Photoproduct for AI agents

Generate images and video in Claude

Ask in plain words. Your agent picks from 1,000+ models, tells you the price, and makes it from your credits.

Web, desktop and the iOS and Android apps. One link, then sign in. Tested end to end on claude.ai.

  • No subscription
  • Credits never expire
  • A daily limit per app

A real Photoproduct clip: MiniMax H3, 5 seconds, 13 credits.

The models people ask for. And a thousand more.

Video

Veo 3.1 · Kling 3.0 · Kling O3 · Seedance 2.0 · MiniMax H3 · Happy Horse 1.1 · Wan 3.0 · FLUX 3 · Pixverse v6 · LTX 2.3

Images

Nano Banana Pro · GPT Image 2.5 · Seedream 5 · FLUX.2 Pro · Ideogram · Recraft

Talking & lip sync

HeyGen Avatar 4 · OmniHuman 1.5 · Sync Lipsync · LatentSync

Upscale & restore

Topaz · SeedVR · Recraft Crisp

Music & voice

ElevenLabs v3 · ElevenLabs Music · Lyria 3 · MiniMax Music · Stable Audio 2.5

3D

Hunyuan3D

Made with Photoproduct

One agent. A whole creative factory.

Images, video and sound, each from one prompt. Open any piece to see exactly how.

Say it, get it

One sentence to your agent.

What you would type, and what came back.

01 · Product photo to scene

“Put my serum bottle on wet stone by a spa pool, morning light. Keep the label exactly.”

Nano Banana Pro edit, 3 credits. Six scene models compared in our blog, 1 to 3 credits each.

Your packshot

The plain serum packshot on white

After · Nano Banana Pro edit · 3 credits

The serum bottle on wet stone beside a steaming spa pool under a bright window

02 · Image to video

“Animate this photo: the camera slowly pushes in, the stream keeps moving.”

The still: FLUX.2 Pro, 1 credit. The clip: MiniMax H3 with sound, 13 credits. Kling 2.5 Turbo or Pixverse v6 make it for 5.

Before

A steel bottle with a teal strap on wet slate by a mountain stream

After · MiniMax H3 · 13 credits

The steel bottle by the stream as the camera pushes in, made with MiniMax H3

03 · A character that talks

“Make her dance, turn away and back, and say: "Photoproduct. Twelve hundred models, one endpoint."”

One MiniMax H3 call made the motion and the voice together: 5 seconds, 2K, 13 credits. For lip sync to your own audio there are Sync, LatentSync, OmniHuman and HeyGen Avatar 4.

Before

The same woman in a charcoal technical jacket on a grey field

After · MiniMax H3 · 13 credits

The character dances, turns away and back, and speaks a line to camera

04 · The same character, every time

“Lock this face, then put her in a technical jacket and make a character sheet.”

FLUX.2 Pro and its editor, 1 credit per step, each step fed the one before by its job reference.

Before

A character reference headshot: a woman with an auburn bob on a flat grey field

After · FLUX.2 Pro edit · 1 credit

A three-panel character sheet: jacket front, jacket back and a portrait

05 · Music and voice

“Make a calm 20-second instrumental bed for my product film.”

Stable Audio 2.5, 4 credits. For voice there are ElevenLabs v3 and others, from 2 credits; for songs, Lyria 3 and MiniMax Music.

Stable Audio 2.54 credits

Restrained instrumental bed for a product film. Slow warm analog synth pad, soft muted plucked arpeggio, low sustained sub-bass, sparse rim-click pulse. Calm, unhurried, confident, slightly cold and technical. Instrumental only, no vocals.

Why Photoproduct

Pay for what you make. Nothing more.

No subscription

Buy credits when you need them. They never expire, and a failed run costs nothing.

Same price as the web

Your agent pays the billed price you see in the studio, for every model, with no agent surcharge.

A daily limit per app

Each app you connect gets its own daily credit limit, so an agent cannot run away with your balance.

1,000+ models

Images, video, lip sync, upscaling, music, speech and 3D, from every major model maker, behind one connection.

What credits buy

$10 goes further than you think.

