Deep Principle releases Hydrogen-1.0, a 753B open-weight scientific model under the MIT License

Deep Principle has released Hydrogen-1.0, a 753B open-weight model for scientific discovery in chemistry, materials science and life sciences. The weights ship under the MIT License, according to MarkTechPost's report on the release.

What was released

Hydrogen-1.0 is built by post-training Z.ai's GLM-5.3 on three kinds of data: domain data, laboratory records and agent discovery trajectories. The model targets chemistry, materials science and life sciences.

Key facts

• Developer: Deep Principle
• Base model: Z.ai's GLM-5.3
• Parameters: 753B
• License: MIT License, with open weights
• Training data: domain data, laboratory records and agent discovery trajectories
• Target domains: chemistry, materials science, life sciences

Benchmark claim

The release reports a win over GPT-6-Astra on FrontierScience-Research. This is a result reported with the release, and the underlying scores are not given in the summary available at publication. Context window, pricing and training cutoff date are not yet disclosed.

What this means

Hydrogen-1.0 is a domain-specialized model built on an existing base model (GLM-5.3) rather than trained from scratch. Its training mix includes agent discovery trajectories alongside domain data and laboratory records. The MIT License is a permissive license, and open weights mean the model can be downloaded and run directly. The FrontierScience-Research comparison against GPT-6-Astra is the main performance claim reported so far.

The original coverage is from MarkTechPost.