GLM 5.2 is now available via a unified Model API

hpcaitech1 pts1 comments

HPC-AI | Access World-Class AI Models via Single Unified API

Model APIs

What's new

Instant Access for Frontier Open-Source AI Models<br>Build AI Apps and Agents with High-Performance Model APIs — No Deployment Required.<br>Start Free Trial

Everything You Need to Run AI Models

Z.ai: GLM 5.1

Input:$0.878 ~ $1.17/M<br>Output:$3.51 ~ $4.10/M

4 supported capabilities for production workloads<br>→Learn More

MoonshotAI: Kimi K2.5

Input:$0.60/M<br>Output:$3.00/M

4 supported capabilities for production workloads<br>→Learn More

DeepSeek: DeepSeek V4 Pro

Input:$1.74/M<br>Output:$3.48/M

4 supported capabilities for production workloads<br>→Learn More

OpenAI: GPT-5.5

Input:$5.00 ~ $10.00/M<br>Output:$30.00 ~ $45.00/M

6 supported capabilities for production workloads<br>→Learn More

Anthropic: Claude Opus 4.7

Input:$5.00/M<br>Output:$25.00/M

4 supported capabilities for production workloads<br>→Learn More

Z.ai: GLM 5.2

Input:$1.40/M<br>Output:$4.40/M

4 supported capabilities for production workloads<br>→Learn More

DeepSeek: DeepSeek V4 Flash

Input:$0.14/M<br>Output:$0.28/M

4 supported capabilities for production workloads<br>→Learn More

MoonshotAI: Kimi K2.6

Input:$0.95/M<br>Output:$4.00/M

4 supported capabilities for production workloads<br>→Learn More

→Explore More

Enterprise-Grade AI Infrastructure<br>Model APIs is built on scalable compute infrastructure designed to support large-scale AI workloads with high availability and performance.

GPU-accelerated inference

Elastic resource scaling

High availability architecture

Model APIs Pricing<br>modelprice($/M tokens)

Input Price<br>Output Price<br>Cache Read<br>No information available

Need more models?

Power Your Entire AI Ecosystem<br>Seamlessly connect Model APIs with the industry's leading agents, frameworks, and developer tools.<br>Open Claw<br>Open Code<br>Kilo Code<br>Aider<br>AutoGen<br>AutoGPT<br>Claude Code<br>Cline<br>Codex<br>Continue<br>CrewAI<br>Cursor<br>Dify<br>Droid<br>Flowise<br>Goose<br>Grok CLI<br>HammerAI<br>Haystack<br>Helicone<br>Hermes Agent<br>Janitor AI<br>LangChain<br>LangGraph<br>LiteLLM<br>LlamaIndex<br>MonkeyCode<br>Nexu Link<br>Nous Research API<br>Novelcrafter<br>Open WebUI<br>OpenHands<br>Portkey<br>Qwen Code<br>Roo Code<br>SillyTavern<br>SuperAGI<br>Zed<br>ZeroClaw

Get Started in Three Steps<br>Start building AI applications in minutes<br>Step 1<br>Create Account<br>Sign up and generate API key.

Step 2<br>Select a Model<br>Choose the AI model you need.

Step 3<br>Build Application<br>Integrate AI using simple APIs.

Start Building with Model APIs<br>Access powerful AI models and build intelligent applications faster.<br>Get StartedContact Sales

Frequently Asked Questions<br>What is HPC-AI.com Model APIs?HPC-AI.com Model APIs is a high-performance Model as a Service platform.<br>Instead of training and deploying your own models, you can access powerful inference capabilities through simple API calls on a pay-per-token basis.

Do you offer a free trial for users?

What models are available?

Do you offer multiple versions of models?

How do I integrate with the platform?

Is streaming supported?

Is multimodal supported?

How am I charged?

What are the prices?

Is customer data used for model training?

Will models be deprecated?

How do I log in?

Is there a limit on API Keys?

Is Rate Limiting supported?

Will more models be added?

Contact

model supported models apis input output

Related Articles