Deep Principle Releases Hydrogen-1.0 Open-Weight Science Model

1 reports archived1 independent sourcesupdated 11/3/2026, 02:45:09

Synthesis & Latest Updates

1 Sources Cross-Validated

Deep Principle released Hydrogen-1.0, an open-weight science model post-trained on GLM-5.3 using full lab notebook records and discovery trajectories, including failed attempts. It scores 38.3% on FrontierScience-Research, surpassing GPT-6-Astra's 33.3%, though it trails in NMR structure elucidation. The 753B parameter model features a 1M context window, is licensed under MIT, and supports BF16/NVFP4/MXFP4 quantization.

LATEST/Deep Principle released Hydrogen-1.0, an open-weight science model post-trained on GLM-5.3 using full lab notebook records and discovery trajectories, including failed attempts. It scores 38.3% on FrontierScience-Research, surpassing GPT-6-Astra's 33.3%, though it trails in NMR structure elucidation. The 753B parameter model features a 1M context window, is licensed under MIT, and supports BF16/NVFP4/MXFP4 quantization.

Hourly heat curve

6 fully observed hours

Now

16.1

Peak

17.803:00

24h change

–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first

Marktechpost (@Marktechpost)T2Nov 3, 02:45·78 pts
Deep Principle Releases Hydrogen-1.0 Open-Weight Science Model
- Hydrogen-1.0 enhances scientific discovery by RL training on complete experimental trajectories, including failures
- Outperforms GPT-6-Astra on FrontierScience-Research (38.3% vs 33.3%) but lags in NMR structure elucidation
- Features 753B parameters, 1M context, MIT license, and efficient NVFP4/MXFP4 quantized weights
View Related Reports & Analysis Original Source