для человека", "code": "машинный код" }. Codes: unknown_model, missing_param (+ param field), invalid_input, invalid_attachment, prompt_too_long, input_too_long, no_auth, no_key, insufficient_balance, unpaid_shortfall, rate_limited, generation_failed, unavailable.

Estimate Cost in Advance

curl https://nordrouter.com/media/estimate \
  -H "Authorization: Bearer sk-nr-YOUR-KEY" -H "Content-Type: application/json" \
  -d '{ "model": "video/kling-3.0", "input": { "duration": 10, "mode": "pro", "audio": true } }'

{ "model": "video/kling-3.0", "usd": 1.6875 }

---

The parameters in the tables were verified against the service catalog on October 19, 2026; for the complete, up-to-date list of models, parameters, and prices, see GET /media/models.