Hedra
Hosted inference and model-building service for visual and multimodal AI
Linked from hedra.com · checked 29 Oct
- Tool type
- Hosted visual and multimodal AI inference service
- Made for
- Developers and enterprise teams building visual AI
- Runs on
- Hedra Cloud, customer hybrid environments, or on-prem clusters
- API calls are made to api.hedra.com; Studio is a web app.
What it does
Hedra accepts image, text, audio and video inputs through its API or Studio web app and returns generated media, speech, music or text answers. Developers can send an image and a question to the chat completions endpoint for an immediate text response, or submit asynchronous jobs to produce video, image, speech or audio.
Teams use the API to add visual generation, lip-sync, upscaling, speech, music or scene understanding to their own applications, paying per output or per token from a prepaid balance. Hedra also offers its World Model Runtime for diffusion, video and world models, along with managed, hybrid or on-premises deployment and custom model work with its researchers.
Hedra suits developers and product teams building visual or multimodal products who want hosted inference without running their own serving stack. Access requires creating an API key with chat:write access and funding the API wallet, which is separate from Studio credits. Dedicated serving, training and GPU execution are quoted per engagement, and per-model rates and limits can change, so check the model catalog and rate table before committing.
What it can do
- Run image and text reasoning through a chat completions endpoint
- Generate video, image and speech via asynchronous model jobs
- Upscale and restore visual media
- Add lip-sync and avatar generation
- Create and edit audio, including music and sound effects
- Serve custom models through managed, hybrid or on-prem deployment
Before you choose
- API calls are billable, and reasoning tokens count toward the output budget.
- Image URLs must be accessible to the service.
- Model limits and availability can change; check the model catalog and rate table.
Listed under
Get started
The API wallet is separate from Studio credits.
- Create an API key with chat:write access in the Hedra console.
- Fund the API wallet, which is separate from Studio subscription credits.
- Set HEDRA_API_KEY in your environment and check model capabilities at the models endpoint before sending requests.
Plans and prices
Prices in USDMedia rates are charged per second, minute, megapixel or generation; LLM and VLM rates are per million tokens with input, cached and output charges. Dedicated serving, training and GPU execution are quoted per engagement. Voice cloning is listed as free. Check the rate table for full units.
- Prepaid dollar balance
- No subscription, seats, or minimum
Source: hedra.com/pricing, checked 29 Oct 2026. Prices can change; check before you buy.
Sources for this overview
Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.
- Hedra — Visual Inference Models and Infrastructure ↗www.hedra.com
- Hedra Pricing | API, Infrastructure, and Studio ↗www.hedra.com
- Run vision and reasoning inference - Hedra ↗www.hedra.com
Ratings and reviews
No ratings for Hedra yet.
What’s changed
Free accountSignals
Free accountData and security
Free accountCompare with similar tools
Free accountWhere this comes from
How we research- Company website
- hedra.com: product, pricing and policy pages
- YC directory mirror
- Batch and company name
- Social accounts
- Accounts the company links, and its posts on X
- Human review
- Checked by a person
- knowledge_base
Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.