diff --git a/mteb/model_meta.py b/mteb/model_meta.py
index b105f301b6..c88326edc6 100644
--- a/mteb/model_meta.py
+++ b/mteb/model_meta.py
@@ -66,7 +66,8 @@ class ModelMeta(BaseModel):
         release_date: The date the model's revision was released.
         license: The license under which the model is released. Required if open_weights is True.
         open_weights: Whether the model is open source or proprietary.
-        public_training_code: Whether the code used to train the model is publicly available.
+        public_training_code: A link to the publicly available training code. If none it is assumed that the training code is not publicly available.
+        public_training_data: A link to the publicly available training data. If none it is assumed that the training data is not publicly available.
         similarity_fn_name: The distance metric used by the model.
         framework: The framework the model is implemented in, can be a list of frameworks e.g. `["Sentence Transformers", "PyTorch"]`.
         reference: A URL to the model's page on huggingface or another source.
@@ -94,6 +95,7 @@ class ModelMeta(BaseModel):
     license: str | None
     open_weights: bool | None
     public_training_code: str | None
+    public_training_data: str | None
     framework: list[FRAMEWORKS]
     reference: STR_URL | None = None
     similarity_fn_name: DISTANCE_METRICS | None
diff --git a/mteb/models/arctic_models.py b/mteb/models/arctic_models.py
index 66822d41b0..f765b01bff 100644
--- a/mteb/models/arctic_models.py
+++ b/mteb/models/arctic_models.py
@@ -102,7 +102,8 @@
     use_instructions=True,
     adapted_from="sentence-transformers/all-MiniLM-L6-v2",
     superseded_by=None,
-    public_training_code=None,  # couldn't find
+    public_training_code=None,
+    public_training_data=None,
     training_datasets={
         # source: https://arxiv.org/pdf/2405.05374
         # splits not specified to assuming everything
@@ -151,7 +152,8 @@
     use_instructions=True,
     adapted_from="intfloat/e5-small-unsupervised",
     superseded_by=None,
-    public_training_code=None,  # couldn't find
+    public_training_code=None,
+    public_training_data=None,  # couldn't find
     training_datasets={
         # source: https://arxiv.org/pdf/2405.05374
         # splits not specified to assuming everything
@@ -200,7 +202,8 @@
     use_instructions=True,
     adapted_from="intfloat/e5-base-unsupervised",
     superseded_by="Snowflake/snowflake-arctic-embed-m-v1.5",
-    public_training_code=None,  # couldn't find
+    public_training_code=None,
+    public_training_data=None,  # couldn't find
     training_datasets={
         # source: https://arxiv.org/pdf/2405.05374
         # splits not specified to assuming everything
@@ -249,7 +252,8 @@
     use_instructions=True,
     adapted_from="nomic-ai/nomic-embed-text-v1-unsupervised",
     superseded_by="Snowflake/snowflake-arctic-embed-m-v2.0",
-    public_training_code=None,  # couldn't find
+    public_training_code=None,
+    public_training_data=None,  # couldn't find
     training_datasets={
         # source: https://arxiv.org/pdf/2405.05374
         # splits not specified to assuming everything
@@ -298,7 +302,8 @@
     use_instructions=True,
     adapted_from="intfloat/e5-base-unsupervised",
     superseded_by="Snowflake/snowflake-arctic-embed-l-v2.0",
-    public_training_code=None,  # couldn't find
+    public_training_code=None,
+    public_training_data=None,  # couldn't find
     training_datasets={
         # source: https://arxiv.org/pdf/2405.05374
         # splits not specified to assuming everything
@@ -350,6 +355,7 @@
     adapted_from=None,
     superseded_by="Snowflake/snowflake-arctic-embed-m-v2.0",
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -375,7 +381,8 @@
     use_instructions=True,
     adapted_from="Alibaba-NLP/gte-multilingual-base",
     superseded_by=None,
-    public_training_code=None,  # couldn't find
+    public_training_code=None,
+    public_training_data=None,  # couldn't find
     training_datasets={
         # source: https://arxiv.org/pdf/2405.05374
         # splits not specified to assuming everything
@@ -423,7 +430,8 @@
     use_instructions=True,
     adapted_from="BAAI/bge-m3-retromae",
     superseded_by=None,
-    public_training_code=None,  # couldn't find
+    public_training_code=None,
+    public_training_data=None,  # couldn't find
     training_datasets={
         # source: https://arxiv.org/pdf/2405.05374
         # splits not specified to assuming everything
diff --git a/mteb/models/bge_models.py b/mteb/models/bge_models.py
index d8270c573b..d9eb64246d 100644
--- a/mteb/models/bge_models.py
+++ b/mteb/models/bge_models.py
@@ -372,7 +372,8 @@
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
-    public_training_code=None,  # seemingly released (at least for some models, but the link is broken
+    public_training_code=None,
+    public_training_data="https://data.baai.ac.cn/details/BAAI-MTP",
     training_datasets=bge_training_data,
 )
 
@@ -397,6 +398,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,  # seemingly released (at least for some models, but the link is broken
+    public_training_data="https://data.baai.ac.cn/details/BAAI-MTP",
     training_datasets=bge_training_data,
 )
 
@@ -421,6 +423,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,  # seemingly released (at least for some models, but the link is broken
+    public_training_data="https://data.baai.ac.cn/details/BAAI-MTP",
     training_datasets=bge_training_data,
 )
 
@@ -444,7 +447,8 @@
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
-    public_training_code=None,  # seemingly released (at least for some models, but the link is broken
+    public_training_code=None,
+    public_training_data=None,
     training_datasets=bge_chinese_training_data,
 )
 
@@ -468,7 +472,8 @@
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
-    public_training_code=None,  # seemingly released (at least for some models, but the link is broken
+    public_training_code=None,
+    public_training_data=None,
     training_datasets=bge_chinese_training_data,
 )
 
@@ -492,7 +497,8 @@
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
-    public_training_code=None,  # seemingly released (at least for some models, but the link is broken
+    public_training_code=None,
+    public_training_data=None,
     training_datasets=bge_chinese_training_data,
 )
 
@@ -516,6 +522,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=bgem3_training_data,
 )
 
@@ -549,5 +556,6 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,  # not disclosed
 )
diff --git a/mteb/models/bm25.py b/mteb/models/bm25.py
index ea56fd432b..6e3d3747d9 100644
--- a/mteb/models/bm25.py
+++ b/mteb/models/bm25.py
@@ -139,5 +139,6 @@ def encode(self, texts: list[str], **kwargs):
     framework=[],
     use_instructions=False,
     public_training_code="https://github.com/xhluca/bm25s",
+    public_training_data=None,
     training_datasets=None,
 )
diff --git a/mteb/models/cohere_models.py b/mteb/models/cohere_models.py
index 8718a2e2a3..60ff63ee81 100644
--- a/mteb/models/cohere_models.py
+++ b/mteb/models/cohere_models.py
@@ -234,7 +234,8 @@ def encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    public_training_code=None,  # assumed
+    public_training_code=None,
+    public_training_data=None,  # assumed
     training_datasets=None,
 )
 
