Move from one lab's models to every model, plus your own
Flow is Google's creative studio on Veo 3.1, Gemini Omni and Nano Banana, sold through Google AI plans. Hedra Agent and Workspaces run those same models beside Kling, Seedance, MiniMax and open weights, and Hedra post-trains models to your data on GPU infrastructure it runs itself.
export HEDRA_API_KEY="<key_id>:<secret>"
curl -X POST \
"https://api.hedra.com/v3/models/veo-31" \
-H "Authorization: Key $HEDRA_API_KEY" \
-H "Content-Type: application/json" \
-d '{ "input": {
"prompt": "A slow dolly through a rain-lit Tokyo alley at night",
"aspect_ratio": "16:9",
"resolution": "1080p",
"duration_ms": 8000
} }'A more modern way to work
Skip the model aggregators. Hedra Agent and Workspaces give your team an autonomous agent that supercharges your workflows with a superintelligent understanding of the latest models: brief it in a shared space and it plans, generates, and edits across video, image, speech, and music, with your team and your agents working alongside it.
GPU infrastructure, an inference engine, and post-training research
Hedra operates GPU clusters and the inference engine that runs diffusion and autoregressive media models on them 8x faster. The same research team post-trains models for customers: supervised fine-tuning, LoRA training, and GRPO, on your data.
Hedra and Google Flow: focus at a glance
Flow is a creative studio for one lab's models, priced in Google AI credits. Hedra is the infrastructure and research company that serves those models beside everyone else's and trains yours.
| Dimension | Hedra | Google Flow |
|---|---|---|
| What it is | GPU infrastructure, inference and post-training research | A creative studio for video and image inside Google's AI plans |
| Core technology | Hedra's inference engine, its own GPU clusters, and post-training (SFT, LoRA, GRPO) | Veo 3.1, Gemini Omni and Nano Banana with Scenebuilder, frames-to-video and extension tools |
| Models | Open-weight models on Hedra's engine, models post-trained for you, and partner models through a gateway | Google's own models only |
| How you connect | API, SDK, CLI, MCP, and Hedra Workspaces | Flow at labs.google, with Google AI Plus, Pro and Ultra credits |
| Where inference runs | Hedra's GPU clusters, or your own | Google Cloud |
| Best fit | Teams building visual products who need inference, capacity and custom models | Creators on a Google AI plan who want Google's models |
Flow's models on Hedra, beside every other leading model
Video rates are per second of generated clip; image rates are per generation. Google's models are reached through the gateway; open weights run on Hedra's engine.
| Provider | Model ID | Rate |
|---|---|---|
| veo-31 | 39.29¢/s · 1080p | |
| gemini-omni-flash | 12.67¢/s · 720p | |
| nano-banana-pro | 15¢/image · 1K | |
| Kling | kling-v3 | 16.8¢/s · 1080p |
| ByteDance | seedance-25 | 79.09¢/s · 1080p |
| MiniMax | minimax-h3 | 6¢/s · 768p |
| Alibaba | wan-3-0 | 20¢/s · 1080p |
FAQs
- Are these the same models Flow uses?
- Yes. Veo 3.1, Gemini Omni Flash and the Nano Banana models are Google's, served by Google and reached through Hedra's gateway with the same request shape as every other model in the catalog. Flow's editing tools are Flow's; the models are the same.
- What does Hedra offer that Flow does not?
- Every other maker's model beside Google's, open-weight models running on Hedra's own inference engine, custom models post-trained on your data with supervised fine-tuning, LoRA training and GRPO, and deployment on your own hardware. Flow is one lab's models inside one subscription.
- How does pricing compare?
- Flow is sold in Google AI credits: as of September 2026 the free tier gives 50 daily credits, Google AI Plus is $4.99 a month for 200, Pro is $19.99 for 1,000, and Ultra is $99.99 or $199.99 for 10,000 or 25,000. Hedra bills per second of video or per image at the rates above, with no plan to buy first.