Starter

$1050 credits

Buys, for example

  • …FLUX.2 Pro imagesreading today’s price
  • …Nano Banana Pro imagesreading today’s price
  • …Kling 2.5 Turbo clips, 5 sreading today’s price
Most chosen

Popular

$30200 credits

Buys, for example

  • …Nano Banana Pro imagesreading today’s price
  • …Kling 2.5 Turbo clips, 5 sreading today’s price
  • …MiniMax H3 clips with sound, 5 s 2Kreading today’s price

Pro

$1001,000 credits

Buys, for example

  • …Nano Banana Pro imagesreading today’s price
  • …MiniMax H3 clips with sound, 5 s 2Kreading today’s price
  • …Veo 3.1 Fast clips, 8 sreading today’s price

Billed prices from today's catalogue, the same in the browser and through your agent. Mix and match: one balance covers every model. Buy credits →

Questions

Before you connect

Is it safe to let an agent spend my credits?
You sign in with Google on our page; the app never sees a password or an API key. Each app you connect gets its own daily credit limit, which you set when you sign in. Claude and Codex ask before each run unless you tell them not to, and a failed run costs nothing.
What does it cost?
No subscription: buy credits when you need them ($10, $30 or $100) and they never expire. Each model has a price per run, shown before it starts, the same through your agent as in the browser. Images start at 1 credit, five-second clips at about 5.
Which Claude and ChatGPT plans work?
Claude: every plan, Free included (one custom connector on Free); on Team and Enterprise an owner adds it for the workspace. ChatGPT: you add it as a custom MCP server on chatgpt.com/plugins, which worked on a Free plan in our test; in a Team or Enterprise workspace an admin may need to allow custom servers first.
Does it work on my phone?
Yes. Add the connector to Claude on the web or the desktop app, and it is there in the Claude iOS and Android apps too. The page here works on a phone, and so does the Create studio.
How do I disconnect an app?
Open Account → Agentic AI and revoke it under Connected apps; it stops working at once. You can also remove the connector in the app itself.
Which models can my agent use?
All of them: the same catalogue as the browser studio, from Veo, Kling and Seedance to Nano Banana Pro, FLUX, ElevenLabs and more. Your agent can list them with their prices and measured success rates before it picks.

Your agent can make it.
Give it the tools.

FOR DEVELOPERS & AGENTSSTREAMABLE HTTP / MCP

For developers

One Streamable HTTP endpoint with OAuth 2.1 sign-in. Discover models and prices for free, run any of them, and chain finished jobs into the next call.

—
models
—
featured
—
MCP tools

01 / SET A BUDGET

Price before you create.

Use list_models with max_credits to shop within a budget. Estimates are in whole credits, with input-dependent and variable-length prices labelled.

02 / CHOOSE WITH CONFIDENCE

Real results, honest signals.

The MCP catalog reports success_rate and p50_seconds. Missing values mean unmeasured, never a hidden zero.

03 / KEEP CREATING

Chain jobs, skip the downloads.

Pass a completed job_ref or its public output URL into the next job’s file input. Build image-to-video workflows without shuttling bytes.

One endpoint. Your workflow.

Point a client at https://photoproduct.io/mcp. It finds the sign-in on its own: protected resource metadata at /.well-known/oauth-protected-resource, an OAuth 2.1 server with PKCE, client ID metadata documents and dynamic registration, scoped to this endpoint.

For a client that only takes a header, create a bearer token in Account → Agentic AI (shown once) and send it as Authorization: Bearer.

MCP quotes use $0.10 per credit, rounded up to whole credits; a pack's own price per credit is on Buy credits. Parameters and inputs can change a model's estimate.

MCP client config (OAuth)
{
  "mcpServers": {
    "photoproduct": {
      "url": "https://photoproduct.io/mcp"
    }
  }
}
With a bearer token instead
{
  "mcpServers": {
    "photoproduct": {
      "url": "https://photoproduct.io/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_TOKEN"
      }
    }
  }
}
Example tool call
{
  "jsonrpc": "2.0", "id": 1, "method": "tools/call",
  "params": {
    "name": "list_models",
    "arguments": {"category": "text-to-image", "max_credits": 2}
  }
}

Find image models within 2 credits. Send this tools/call request in a connected MCP session; discovery does not start a paid generation.

02 / LIVE CATALOG PLAYGROUND

Find a model. Build your tool call.

Live catalog · refreshed every five minutes · no account required

Loading the public catalog…

Choose a category and a budget, or inspect a model above. Send the request below to see the live catalog response.

This is live, read-only discovery. Browsing and copying requests do not start a generation.

To generate, connect your app above and sign in with Google.

Public catalog request
GET /api/public/models?limit=6&offset=0
No account required · no credits spent
Use these filters in an MCP client
MCP tool call
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "list_models",
    "arguments": {
      "compact": true,
      "limit": 6,
      "offset": 0
    }
  }
}

The public request returns catalog summaries. Authenticated list_models also provides full model parameters and defaults.

Response

Send the request to see the live JSON response here.
View public catalog JSON ↗

03 / A RECORDED CHAIN

Five calls, one character, image to video and sound.