@@ -257,7 +258,8 @@ def encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    public_training_code=None,  # assumed
+    public_training_code=None,
+    public_training_data=None,  # assumed
     training_datasets=None,
 )
 
@@ -280,7 +282,8 @@ def encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    public_training_code=None,  # assumed
+    public_training_code=None,
+    public_training_data=None,  # assumed
     training_datasets=None,
 )
 
@@ -303,6 +306,7 @@ def encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    public_training_code=None,  # assumed
+    public_training_code=None,
+    public_training_data=None,  # assumed
     training_datasets=None,
 )
diff --git a/mteb/models/colbert_models.py b/mteb/models/colbert_models.py
index 87b5fdb93a..89b09de28e 100644
--- a/mteb/models/colbert_models.py
+++ b/mteb/models/colbert_models.py
@@ -153,6 +153,7 @@ def similarity(self, a: np.ndarray, b: np.ndarray) -> np.ndarray:
     open_weights=True,
     revision="c1e84128e85ef755c096a95bdb06b47793b13acf",
     public_training_code=None,
+    public_training_data=None,
     release_date="2024-09-21",
     n_parameters=110 * 1e6,
     max_tokens=180,  # Reduced for Benchmarking - see ColBERT paper
@@ -205,6 +206,7 @@ def similarity(self, a: np.ndarray, b: np.ndarray) -> np.ndarray:
     open_weights=True,
     revision="4cf816e5e2b03167b132a3c847a9ecd48ba708e1",
     public_training_code=None,
+    public_training_data=None,
     release_date="2024-08-16",
     n_parameters=559 * 1e6,
     max_tokens=8192,
diff --git a/mteb/models/e5_instruct.py b/mteb/models/e5_instruct.py
index f4d5909350..c89b64fc72 100644
--- a/mteb/models/e5_instruct.py
+++ b/mteb/models/e5_instruct.py
@@ -15,6 +15,16 @@
 E5_INSTRUCTION = "Instruct: {instruction}\nQuery: "
 
 
+E5_MISTRAL_TRAINING_DATA = {
+    **E5_TRAINING_DATA,
+    "FEVER": ["train"],
+    "FEVERHardNegatives": ["train"],
+    "FEVER-PL": ["train"],  # translation not trained on
+    "HotpotQA": ["train"],
+    "HotpotQAHardNegatives": ["train"],
+    "HotpotQA-PL": ["train"],  # translation not trained on
+}
+
 e5_instruct = ModelMeta(
     loader=partial(  # type: ignore
         instruct_wrapper,
@@ -40,6 +50,7 @@
     license="mit",
     max_tokens=514,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=E5_TRAINING_DATA,
 )
 
@@ -70,5 +81,6 @@
     license="mit",
     max_tokens=32768,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=E5_TRAINING_DATA,
 )
diff --git a/mteb/models/e5_models.py b/mteb/models/e5_models.py
index ace25ca08d..0ad15e7320 100644
--- a/mteb/models/e5_models.py
+++ b/mteb/models/e5_models.py
@@ -126,6 +126,16 @@
     "NQ-PL": ["train"],  # translation not trained on
 }
 
+ME5_TRAINING_DATA = {
+    **E5_TRAINING_DATA,
+    "FEVER": ["train"],
+    "FEVERHardNegatives": ["train"],
+    "FEVER-PL": ["train"],  # translation not trained on
+    "HotpotQA": ["train"],
+    "HotpotQAHardNegatives": ["train"],
+    "HotpotQA-PL": ["train"],  # translation not trained on
+}
+
 e5_mult_small = ModelMeta(
     loader=partial(  # type: ignore
         sentence_transformers_loader,
@@ -147,7 +157,8 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
-    training_datasets=E5_TRAINING_DATA,
+    public_training_data=None,
+    training_datasets=ME5_TRAINING_DATA,
 )
 
 e5_mult_base = ModelMeta(
@@ -170,7 +181,8 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
-    training_datasets=E5_TRAINING_DATA,
+    public_training_data=None,
+    training_datasets=ME5_TRAINING_DATA,
 )
 
 e5_mult_large = ModelMeta(
@@ -194,7 +206,8 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
-    training_datasets=E5_TRAINING_DATA,
+    public_training_data=None,
+    training_datasets=ME5_TRAINING_DATA,
 )
 
 e5_eng_small_v2 = ModelMeta(
@@ -217,6 +230,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=E5_TRAINING_DATA,
 )
 
@@ -241,6 +255,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=E5_TRAINING_DATA,
 )
 
@@ -267,6 +282,7 @@
     superseded_by=None,
     adapted_from=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=E5_TRAINING_DATA,
 )
 
@@ -293,6 +309,7 @@
     superseded_by=None,
     adapted_from=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=E5_TRAINING_DATA,
 )
 
@@ -319,6 +336,7 @@
     superseded_by="intfloat/e5-large-v2",
     adapted_from=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=E5_TRAINING_DATA,
 )
 
@@ -345,5 +363,6 @@
     superseded_by="intfloat/e5-base-v2",
     adapted_from=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=E5_TRAINING_DATA,
 )
diff --git a/mteb/models/google_models.py b/mteb/models/google_models.py
index 08065f7af0..40d316fee7 100644
--- a/mteb/models/google_models.py
+++ b/mteb/models/google_models.py
@@ -151,7 +151,8 @@ def encode(
     similarity_fn_name="cosine",  # assumed
     framework=["API"],
     use_instructions=True,
-    public_training_code=None,  # assumed
+    public_training_code=None,
+    public_training_data=None,  # assumed
     training_datasets=None,
 )
 
@@ -173,7 +174,8 @@ def encode(
     similarity_fn_name="cosine",  # assumed
     framework=["API"],
     use_instructions=True,
-    public_training_code=None,  # assumed
+    public_training_code=None,
+    public_training_data=None,  # assumed
     training_datasets=None,
 )
 
@@ -195,6 +197,7 @@ def encode(
     similarity_fn_name="cosine",  # assumed
     framework=["API"],
     use_instructions=True,
-    public_training_code=None,  # assumed
+    public_training_code=None,
+    public_training_data=None,  # assumed
     training_datasets=None,
 )
diff --git a/mteb/models/gritlm_models.py b/mteb/models/gritlm_models.py
index a68502b06d..d15c1f4a55 100644
--- a/mteb/models/gritlm_models.py
+++ b/mteb/models/gritlm_models.py
@@ -11,6 +11,18 @@
 logger = logging.getLogger(__name__)
 
