SalamandraTA 2B Academic
SalamandraTA 2B Academic adapts BSC's translation model to academic text.
Model and training
The release fine-tunes SalamandraTA 2B Instruct on ACAData, an academic parallel corpus. It retains the small 2B architecture and the 8,192-token sequence limit. Its card describes translation use and does not establish general chat capability. The academic training makes this a distinct derivative, not a prompt preset for the general translation model.SalamandraTA 2B Academic: official model card
Practical use
Practical guidance. Use paired academic passages to check terminology, equations in text and citation preservation. A specialised model can still alter cautious language or introduce claims absent from the source. Compare it with the general TA 2B model on both academic and ordinary documents to measure the effect of domain tuning.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | salamandraTA-2B-academic |
| Context window | 8,192 tokens Maximum sequence length documented for these weights; the prompt and generated text share the window. SalamandraTA 2B Academic: official model card |
| License | Apache License 2.0.SalamandraTA 2B Academic: official model card |
| Model size | 2,253,490,176 parameters.SalamandraTA 2B Academic: official model card |
Related models
- SalamandraTA 2B
Barcelona Supercomputing Center · SalamandraTA
SalamandraTA 2B is a small translation model derived from BSC's Salamandra 2B base.
Launch date not confirmedDownloadable - SalamandraTA 7B v3
Barcelona Supercomputing Center · SalamandraTA
SalamandraTA 7B v3 is BSC's expanded translation model with terminology and structured-document support.
Launch date not confirmedDownloadable