American tech giant NVIDIA is providing free access to five cutting-edge Chinese neural networks. Users will not need bank cards, paid subscriptions, or complex setups. Simply obtain a single nvapi key to connect DeepSeek, Qwen, Kimi, MiniMax, and GLM through the familiar OpenAI-compatible interface.
This initiative has already generated widespread resonance in the professional community. Developers can test the models on the build.nvidia.com platform, each tailored for a specific range of tasks.
What's included in the free starter kit
On the build.nvidia.com platform, developers can test five powerful neural networks. Each is tailored for a specific range of tasks.
| AI Model Name | Primary Specialization and Purpose |
| DeepSeek V4 Flash | Ultra-fast processing of text queries |
| MiniMax M3 | High-quality assistance in writing software code |
| Qwen3.5-397B | Solving complex logical problems and reasoning |
| Kimi K2.6 | Creating autonomous agents and long chains of thought |
| GLM 5.1 | Reliable all-purpose neural network for everyday use |
The main advantage of this format lies in complete freedom of action. There's no need to pay for separate subscriptions for different platforms, change familiar work tools, or adapt to the requirements of a specific vendor. A single free key opens access not only to this five but to the entire NVIDIA catalog, which contains over 100 AI options.
The established limit is about 40 requests per minute. According to developers, this volume is quite sufficient for creating personal and commercial prototypes.
How to quickly connect neural networks
The initial setup process takes just a couple of minutes. You need to complete a few simple steps:
- Create a profile on the build.nvidia.com website and confirm your email.
- Generate a personal nvapi authorization key in your account.
- Specify the address integrate.api.nvidia.com/v1 as the base URL.
- Select the desired neural network from the list and start testing.
To send requests, you will need the exact technical identifiers:
- minimaxai/minimax-m3
- qwen/qwen3.5-397b-a17b
- moonshotai/kimi-k2.6
- zhipuai/glm-5.1
- deepseek/deepseek-v4-flash
As a result, project authors gain access to products capable of competing with GPT and Claude, without spending a single ruble.
Expert opinion: This is a strategically important step from NVIDIA. The company is not just expanding its ecosystem but is effectively democratizing access to the most advanced LLM models from China. For developers, especially those working at the intersection of AI and blockchain, this is an opportunity to integrate powerful language models into their projects without initial capital investment. The AI services market is becoming increasingly competitive, and free API keys from NVIDIA are a serious argument for retaining the developer audience.