Overdo is an AI platform by AUOTAM Inc. that combines agent, chat, code, image, video, and calls in one application. It runs on Overdo's own GPU fleet, which keeps responses fast and data private.

How much does Overdo cost?

Overdo Chat is free with daily limits. Flash is $29/mo for the attended agent and more generation. Pro is $99/mo for autonomous runs, video and the highest limits. Enterprise GPU cloud pricing on request.

Can I resell Overdo's models?

Yes — the Models API provides drop-in compatible endpoints. You set your own pricing and sell under your own brand; Overdo handles the infrastructure.

Every capability, one key.

The inference API behind the Overdo workspace — chat, image, video, calls, and sites, served from our own GPU fleet. Included with Pro. White-label it on Scale.

Get API access — Pro Talk about Scale

The real endpoint surface

These are the live routes the workspace itself calls — not a compatibility shim.

Method	Endpoint	Capability
POST	/api/chat	Streaming chat & vision — web-aware, file-aware, memory-backed
POST	/api/enhance	Prompt enhancement — rewrites a rough idea into a generation-ready prompt
POST	/api/image	Text-to-image and image-variation batches on the fleet
POST	/api/video	Text-to-video clips, queued and rendered on GPUs
POST	/api/tts	Text-to-speech, streaming audio
POST	/api/stt	Speech-to-text transcription
POST	/api/site	Prompt → full website markup, streamed
POST	/api/publish	Publish a generated site to a shareable URL
GET	/api/sessions	List, read, and manage chat sessions
GET	/api/memory	Read and update durable per-user memory
GET	/api/usage	Live quota & credit usage for the key

Ship it tonight.

Plain HTTP, streaming responses, JSON in and out. No SDK required — though the same call shapes work from Python, Go, or a serverless function.

Streaming on every generative endpoint — tokens, audio, frames
Capacity kept warm for your workloads — no cold-start queue
One key covers every mode — no per-model juggling

generate.shcURL

curl https://overdo.ai/api/image \
  -H "authorization: Bearer $OVERDO_KEY" \
  -d '{"prompt":"a corgi riding a horse
        through a sunny meadow"}'

# → 4 renders, streamed as they're done

curl https://overdo.ai/api/video \
  -H "authorization: Bearer $OVERDO_KEY" \
  -d '{"prompt":"waves on rocks, cinematic"}'

# → mp4, ~8s, 1080p

Same rails as the product

No second-class endpoints — the API is the same code path the Overdo workspace calls. What the app does, your key can do.

Fleet-backed, not rented

Generations run on Overdo's dedicated GPU capacity — kept warm for your workloads, so cold starts are rare, not the default.

White-label & resell

On Scale: sub-keys per customer, usage metering, spend caps, and your name on the invoice. Overdo stays invisible.

Build on the same fleet.

Start building — Pro See pricing