 
+GRIT_LM_TRAINING_DATA = {
+    **E5_TRAINING_DATA,  # source https://arxiv.org/pdf/2402.09906
+    # also uses medi2 which contains fever and hotpotqa:
+    "FEVER": ["train"],
+    "FEVERHardNegatives": ["train"],
+    "FEVER-PL": ["train"],  # translation not trained on
+    "HotpotQA": ["train"],
+    "HotpotQAHardNegatives": ["train"],
+    "HotpotQA-PL": ["train"],  # translation not trained on
+}
+
+
 def gritlm_instruction(instruction: str = "", prompt_type=None) -> str:
     return (
         "<|user|>\n" + instruction + "\n<|embed|>\n" if instruction else "<|embed|>\n"
@@ -38,9 +50,10 @@ def gritlm_instruction(instruction: str = "", prompt_type=None) -> str:
     similarity_fn_name="cosine",
     framework=["GritLM", "PyTorch"],
     use_instructions=True,
-    training_datasets=E5_TRAINING_DATA,  # source https://arxiv.org/pdf/2402.09906
+    training_datasets=GRIT_LM_TRAINING_DATA,
     # section 3.1 "We finetune our final models from Mistral 7B [68] and Mixtral 8x7B [69] using adaptations of E5 [160] and the Tülu 2 data
     public_training_code="https://github.com/ContextualAI/gritlm",
+    public_training_data=None,
 )
 gritlm8x7b = ModelMeta(
     loader=partial(  # type: ignore
@@ -63,7 +76,8 @@ def gritlm_instruction(instruction: str = "", prompt_type=None) -> str:
     similarity_fn_name="cosine",
     framework=["GritLM", "PyTorch"],
     use_instructions=True,
-    training_datasets=E5_TRAINING_DATA,  # source https://arxiv.org/pdf/2402.09906
+    training_datasets=GRIT_LM_TRAINING_DATA,
     # section 3.1 "We finetune our final models from Mistral 7B [68] and Mixtral 8x7B [69] using adaptations of E5 [160] and the Tülu 2 data
     public_training_code="https://github.com/ContextualAI/gritlm",
+    public_training_data=None,
 )
diff --git a/mteb/models/gte_models.py b/mteb/models/gte_models.py
index da265e79c2..4de4b610f2 100644
--- a/mteb/models/gte_models.py
+++ b/mteb/models/gte_models.py
@@ -46,6 +46,7 @@ def instruction_template(
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
     max_tokens=131072,
 )
@@ -77,6 +78,7 @@ def instruction_template(
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -107,6 +109,7 @@ def instruction_template(
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -130,6 +133,7 @@ def instruction_template(
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,  # Not disclosed
 )
 
@@ -153,6 +157,7 @@ def instruction_template(
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,  # Not disclosed
 )
 
@@ -176,6 +181,7 @@ def instruction_template(
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,  # Not disclosed
 )
 
@@ -291,6 +297,7 @@ def instruction_template(
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
-    public_training_code=None,  # couldn't find
+    public_training_code=None,
+    public_training_data=None,  # couldn't find
     training_datasets=gte_multi_training_data,
 )
diff --git a/mteb/models/ibm_granite_models.py b/mteb/models/ibm_granite_models.py
index 78bad6097f..63679879c2 100644
--- a/mteb/models/ibm_granite_models.py
+++ b/mteb/models/ibm_granite_models.py
@@ -42,6 +42,7 @@
     adapted_from=None,
     superseded_by=None,
     public_training_code=None,
+    public_training_data=None,
     use_instructions=False,
     training_datasets=None,
 )
@@ -67,6 +68,7 @@
     adapted_from=None,
     superseded_by=None,
     public_training_code=None,
+    public_training_data=None,
     use_instructions=False,
     training_datasets=None,
 )
@@ -92,6 +94,7 @@
     adapted_from=None,
     superseded_by=None,
     public_training_code=None,
+    public_training_data=None,
     use_instructions=False,
     training_datasets=None,
 )
@@ -117,6 +120,7 @@
     adapted_from=None,
     superseded_by=None,
     public_training_code=None,
+    public_training_data=None,
     use_instructions=False,
     training_datasets=None,
 )
diff --git a/mteb/models/inf_models.py b/mteb/models/inf_models.py
index dc31adccd2..0d40ff3ef2 100644
--- a/mteb/models/inf_models.py
+++ b/mteb/models/inf_models.py
@@ -26,5 +26,6 @@
     use_instructions=True,
     adapted_from="Alibaba-NLP/gte-Qwen2-7B-instruct",
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
diff --git a/mteb/models/jasper_models.py b/mteb/models/jasper_models.py
index 1dc06d5640..dbd1615ad8 100644
--- a/mteb/models/jasper_models.py
+++ b/mteb/models/jasper_models.py
@@ -93,4 +93,5 @@ def encode(
     training_datasets=nvidia_training_datasets,  #  "In jasper model the teacher model is nvidia/NV-Embed-v2", source https://huggingface.co/infgrad/jasper_en_vision_language_v1
     # "non_mteb": ["BAAI/Infinity-MM", "HuggingFaceFW/fineweb-edu"],
     public_training_code=None,
+    public_training_data=None,
 )
diff --git a/mteb/models/jina_models.py b/mteb/models/jina_models.py
index 265d512371..4f1b58a352 100644
--- a/mteb/models/jina_models.py
+++ b/mteb/models/jina_models.py
@@ -224,6 +224,7 @@ def encode(
     reference="https://huggingface.co/jinaai/jina-embeddings-v3",
     training_datasets=None,
     public_training_code=None,
+    public_training_data=None,
 )
 
