add richinfoai models

mteb/models/overview.py

@@ -64,6 +64,7 @@
     repllama_models,
     rerankers_custom,
     rerankers_monot5_based,
+    richinfoai_models,
     ru_sentence_models,
     salesforce_models,
     sentence_transformers_models,
@@ -131,6 +132,7 @@
     repllama_models,
     rerankers_custom,
     rerankers_monot5_based,
+    richinfoai_models,
     ru_sentence_models,
     salesforce_models,
     sentence_transformers_models,

mteb/models/richinfoai_models.py

@@ -0,0 +1,27 @@
+from __future__ import annotations
+
+from mteb.model_meta import ModelMeta
+from mteb.models.bge_models import bge_full_data, bge_m3_training_data
+from mteb.models.stella_models import stella_zh_datasets
+
+ritrieve_zh_v1 = ModelMeta(
+    name="richinfoai/ritrieve_zh_v1",
+    languages=["zho_Hans"],
+    open_weights=True,
+    revision="f8d5a707656c55705027678e311f9202c8ced12c",
+    release_date="2025-03-25",
+    n_parameters=int(326 * 1e6),
+    memory_usage_mb=1242,
+    embed_dim=1792,
+    license="mit",
+    max_tokens=512,
+    reference="https://huggingface.co/richinfoai/ritrieve_zh_v1",
+    similarity_fn_name="cosine",
+    framework=["Sentence Transformers", "PyTorch"],
+    use_instructions=False,
+    superseded_by=None,
+    adapted_from=None,
+    public_training_code=None,
+    public_training_data=None,
+    training_datasets={**stella_zh_datasets, **bge_full_data, **bge_m3_training_data},
+)