Fix: minicpmv2 (#1705)

* updmini cpm * flash_attn implementation * remove flash attn
embeddings-benchmark · Jan 6, 2025 · 222bb35 · 222bb35
1 parent 4a1c8e6
commit 222bb35
Showing 1 changed file with 7 additions and 0 deletions.
diff --git a/mteb/models/misc_models.py b/mteb/models/misc_models.py
@@ -2,6 +2,8 @@
 
 from functools import partial
 
+import torch
+
 from mteb.model_meta import ModelMeta, sentence_transformers_loader
 
 Haon_Chen__speed_embedding_7b_instruct = ModelMeta(
@@ -1556,6 +1558,11 @@
         sentence_transformers_loader,
         model_name="openbmb/MiniCPM-Embedding",
         revision="c0cb2de33fb366e17c30f9d53142ff11bc18e049",
+        # https://huggingface.co/openbmb/MiniCPM-Embedding/blob/c0cb2de33fb366e17c30f9d53142ff11bc18e049/README.md?code=true#L405
+        model_kwargs={
+            # "attn_implementation": "flash_attention_2",
+            "torch_dtype": torch.float16,
+        },
         trust_remote_code=True,
     ),
     name="openbmb/MiniCPM-Embedding",