 
@@ -245,6 +246,7 @@ def encode(
     adapted_from=None,
     training_datasets=None,
     public_training_code=None,
+    public_training_data=None,
 )
 
 jina_embeddings_v2_small_en = ModelMeta(
@@ -265,6 +267,7 @@ def encode(
     adapted_from=None,
     training_datasets=None,
     public_training_code=None,
+    public_training_data=None,
 )
 
 jina_embedding_b_en_v1 = ModelMeta(
@@ -285,6 +288,7 @@ def encode(
     adapted_from=None,
     training_datasets=None,
     public_training_code=None,
+    public_training_data=None,
 )
 
 jina_embedding_s_en_v1 = ModelMeta(
@@ -305,4 +309,5 @@ def encode(
     adapted_from=None,
     training_datasets=None,
     public_training_code=None,
+    public_training_data=None,
 )
diff --git a/mteb/models/linq_models.py b/mteb/models/linq_models.py
index 11cfa74ed1..ead10ebf71 100644
--- a/mteb/models/linq_models.py
+++ b/mteb/models/linq_models.py
@@ -40,5 +40,6 @@ def instruction_template(
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
diff --git a/mteb/models/llm2vec_models.py b/mteb/models/llm2vec_models.py
index a5f1a69a36..28197e5c84 100644
--- a/mteb/models/llm2vec_models.py
+++ b/mteb/models/llm2vec_models.py
@@ -126,6 +126,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     use_instructions=True,
     public_training_code="https://github.com/McGill-NLP/llm2vec/tree/250292a307428240d801fadd85825464e71c3277/train_configs",
     training_datasets=llm2vec_supervised_training_data,
+    public_training_data=None,
 )
 
 llm2vec_llama3_8b_unsupervised = ModelMeta(
@@ -151,6 +152,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     use_instructions=True,
     public_training_code="https://github.com/McGill-NLP/llm2vec/tree/250292a307428240d801fadd85825464e71c3277/train_configs",
     training_datasets={},
+    public_training_data=None,
 )
 
 
@@ -177,6 +179,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     use_instructions=True,
     public_training_code="https://github.com/McGill-NLP/llm2vec/tree/250292a307428240d801fadd85825464e71c3277/train_configs",
     training_datasets=llm2vec_supervised_training_data,
+    public_training_data=None,
 )
 
 llm2vec_mistral7b_unsupervised = ModelMeta(
@@ -202,6 +205,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     use_instructions=True,
     public_training_code="https://github.com/McGill-NLP/llm2vec/tree/250292a307428240d801fadd85825464e71c3277/train_configs",
     training_datasets={},
+    public_training_data=None,
 )
 
 llm2vec_llama2_7b_supervised = ModelMeta(
@@ -227,6 +231,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     use_instructions=True,
     public_training_code="https://github.com/McGill-NLP/llm2vec/tree/250292a307428240d801fadd85825464e71c3277/train_configs",
     training_datasets=llm2vec_supervised_training_data,
+    public_training_data=None,
 )
 
 llm2vec_llama2_7b_unsupervised = ModelMeta(
@@ -252,6 +257,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     use_instructions=True,
     public_training_code="https://github.com/McGill-NLP/llm2vec/tree/250292a307428240d801fadd85825464e71c3277/train_configs",
     training_datasets={},
+    public_training_data=None,
 )
 
 llm2vec_sheared_llama_supervised = ModelMeta(
@@ -277,6 +283,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     use_instructions=True,
     public_training_code="https://github.com/McGill-NLP/llm2vec/tree/250292a307428240d801fadd85825464e71c3277/train_configs",
     training_datasets=llm2vec_supervised_training_data,
+    public_training_data=None,
 )
 
