1.25.16

1.25.16 (2025-01-01)

Fix

• fix: Update BUCC dataset revision (#1674)

• trust remote code

• Update revision (343edc4)

1.25.15

1.25.15 (2025-01-01)

Fix

• fix: Use prompts instead of prompt names for voyage (#1665)

• fix prompt names

• lint

• change input type (82e9949)

• fix: remove model as a parameter for MulticlassClassification (#1666)

remove model parameter (5cfcc77)

1.25.14

1.25.14 (2025-01-01)

Fix

• fix: Updated metadata for CPM (#1670)

• fix: Pass trust_remote_code=True to CPM model

Fixes #1651

• fix: Updated metadata for cpm (f99a178)

1.25.13

1.25.13 (2025-01-01)

Fix

• fix: Pass trust_remote_code=True to CPM model (#1669)

Fixes #1651 (f426159)

1.25.12

1.25.12 (2025-01-01)

Fix

• fix: Use batch size kwargs for openai APIs (#1668)

Fixes #1645 (663653e)

1.25.11

1.25.11 (2025-01-01)

Fix

• fix: Update gritlm kwargs (#1643)

• Fix kwarg

• format

---

Co-authored-by: Kenneth Enevoldsen <[email protected]> (19cbf64)

1.25.10

1.25.10 (2025-01-01)

Fix

• fix: Cast all Model2Vec outputs as floats (#1667)

cast all outputs as floats (fa0ed6b)

1.25.9

1.25.9 (2024-12-30)

Fix

• fix: add missing benchmark to benchmarks.py (#1641)

add missing benchmark (27eb549)

1.25.8

1.25.8 (2024-12-30)

Fix

• fix: output_folder for co2 evaluation (#1642) (366b2ce)

1.25.7

1.25.7 (2024-12-29)

Fix

• fix: Correction of discrepancies for gte-Qweb model (#1637) (2de61b1)

Unknown

• Speed up test_save_predictions (#1631) (1b06601)