Together AI supplies hosted inference, dedicated endpoints and fine-tuning for supported models.
Best for: Hosted open-model inference and fine-tuning
Coding & DevelopmentAI Infrastructure & Developer APIs
Back to discovery
š§©
Pricing unconfirmedInformation reviewed 2026-11-05
Hedra
About this listing
Reviewed from the official site on 5 Nov 2026 after the company asked us to update the listing. Pricing tiers are not shown on the home page.
Official sourceTraining and inference infrastructure for multimodal and physical AI: post-training, GPU sandboxes, hosted inference and a developer API.
Best suited to
Hedra is a research and infrastructure company.
Meet Hedra
Hedra is a research and infrastructure company. Its site describes post-training of open models and vision-language-action (VLA) models, GPU sandboxes for training code and evaluation, hosted inference on Hedra Cloud or your own infrastructure, and a World Model Runtime for diffusion, video and world action models on NVIDIA and AMD GPUs. It also offers a developer API that uses an API key and a prepaid USD wallet. It suits AI and robotics teams that train or serve their own models, and the company says its focus is now training and inference infrastructure.
A practical workflow
- Read the developer overview and model catalog to find the model and operation you need
- Create an API key and add a small amount to the prepaid wallet
- Send one small test request and record the output and cost
- Compare cost and response time with another inference provider for the same model
- Decide whether training, hosted inference or your own infrastructure fits
Illustrative input
This example describes a task, not a tested output. Adjust the source material and settings to your own situation.
Serve our fine-tuned vision model through a hosted endpoint and report the cost and response time for 100 test requests.
Result check: Compare the measured cost and response time with your budget and with at least one other inference provider.
Access and alternatives
The Hedra home page does not show pricing tiers. API access uses an API key and a prepaid USD wallet, so check the developer overview and your account for current rates before committing.
Choose another workflow when its emphasis fits better: fal for pay-per-use hosted generative model APIs; Replicate for running open models through a simple API; Together AI for hosted open-model inference and fine-tuning.
What you can do
Post-training of open and vision-language-action models
GPU sandboxes for training code and evaluation
Hosted inference on Hedra Cloud or your own infrastructure
World Model Runtime for diffusion, video and world action models
Sovereign deployment within your jurisdiction
Developer API with an API key and prepaid USD wallet
Where it helps
- Covers training, evaluation and serving in one provider
- Runs on NVIDIA and AMD GPUs
- Lets teams keep weights and data in their own jurisdiction
What to consider
- Aimed at AI and robotics teams, not casual users
- The home page shows no pricing tiers
- Wallet-based API billing needs cost tracking
Pricing & free access
Official pricingPricing unconfirmed
The Hedra home page does not show pricing tiers. API access uses an API key and a prepaid USD wallet, so check the developer overview and your account for current rates before committing.
Plan information can change. Prices and conditions are set by the provider. Information checked 2026-11-05.
Source for access detailsGet free-plan changes by email
One email a week: new free AI tools, prompts and free-plan changes. No spam, unsubscribe any time.
By subscribing you agree to receive this email from Genius Forges and to us storing your address to send it. We may ask you to confirm by email. Every email has an unsubscribe link. See our privacy policy.
Questions about Hedra
Who is Hedra useful for?
Hedra is a research and infrastructure company. Its site describes post-training of open models and vision-language-action (VLA) models, GPU sandboxes for training code and evaluation, hosted inference on Hedra Cloud or your own infrastructure, and a World Model Runtime for diffusion, video and world action models on NVIDIA and AMD GPUs. It also offers a developer API that uses an API key and a prepaid USD wallet. It suits AI and robotics teams that train or serve their own models, and the company says its focus is now training and inference infrastructure..
How is Hedra priced?
The Hedra home page does not show pricing tiers. API access uses an API key and a prepaid USD wallet, so check the developer overview and your account for current rates before committing.
How should I start a small evaluation?
Read the developer overview and model catalog to find the model and operation you need. Create an API key and add a small amount to the prepaid wallet. Send one small test request and record the output and cost.
What should I check in the result?
Compare the measured cost and response time with your budget and with at least one other inference provider.
When should I choose a different provider?
fal for pay-per-use hosted generative model APIs; Replicate for running open models through a simple API; Together AI for hosted open-model inference and fine-tuning.
Did this help you complete your task?
Tell us what needs improving. No account or private output required.
Report incorrect information or a broken linkReview evidence
Documentation-based review Ā· 2026-11-05
Rewritten from the official site on 5 Nov 2026 around its current infrastructure focus. The workflow and example are editorial suggestions; no account-level product test was performed.
Put it to work
Other tools to consider
Explore all toolsReplicate hosts models with versioned inputs and APIs for generating or processing media and text.
Best for: Running hosted AI models through an API
Coding & DevelopmentAI Infrastructure & Developer APIs
fal offers hosted generative-media models and developer APIs.
Best for: Image, video and audio generation APIs
Image GenerationAI Infrastructure & Developer APIs