 llm2vec_sheared_llama_unsupervised = ModelMeta(
@@ -302,4 +309,5 @@ def loader_inner(**kwargs: Any) -> Encoder:
     use_instructions=True,
     public_training_code="https://github.com/McGill-NLP/llm2vec/tree/250292a307428240d801fadd85825464e71c3277/train_configs",
     training_datasets={},
+    public_training_data=None,
 )
diff --git a/mteb/models/misc_models.py b/mteb/models/misc_models.py
index 5233ecec6b..ba6e3e8163 100644
--- a/mteb/models/misc_models.py
+++ b/mteb/models/misc_models.py
@@ -22,6 +22,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/Haon-Chen/speed-embedding-7b-instruct",
     similarity_fn_name="cosine",
@@ -42,6 +43,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Gameselo/STS-multilingual-mpnet-base-v2",
     similarity_fn_name="cosine",
@@ -62,6 +64,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1",
     similarity_fn_name="cosine",
@@ -82,6 +85,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/HIT-TMG/KaLM-embedding-multilingual-mini-v1",
     similarity_fn_name="cosine",
@@ -102,6 +106,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/Hum-Works/lodestone-base-4096-v1",
     similarity_fn_name="cosine",
@@ -164,6 +169,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Jaume/gemma-2b-embeddings",
     similarity_fn_name="cosine",
@@ -184,6 +190,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/BeastyZ/e5-R-mistral-7b",
     similarity_fn_name="cosine",
@@ -210,6 +217,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Lajavaness/bilingual-embedding-base",
     similarity_fn_name="cosine",
@@ -235,6 +243,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Lajavaness/bilingual-embedding-large",
     similarity_fn_name="cosine",
@@ -260,6 +269,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Lajavaness/bilingual-embedding-small",
     similarity_fn_name="cosine",
@@ -280,6 +290,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Mihaiii/Bulbasaur",
     similarity_fn_name="cosine",
@@ -301,6 +312,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Mihaiii/Ivysaur",
     similarity_fn_name="cosine",
@@ -322,6 +334,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Mihaiii/Squirtle",
     similarity_fn_name="cosine",
@@ -343,6 +356,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Mihaiii/Venusaur",
     similarity_fn_name="cosine",
@@ -364,6 +378,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Mihaiii/Wartortle",
     similarity_fn_name="cosine",
@@ -385,6 +400,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Mihaiii/gte-micro",
     similarity_fn_name="cosine",
@@ -405,6 +421,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Mihaiii/gte-micro-v4",
     similarity_fn_name="cosine",
@@ -425,6 +442,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/OrdalieTech/Solon-embeddings-large-0.1",
     similarity_fn_name="cosine",
@@ -445,6 +463,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka",
     similarity_fn_name="cosine",
@@ -465,6 +484,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Omartificial-Intelligence-Space/Arabic-MiniLM-L12-v2-all-nli-triplet",
     similarity_fn_name="cosine",
@@ -487,6 +507,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Omartificial-Intelligence-Space/Arabic-all-nli-triplet-Matryoshka",
     similarity_fn_name="cosine",
@@ -509,6 +530,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Omartificial-Intelligence-Space/Arabic-labse-Matryoshka",
     similarity_fn_name="cosine",
@@ -531,6 +553,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Omartificial-Intelligence-Space/Arabic-mpnet-base-all-nli-triplet",
     similarity_fn_name="cosine",
@@ -553,6 +576,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka",
     similarity_fn_name="cosine",
@@ -573,6 +597,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/consciousAI/cai-lunaris-text-embeddings",
     similarity_fn_name="cosine",
@@ -593,6 +618,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/consciousAI/cai-stellaris-text-embeddings",
     similarity_fn_name="cosine",
@@ -613,6 +639,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/manu/bge-m3-custom-fr",
     similarity_fn_name="cosine",
@@ -633,6 +660,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/manu/sentence_croissant_alpha_v0.2",
     similarity_fn_name="cosine",
@@ -653,6 +681,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/manu/sentence_croissant_alpha_v0.3",
     similarity_fn_name="cosine",
@@ -673,6 +702,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/manu/sentence_croissant_alpha_v0.4",
     similarity_fn_name="cosine",
@@ -694,6 +724,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/thenlper/gte-base",
     similarity_fn_name="cosine",
@@ -714,6 +745,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/thenlper/gte-large",
     similarity_fn_name="cosine",
@@ -734,6 +766,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/thenlper/gte-small",
     similarity_fn_name="cosine",
@@ -754,6 +787,7 @@
     license="gpl-3.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/OrlikB/KartonBERT-USE-base-v1",
     similarity_fn_name="cosine",
@@ -774,6 +808,7 @@
     license="lgpl",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/OrlikB/st-polish-kartonberta-base-alpha-v1",
     similarity_fn_name="cosine",
@@ -794,6 +829,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/sdadas/mmlw-e5-base",
     similarity_fn_name="cosine",
@@ -814,6 +850,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/dwzhu/e5-base-4k",
     similarity_fn_name="cosine",
@@ -834,6 +871,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/sdadas/mmlw-e5-large",
     similarity_fn_name="cosine",
@@ -854,6 +892,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/sdadas/mmlw-e5-small",
     similarity_fn_name="cosine",
@@ -874,6 +913,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/sdadas/mmlw-roberta-base",
     similarity_fn_name="cosine",
@@ -894,6 +934,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/sdadas/mmlw-roberta-large",
     similarity_fn_name="cosine",
@@ -960,6 +1001,7 @@
     license="bigscience-bloom-rail-1.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/izhx/udever-bloom-1b1",
     similarity_fn_name="cosine",
@@ -1026,6 +1068,7 @@
     license="bigscience-bloom-rail-1.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/izhx/udever-bloom-3b",
     similarity_fn_name="cosine",
@@ -1092,6 +1135,7 @@
     license="bigscience-bloom-rail-1.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/izhx/udever-bloom-560m",
     similarity_fn_name="cosine",
@@ -1158,6 +1202,7 @@
     license="bigscience-bloom-rail-1.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/izhx/udever-bloom-7b1",
     similarity_fn_name="cosine",
@@ -1178,6 +1223,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/avsolatorio/GIST-Embedding-v0",
     similarity_fn_name="cosine",
@@ -1198,6 +1244,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/avsolatorio/GIST-all-MiniLM-L6-v2",
     similarity_fn_name="cosine",
@@ -1218,6 +1265,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/avsolatorio/GIST-large-Embedding-v0",
     similarity_fn_name="cosine",
@@ -1238,6 +1286,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/avsolatorio/GIST-small-Embedding-v0",
     similarity_fn_name="cosine",
@@ -1258,6 +1307,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/bigscience/sgpt-bloom-7b1-msmarco",
     similarity_fn_name="cosine",
@@ -1278,6 +1328,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/aari1995/German_Semantic_STS_V2",
     similarity_fn_name="cosine",
@@ -1299,6 +1350,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/abhinand/MedEmbed-small-v0.1",
     similarity_fn_name="cosine",
@@ -1325,6 +1377,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/avsolatorio/NoInstruct-small-Embedding-v0",
     similarity_fn_name="cosine",
@@ -1345,6 +1398,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/brahmairesearch/slx-v0.1",
     similarity_fn_name="cosine",
@@ -1365,6 +1419,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/deepfile/embedder-100p",
     similarity_fn_name="cosine",
@@ -1385,6 +1440,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/deepvk/USER-bge-m3",
     similarity_fn_name="cosine",
@@ -1416,6 +1472,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/infgrad/stella-base-en-v2",
     similarity_fn_name="cosine",
@@ -1436,6 +1493,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/malenia1/ternary-weight-embedding",
     similarity_fn_name="cosine",
@@ -1456,6 +1514,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/omarelshehy/arabic-english-sts-matryoshka",
     similarity_fn_name="cosine",
@@ -1486,6 +1545,7 @@
     license=None,
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/openbmb/MiniCPM-Embedding",
     similarity_fn_name="cosine",
@@ -1516,6 +1576,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/shibing624/text2vec-base-multilingual",
     similarity_fn_name="cosine",
@@ -1537,6 +1598,7 @@
     license="apache-2.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/silma-ai/silma-embeddding-matryoshka-v0.1",
     similarity_fn_name="cosine",
@@ -1557,6 +1619,7 @@
     license="mit",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch"],
     reference="https://huggingface.co/zeta-alpha-ai/Zeta-Alpha-E5-Mistral",
     similarity_fn_name="cosine",
@@ -1577,6 +1640,7 @@
     license="apache-2",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/DMetaSoul/sbert-chinese-general-v1",
     similarity_fn_name="cosine",
@@ -1601,6 +1665,7 @@
     license="apache-2",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/DMetaSoul/Dmeta-embedding-zh-small/",
     similarity_fn_name="cosine",
@@ -1620,6 +1685,7 @@
     license="not specified",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/lier007/xiaobu-embedding",
     similarity_fn_name="cosine",
@@ -1640,6 +1706,7 @@
     license="not specified",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/lier007/xiaobu-embedding-v2",
     similarity_fn_name="cosine",
@@ -1660,6 +1727,7 @@
     license="not specified",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Classical/Yinka",
     similarity_fn_name="cosine",
@@ -1680,6 +1748,7 @@
     license="cc-by-nc-4.0",
     open_weights=True,
     public_training_code=None,
+    public_training_data=None,
     framework=["PyTorch", "Sentence Transformers"],
     reference="https://huggingface.co/Classical/Yinka",
     similarity_fn_name="cosine",
diff --git a/mteb/models/model2vec_models.py b/mteb/models/model2vec_models.py
index afbf9df627..33da211c7a 100644
--- a/mteb/models/model2vec_models.py
+++ b/mteb/models/model2vec_models.py
@@ -75,7 +75,8 @@ def encode(
     adapted_from="BAAI/bge-base-en-v1.5",
     superseded_by=None,
     training_datasets=bge_training_data,  # distilled
-    public_training_code="https://github.com/MinishLab/model2vec",  #
+    public_training_code="https://github.com/MinishLab/model2vec",
+    public_training_data=None,
 )
 
