Average Ratings 0 Ratings
Average Ratings 0 Ratings
Description
Confidential AI offers a secure platform for private AI inference, agent environments, and confidential computing solutions tailored for AI laboratories, inference service providers, routers, and organizations that manage sensitive data. Its hosted inference service operates open-weight models within hardware-backed trusted execution environments (TEEs), utilizing technologies such as Intel TDX and AMD SEV-SNP confidential virtual machines, paired with NVIDIA GPUs in a confidential computing configuration. The system ensures that prompts, responses, and workload information remain safeguarded from infrastructure operators, allowing clients to authenticate attestation prior to submitting sensitive information. Users have the flexibility to implement an OpenAI-compatible API to establish inference endpoints and agent workloads, or alternatively, they can acquire the confidential-computing stack for deployment on their own premises, bare metal servers, or across major cloud platforms. This service is particularly beneficial for scenarios that involve serving models while restricting operator access to prompts and model assets, as well as executing sensitive AI workloads in a dedicated confidential computing environment. Additionally, the emphasis on robust security measures fosters trust in the processing of sensitive data, enabling clients to focus on innovation without compromising privacy.
Description
Hedra serves as a comprehensive platform for AI inference and model deployment, tailored for teams engaged in developing applications that utilize multimodal and physical AI. It features a managed API that grants access to a suite of models encompassing images, videos, audio, and language, all through a single endpoint with a pay-per-use billing structure. In addition, it includes a browser playground, a catalog of models, API keys, job tracking capabilities, webhooks, and controls for team management. Developers can seamlessly link to supported assistants via the Model Context Protocol server or explore models and submit inference jobs using a command-line interface. Moreover, Hedra facilitates model training and optimization for unique workloads, offering flexible deployment options across managed cloud, hybrid, or client-controlled on-premises environments. Additionally, a no-code Studio workspace empowers creators to effortlessly generate and modify visual content, enhancing the overall user experience. This combination of features makes Hedra an invaluable resource for teams looking to innovate in the AI space.
API Access
Has API
Yes
API Access
Has API
Yes
Screenshots View All
No images available
Screenshots View All
No images available
Integrations
ChatGPT
No
Claude
No
Claude Code
No
Codex CLI
No
Cursor
No
Hermes Agent
No
Model Context Protocol (MCP)
No
OpenAI
Yes
Integrations
ChatGPT
Yes
Claude
Yes
Claude Code
Yes
Codex CLI
Yes
Cursor
Yes
Hermes Agent
Yes
Model Context Protocol (MCP)
Yes
OpenAI
No
Pricing Details
$0.30 per 1M input tokens
Confidential Inference (OpenAI-compatible API, per 1M tokens):
DeepSeek V4.1 Flash (live): $0.30 input, $0.03 cached input, $1.20 output.
Kimi K3 (on request): $3.00 input, $0.30 cached input, $15.00 output.
GLM 5.3 (on request): $1.40 input, $0.26 cached input, $4.40 output.
Qwen3.8 27B (on request): $0.30 input, $0.03 cached input, $1.80 output.
DeepSeek V4 Flash is no longer offered; DeepSeek V4.1 Flash is the current live model.
Confidential GPU VMs (per GPU-hour, confidential computing mode):
NVIDIA B300: $7.50 (available).
NVIDIA B200: $6.50 (available).
NVIDIA H100: $3.25 (on request).
CPU VMs: $0.05 per core-hour plus $0.012 per GB-hour (AMD SEV-SNP and Intel TDX).
Licensed stack (per GPU per year): RTX PRO 6000 $2,789; H100 $3,876; H200 $4,974; B200 $7,625; B300 $9,884. CPU core license $48.85 per core per year. Standard support included; Enterprise support 25% of total license price.
Terms: minimum order $500. 50% on signed order, 50% on completion; monthly in arrears for longer engagements. USD by bank transfer (ACH or international wire). Card payments coming soon.
DeepSeek V4.1 Flash (live): $0.30 input, $0.03 cached input, $1.20 output.
Kimi K3 (on request): $3.00 input, $0.30 cached input, $15.00 output.
GLM 5.3 (on request): $1.40 input, $0.26 cached input, $4.40 output.
Qwen3.8 27B (on request): $0.30 input, $0.03 cached input, $1.80 output.
DeepSeek V4 Flash is no longer offered; DeepSeek V4.1 Flash is the current live model.
Confidential GPU VMs (per GPU-hour, confidential computing mode):
NVIDIA B300: $7.50 (available).
NVIDIA B200: $6.50 (available).
NVIDIA H100: $3.25 (on request).
CPU VMs: $0.05 per core-hour plus $0.012 per GB-hour (AMD SEV-SNP and Intel TDX).
Licensed stack (per GPU per year): RTX PRO 6000 $2,789; H100 $3,876; H200 $4,974; B200 $7,625; B300 $9,884. CPU core license $48.85 per core per year. Standard support included; Enterprise support 25% of total license price.
Terms: minimum order $500. 50% on signed order, 50% on completion; monthly in arrears for longer engagements. USD by bank transfer (ACH or international wire). Card payments coming soon.
Free Trial
Yes
Free Version
No
Pricing Details
$20/month
Hedra Studio offers a free plan; free outputs may be watermarked and are for non-commercial use. Paid Studio plans include Basic at $20/month for 2,000 credits, Pro at $50/month for 7,200 credits, Ultra at $100/month for 18,000 credits, and Teams at $75/month for 14,400 credits. Enterprise plans have custom pricing. API usage is billed per output or token from a prepaid balance; video generation starts at 1.22¢/second, image generation at 0.59¢/generation or 0.5¢/megapixel, and language-model rates vary by model and token type. Dedicated serving, model training, and GPU execution are priced per engagement.
Free Trial
No
Free Version
No
Deployment
Web-Based
Yes
On-Premises
Yes
iPhone App
No
iPad App
No
Android App
No
Windows
No
Mac
No
Linux
No
Chromebook
No
Deployment
Web-Based
Yes
On-Premises
Yes
iPhone App
No
iPad App
No
Android App
No
Windows
No
Mac
No
Linux
No
Chromebook
No
Customer Support
Business Hours
No
Live Rep (24/7)
Yes
Online Support
Yes
Customer Support
Business Hours
No
Live Rep (24/7)
No
Online Support
Yes
Types of Training
Training Docs
Yes
Webinars
No
Live Training (Online)
Yes
In Person
Yes
Types of Training
Training Docs
Yes
Webinars
No
Live Training (Online)
No
In Person
No
Vendor Details
Company Name
Confidential AI
Founded
2025
Country
United States
Website
confidential.ai
Vendor Details
Company Name
Hedra
Founded
2023
Country
United States
Website
www.hedra.com
Product Features
Virtual Machine
Backup Management
No
Graphical User Interface
No
Remote Control
No
VDI
No
Virtual Machine Encryption
No
Virtual Machine Migration
No
Virtual Machine Monitoring
No
Virtual Server
No
Product Features
Alternatives
Alternatives
No Alternatives