01 / A REAL MCP WORKFLOW

One character. Across generations.

A fresh model call has no memory of the last one. This recorded workflow carries identity forward with prompt_guide instructions and completed job references.

5 steps

five successful jobs

Recorded 1 Oct 2026

Original generated face lock with auburn bob and beauty markFinal MiniMax H3 video frame after the character turns back to camera
01 · Original face lock04 · Final video frame

From the original face lock to the final video frame: the character’s face, eye colour, auburn bob and beauty mark remain recognizable.

Playback of saved outputs, not live generation. To create your own, sign in with Google and create an MCP bearer token.

Face lock → outfit → character sheet + video

The sheet and video both use the outfit’s job_ref. The video includes speech; the music bed is generated separately. No image download or re-upload between calls.

No file inputSaved output
Generated face lock: auburn bob, grey-green eyes and a beauty mark

OBSERVED IN THIS RUN

This first plate establishes the character’s face, hair, eyes and beauty mark for the following generations.

STEP 1 / 5 · 1 CREDIT

Face lock

flux-2-pro

A 3:4 identity plate: forehead to upper chest, arms down, hands out of frame. The 12-point face-lock grammar fixes bone structure, hair, makeup and identity markers on a flat 18% gray field with shadowless lighting.

Prompt grammar: prompt_guide · face_lock

Recorded run · parameter excerpt
{
  "model": "flux-2-pro",
  "params": {
    "image_size": "portrait_4_3"
  }
}

RETURNED JOB_REF

job_NeOlT6PjpchlYwtG1KvV5w
Read & copy the full recorded prompt and arguments
Full recorded run arguments
{
  "model": "flux-2-pro",
  "params": {
    "prompt": "A clean cinema-character-reference 3:4 headshot, framed from forehead to upper chest with the face filling most of the frame. A Northern European woman in her early thirties, lean athletic build, fair skin with a cool neutral undertone and a matte finish. Dark auburn hair in a deep copper-brown, cut to a blunt chin-length bob with a centre parting tucked behind both ears, fine baby hairs at the hairline. Strong angular features: high wide cheekbones, a square clean jawline, a straight narrow nose, thin upper lip with a fuller lower lip. Wide-set grey-green eyes, level outer corners. Straight dark brows, natural thickness, low arch. Bare skin register, no visible makeup, no liner, no lash drama. A small flat beauty mark below the outer corner of the left eye and no other facial markings: no freckles, no piercings, no tattoos, no scars. Both arms down at her sides, hands not visible, nothing in frame but head, neck, shoulders and upper chest. Plain black thin-strap camisole, no jewellery, no logos, no graphics. Body squared to camera, head level, neutral relaxed expression, eyes to camera, lips closed. BACKDROP: a single flat 18% neutral gray field, one uniform value at every pixel, corner to corner, no seam line, no gradient, no hotspot, no vignette, no falloff. It is a flat colour field, NOT a photographed backdrop: no surface, no floor, no wall, no corner, no horizon. LIGHT: completely flat shadowless illumination, one enormous soft frontal source at camera position, matched equal fill from camera-left, camera-right, above and below, so both sides read at exactly the same brightness. No key-and-fill ratio, no modelling, no shadow side, no nose shadow, no under-chin shadow, no rim light, no hair light, no kicker, no specular hotspot. Extremely low contrast, even, milky, catalogue-flat. NO SHADOW OUTSIDE THE SUBJECT: no cast shadow, no contact shadow, no drop shadow, no ambient occlusion where the body meets the background, no darkening behind the shoulders, no halo, no edge darkening. Shading exists only on the subject and stops cleanly at the silhouette. No light bleed, no spill, no glow, no bounce, no reflected colour cast on the background. Skin reads matte and velvety with light natural sheen on the high points only, no oily shine, real fine even pore texture, real peach fuzz at the jaw and hairline, subsurface scattering reading as semi-translucent biology. Skin renders at its true tone, never darkened, never pale, never washed-out, never cool-shifted by the background. Never plastic, never doll-skin, never waxy. Even sharpness edge to edge. No depth-of-field falloff, no bokeh, no background blur, no lens vignette, no distortion, no flare, no bloom, no chromatic aberration, no haze. Photographed on a 50mm prime. Photographed, not generated. No text, no letters, no numbers, no watermarks anywhere in the image.",
    "image_size": "portrait_4_3"
  }
}

Pass the full argument object to the MCP run tool. To reproduce the workflow, use your own returned job references; these references identify the recorded jobs.

1 of 5

Recorded 1 Oct 2026 in one agent session: each step fed the next by its job reference.

Read the full SingularityByte experiment ↗