 
@@ -101,6 +102,7 @@ def encode(
     superseded_by=None,
     training_datasets=bge_training_data,  # distilled
     public_training_code="https://github.com/MinishLab/model2vec",
+    public_training_data=None,
 )
 
 m2v_base_output = ModelMeta(
@@ -125,6 +127,7 @@ def encode(
     superseded_by=None,
     training_datasets=bge_training_data,  # distilled
     public_training_code="https://github.com/MinishLab/model2vec",
+    public_training_data=None,
 )
 
 m2v_multilingual_output = ModelMeta(
@@ -149,6 +152,7 @@ def encode(
     superseded_by=None,
     training_datasets=None,
     public_training_code="https://github.com/MinishLab/model2vec",
+    public_training_data=None,
 )
 
 potion_base_2m = ModelMeta(
@@ -173,6 +177,7 @@ def encode(
     superseded_by=None,
     training_datasets=bge_training_data,  # distilled
     public_training_code="https://github.com/MinishLab/model2vec",
+    public_training_data=None,
 )
 
 potion_base_4m = ModelMeta(
@@ -197,6 +202,7 @@ def encode(
     superseded_by=None,
     training_datasets=bge_training_data,  # distilled
     public_training_code="https://github.com/MinishLab/model2vec",
+    public_training_data=None,
 )
 
 potion_base_8m = ModelMeta(
@@ -221,4 +227,5 @@ def encode(
     superseded_by=None,
     training_datasets=bge_training_data,  # distilled
     public_training_code="https://github.com/MinishLab/model2vec",
+    public_training_data=None,
 )
diff --git a/mteb/models/moka_models.py b/mteb/models/moka_models.py
index d3943d78d7..1504b40789 100644
--- a/mteb/models/moka_models.py
+++ b/mteb/models/moka_models.py
@@ -96,7 +96,8 @@
     use_instructions=False,
     superseded_by=None,
     adapted_from=None,
-    public_training_code=None,  # Not published
+    public_training_code=None,
+    public_training_data=None,  # Not published
     training_datasets=m3e_dataset,
 )
 
@@ -117,7 +118,8 @@
     use_instructions=False,
     superseded_by=None,
     adapted_from=None,
-    public_training_code=None,  # Not published
+    public_training_code=None,
+    public_training_data=None,  # Not published
     training_datasets=m3e_dataset,
 )
 
@@ -139,6 +141,7 @@
     use_instructions=False,
     superseded_by=None,
     adapted_from=None,
-    public_training_code=None,  # Not published
+    public_training_code=None,
+    public_training_data=None,  # Not published
     training_datasets=m3e_dataset,
 )
diff --git a/mteb/models/mxbai_models.py b/mteb/models/mxbai_models.py
index 04978a190d..921db17871 100644
--- a/mteb/models/mxbai_models.py
+++ b/mteb/models/mxbai_models.py
@@ -27,5 +27,6 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
diff --git a/mteb/models/no_instruct_sentence_models.py b/mteb/models/no_instruct_sentence_models.py
index a0596b9bd1..9ff5cf901f 100644
--- a/mteb/models/no_instruct_sentence_models.py
+++ b/mteb/models/no_instruct_sentence_models.py
@@ -100,5 +100,6 @@ def encode(  # type: ignore
     adapted_from=None,
     superseded_by=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
diff --git a/mteb/models/nomic_models.py b/mteb/models/nomic_models.py
index 5d9da7b596..772d92902d 100644
--- a/mteb/models/nomic_models.py
+++ b/mteb/models/nomic_models.py
@@ -127,6 +127,7 @@ def encode(  # type: ignore
     adapted_from=None,
     superseded_by=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -154,6 +155,7 @@ def encode(  # type: ignore
     adapted_from=None,
     superseded_by="nomic-ai/nomic-embed-text-v1.5",
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -181,6 +183,7 @@ def encode(  # type: ignore
     adapted_from=None,
     superseded_by=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -209,6 +212,7 @@ def encode(  # type: ignore
     adapted_from=None,
     superseded_by=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -238,5 +242,6 @@ def encode(  # type: ignore
     adapted_from=None,
     superseded_by=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
diff --git a/mteb/models/nvidia_models.py b/mteb/models/nvidia_models.py
index 1f345a62be..1997a85274 100644
--- a/mteb/models/nvidia_models.py
+++ b/mteb/models/nvidia_models.py
@@ -141,6 +141,7 @@ def encode(
     use_instructions=True,
     training_datasets=nvidia_training_datasets,
     public_training_code=None,
+    public_training_data=None,
 )
 
 NV_embed_v1 = ModelMeta(
@@ -164,4 +165,5 @@ def encode(
     use_instructions=True,
     training_datasets=nvidia_training_datasets,
     public_training_code=None,
+    public_training_data=None,
 )
diff --git a/mteb/models/openai_models.py b/mteb/models/openai_models.py
index 863c9d7828..079e7c9361 100644
--- a/mteb/models/openai_models.py
+++ b/mteb/models/openai_models.py
@@ -135,7 +135,8 @@ def _to_numpy(self, embedding_response) -> np.ndarray:
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=False,
-    public_training_code=None,  # assumed
+    public_training_code=None,
+    public_training_data=None,  # assumed
     training_datasets=None,
 )
 text_embedding_3_large = ModelMeta(
@@ -156,7 +157,8 @@ def _to_numpy(self, embedding_response) -> np.ndarray:
     framework=["API"],
     use_instructions=False,
     n_parameters=None,
-    public_training_code=None,  # assumed
+    public_training_code=None,
+    public_training_data=None,  # assumed
     training_datasets=None,
     license=None,
     similarity_fn_name=None,
@@ -179,7 +181,8 @@ def _to_numpy(self, embedding_response) -> np.ndarray:
     framework=["API"],
     use_instructions=False,
     n_parameters=None,
-    public_training_code=None,  # assumed
+    public_training_code=None,
+    public_training_data=None,  # assumed
     training_datasets=None,
     license=None,
     similarity_fn_name=None,
diff --git a/mteb/models/overview.py b/mteb/models/overview.py
index ad93efb314..e444b1105f 100644
--- a/mteb/models/overview.py
+++ b/mteb/models/overview.py
@@ -226,6 +226,7 @@ def model_meta_from_hf_hub(model_name: str) -> ModelMeta:
             embed_dim=None,
             open_weights=True,
             public_training_code=None,
+            public_training_data=None,
             use_instructions=None,
         )
     except Exception as e:
@@ -241,6 +242,7 @@ def model_meta_from_hf_hub(model_name: str) -> ModelMeta:
             license=None,
             open_weights=True,
             public_training_code=None,
+            public_training_data=None,
             similarity_fn_name=None,
             use_instructions=None,
             training_datasets=None,
@@ -273,6 +275,7 @@ def model_meta_from_sentence_transformers(model: SentenceTransformer) -> ModelMe
             license=None,
             open_weights=True,
             public_training_code=None,
+            public_training_data=None,
             use_instructions=None,
             training_datasets=None,
         )
@@ -291,6 +294,7 @@ def model_meta_from_sentence_transformers(model: SentenceTransformer) -> ModelMe
             license=None,
             open_weights=True,
             public_training_code=None,
+            public_training_data=None,
             similarity_fn_name=None,
             use_instructions=None,
             training_datasets=None,
diff --git a/mteb/models/piccolo_models.py b/mteb/models/piccolo_models.py
index bb92b55673..d51487b8ba 100644
--- a/mteb/models/piccolo_models.py
+++ b/mteb/models/piccolo_models.py
@@ -21,6 +21,7 @@
     superseded_by=None,
     adapted_from=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,  # They don't specify
 )
 
