Pharia 1 LLM 7B Control Aligned
Pharia 1 Control Aligned adds safety alignment to the 7B Control model. Aleph Alpha recommends this variant for chat applications.
Model design
The aligned and Control versions share a pretrained decoder backbone. Commercial on-premises deployment is offered through a separate customer agreement.Pharia 1 language model card
Practical use
Practical guidance. Evaluate the aligned model on representative acceptable requests as well as requests the application should decline. Check that its response format still meets downstream requirements. Alignment reduces some risks but does not replace application permissions or validation.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable, Provider product |
| Model identifiers | Aleph-Alpha/Pharia-1-LLM-7B-control-alignedAleph-Alpha/Pharia-1-LLM-7B-control-aligned-hf |
| Context window | 8,192 tokens Published checkpoint context. Pharia 1 language model card |
| License | Open Aleph License for non-commercial research and education.Pharia 1 language model card |
| Languages | Trained in seven languages; evaluated in English, German, French and Spanish.Pharia 1 language model card |
Related models
- Kolibri 1
Aleph Alpha ยท Kolibri
German-and-English reasoning model with 78.1 billion total parameters and 3.46 billion active per token.
Launch date Downloadable