@@ -42,5 +43,6 @@
     superseded_by=None,
     adapted_from=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,  # They don't say
 )
diff --git a/mteb/models/promptriever_models.py b/mteb/models/promptriever_models.py
index a7066817a4..287fd3ef91 100644
--- a/mteb/models/promptriever_models.py
+++ b/mteb/models/promptriever_models.py
@@ -65,6 +65,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     framework=["PyTorch", "Tevatron"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
 )
 
 promptriever_llama3 = ModelMeta(
@@ -90,6 +91,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     framework=["PyTorch", "Tevatron"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
 )
 
 
@@ -116,6 +118,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     framework=["PyTorch", "Tevatron"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
 )
 
 promptriever_mistral_v1 = ModelMeta(
@@ -141,4 +144,5 @@ def loader_inner(**kwargs: Any) -> Encoder:
     framework=["PyTorch", "Tevatron"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
 )
diff --git a/mteb/models/repllama_models.py b/mteb/models/repllama_models.py
index 5ae4c0d8cb..2c5ef6e446 100644
--- a/mteb/models/repllama_models.py
+++ b/mteb/models/repllama_models.py
@@ -150,6 +150,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     framework=["PyTorch", "Tevatron"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
 )
 
 
@@ -176,5 +177,6 @@ def loader_inner(**kwargs: Any) -> Encoder:
     framework=["PyTorch", "Tevatron"],
     use_instructions=True,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
diff --git a/mteb/models/rerankers_custom.py b/mteb/models/rerankers_custom.py
index 5609fdf83a..1a0fd1f6ba 100644
--- a/mteb/models/rerankers_custom.py
+++ b/mteb/models/rerankers_custom.py
@@ -209,6 +209,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
@@ -233,6 +234,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
@@ -289,6 +291,7 @@ def loader_inner(**kwargs: Any) -> Encoder:
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
diff --git a/mteb/models/rerankers_monot5_based.py b/mteb/models/rerankers_monot5_based.py
index 5bc50bad70..c53b364000 100644
--- a/mteb/models/rerankers_monot5_based.py
+++ b/mteb/models/rerankers_monot5_based.py
@@ -301,6 +301,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
@@ -324,6 +325,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
@@ -347,6 +349,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
@@ -370,6 +373,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
@@ -405,6 +409,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     framework=["PyTorch"],
@@ -439,6 +444,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     framework=["PyTorch"],
@@ -473,6 +479,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     framework=["PyTorch"],
@@ -507,6 +514,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     framework=["PyTorch"],
@@ -530,6 +538,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
@@ -553,6 +562,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
@@ -576,6 +586,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
@@ -600,6 +611,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     framework=["PyTorch"],
@@ -728,6 +740,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     framework=["PyTorch"],
@@ -750,6 +763,7 @@ def get_prediction_tokens(self, *args, **kwargs):
     embed_dim=None,
     license=None,
     public_training_code=None,
+    public_training_data=None,
     similarity_fn_name=None,
     use_instructions=None,
     training_datasets=None,
diff --git a/mteb/models/ru_sentence_models.py b/mteb/models/ru_sentence_models.py
index d8c7e84518..297c7f3142 100644
--- a/mteb/models/ru_sentence_models.py
+++ b/mteb/models/ru_sentence_models.py
@@ -23,6 +23,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -41,6 +42,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -59,6 +61,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -77,6 +80,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -124,6 +128,7 @@
         # "CarlBrendt/Summ_Dialog_News": ["train"],
     },
     public_training_code=None,
+    public_training_data=None,
 )
 
 deberta_v1_ru = ModelMeta(
@@ -141,6 +146,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -159,6 +165,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -177,6 +184,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -195,6 +203,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -213,6 +222,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
 )
 
@@ -231,6 +241,7 @@
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=False,
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,  # source model in unknown
     # Not MTEB: {"IlyaGusev/gazeta": ["train"], "zloelias/lenta-ru": ["train"]},
 )
@@ -252,6 +263,7 @@
     training_datasets=None,  # source model in unknown
     # not MTEB: {"IlyaGusev/gazeta": ["train"], "zloelias/lenta-ru": ["train"]},
     public_training_code=None,
+    public_training_data=None,
 )
 
 
@@ -279,6 +291,7 @@
     license="mit",
     similarity_fn_name="cosine",
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,
     framework=["Sentence Transformers", "PyTorch"],
 )
diff --git a/mteb/models/salesforce_models.py b/mteb/models/salesforce_models.py
index 4d4a60b621..235057a6f8 100644
--- a/mteb/models/salesforce_models.py
+++ b/mteb/models/salesforce_models.py
@@ -6,6 +6,8 @@
 from mteb.model_meta import ModelMeta
 from mteb.models.instruct_wrapper import instruct_wrapper
 
+from .e5_instruct import E5_MISTRAL_TRAINING_DATA
+
 
 def instruction_template(
     instruction: str, prompt_type: PromptType | None = None
@@ -13,6 +15,19 @@ def instruction_template(
     return f"Instruct: {instruction}\nQuery: " if instruction else ""
 
 
+SFR_TRAINING_DATA = {  # inherits from e5
+    **E5_MISTRAL_TRAINING_DATA,
+    # From previously released blogpost which now have been taken down:
+    "FiQA2018": ["train"],
+    "FiQA2018-PL": ["train"],
+    "FEVER": ["train"],
+    "FEVERHardNegatives": ["train"],
+    "FEVER-PL": ["train"],  # translation not trained on
+    "HotpotQA": ["train"],
+    "HotpotQAHardNegatives": ["train"],
+    "HotpotQA-PL": ["train"],  # translation not trained on
+}
+
 SFR_Embedding_2_R = ModelMeta(
     loader=partial(  # type: ignore
         instruct_wrapper,
@@ -41,16 +56,8 @@ def instruction_template(
     use_instructions=True,
     adapted_from="intfloat/e5-mistral-7b-instruct",
     public_training_code=None,
-    training_datasets={  # inherits from e5
-        "MSMARCO": ["train"],
-        "MSMARCOHardNegatives": ["train"],
-        "NanoMSMARCORetrieval": ["train"],
-        "MSMARCO-PL": ["train"],  # translation not trained on
-        "NQ": ["train"],
-        "NQHardNegatives": ["train"],
-        "NanoNQRetrieval": ["train"],
-        "NQ-PL": ["train"],  # translation not trained on
-    },
+    public_training_data=None,
+    training_datasets=SFR_TRAINING_DATA,
 )
 
 
@@ -79,14 +86,6 @@ def instruction_template(
     framework=["Sentence Transformers", "PyTorch"],
     use_instructions=True,
     public_training_code=None,
-    training_datasets={  # inherits from e5
-        "MSMARCO": ["train"],
-        "MSMARCOHardNegatives": ["train"],
-        "NanoMSMARCORetrieval": ["train"],
-        "MSMARCO-PL": ["train"],  # translation not trained on
-        "NQ": ["train"],
-        "NQHardNegatives": ["train"],
-        "NanoNQRetrieval": ["train"],
-        "NQ-PL": ["train"],  # translation not trained on
-    },
+    public_training_data=None,
+    training_datasets=SFR_TRAINING_DATA,
 )
diff --git a/mteb/models/sentence_transformers_models.py b/mteb/models/sentence_transformers_models.py
index fa48ae7ccc..eec65049d5 100644
--- a/mteb/models/sentence_transformers_models.py
+++ b/mteb/models/sentence_transformers_models.py
@@ -112,6 +112,7 @@
     adapted_from=None,
     training_datasets=sent_trf_training_dataset,
     public_training_code=None,
+    public_training_data=None,
 )
 
 all_MiniLM_L12_v2 = ModelMeta(
@@ -132,6 +133,7 @@
     adapted_from=None,
     training_datasets=sent_trf_training_dataset,
     public_training_code=None,
+    public_training_data=None,
 )
 
 paraphrase_multilingual_MiniLM_L12_v2 = ModelMeta(
@@ -152,6 +154,7 @@
     adapted_from=None,
     training_datasets=sent_trf_training_dataset,  # assumed (probably some parallel as well)
     public_training_code=None,
+    public_training_data=None,
 )
 
 paraphrase_multilingual_mpnet_base_v2 = ModelMeta(
@@ -183,6 +186,7 @@
     # "yahoo-answers": yahoo_answers_train_dataset,
     # "stack-exchange": stack_exchange_train_dataset,
     public_training_code=None,
+    public_training_data=None,
 )
 
 labse = ModelMeta(
@@ -203,6 +207,7 @@
     adapted_from=None,
     training_datasets=None,  # scraped and mined webdata including CC, wiki, see section 3.1 https://aclanthology.org/2022.acl-long.62.pdf
     public_training_code="https://www.kaggle.com/models/google/labse/tensorFlow2/labse/2?tfhub-redirect=true",
+    public_training_data=None,
 )
 
 multi_qa_MiniLM_L6_cos_v1 = ModelMeta(
@@ -223,6 +228,7 @@
     adapted_from="nreimers/MiniLM-L6-H384-uncased",
     training_datasets=sent_trf_training_dataset,  # assumed
     public_training_code=None,
+    public_training_data=None,
 )
 
 all_mpnet_base_v2 = ModelMeta(
@@ -243,6 +249,7 @@
     adapted_from=None,
     training_datasets=sent_trf_training_dataset,
     public_training_code=None,
+    public_training_data=None,
 )
 
 
@@ -274,4 +281,5 @@
         # "sentence-transformers/natural-questions": ["train"],
     },
     public_training_code=None,
+    public_training_data=None,
 )
diff --git a/mteb/models/stella_models.py b/mteb/models/stella_models.py
index 44aa1f8604..7210b287cb 100644
--- a/mteb/models/stella_models.py
+++ b/mteb/models/stella_models.py
@@ -30,6 +30,7 @@
     reference="https://huggingface.co/dunzhang/stella_en_400M_v5",
     training_datasets=None,
     public_training_code=None,
+    public_training_data=None,
 )
 
 stella_en_1_5b = ModelMeta(
@@ -56,6 +57,7 @@
     reference="https://huggingface.co/dunzhang/stella_en_1.5B_v5",
     training_datasets=None,
     public_training_code=None,
+    public_training_data=None,
 )
 
 stella_large_zh_v3_1792d = ModelMeta(
@@ -75,6 +77,7 @@
     superseded_by="dunzhang/stella-mrl-large-zh-v3.5-1792d",
     adapted_from=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets={
         # Not in MTEB:
         # - infgrad/dialogue_rewrite_llm
@@ -99,6 +102,7 @@
     superseded_by=None,
     adapted_from=None,
     public_training_code=None,
+    public_training_data=None,
     training_datasets={
         # Not in MTEB:
         # - infgrad/dialogue_rewrite_llm
@@ -124,6 +128,7 @@
     superseded_by=None,
     adapted_from="dunzhang/stella-large-zh-v3-1792d",
     public_training_code=None,
+    public_training_data=None,
     training_datasets=None,  # Not specified
 )
 
@@ -144,6 +149,7 @@
     superseded_by=None,
     adapted_from="dunzhang/stella-mrl-large-zh-v3.5-1792d",
     public_training_code=None,
+    public_training_data=None,
     training_datasets={
         # It's a bit unclear what they have trained on to be honest, because they don't list all
         # And they also have some rather cryptic description of their training procedure, but at
diff --git a/mteb/models/text2vec_models.py b/mteb/models/text2vec_models.py
index 12322e69e9..86a9bcca4f 100644
--- a/mteb/models/text2vec_models.py
+++ b/mteb/models/text2vec_models.py
@@ -21,7 +21,8 @@
     use_instructions=False,
     superseded_by=None,
     adapted_from=None,
-    public_training_code=None,  # Couldn't find it
+    public_training_code=None,
+    public_training_data=None,  # Couldn't find it
     training_datasets={
         # source: https://huggingface.co/shibing624/text2vec-base-chinese
         # Not in MTEB
@@ -46,7 +47,8 @@
     use_instructions=False,
     superseded_by=None,
     adapted_from=None,
-    public_training_code=None,  # Couldn't find it
+    public_training_code=None,
+    public_training_data=None,  # Couldn't find it
     training_datasets={
         # source: https://huggingface.co/shibing624/text2vec-base-chinese
         # Not in MTEB
@@ -87,7 +89,8 @@
     use_instructions=False,
     superseded_by=None,
     adapted_from="sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
-    public_training_code=None,  # Couldn't find it
+    public_training_code=None,
+    public_training_data=None,  # Couldn't find it
     training_datasets={
         # source: https://huggingface.co/shibing624/text2vec-base-chinese
         # Not in MTEB
diff --git a/mteb/models/uae_models.py b/mteb/models/uae_models.py
index bd8be48693..8d97703ef6 100644
--- a/mteb/models/uae_models.py
+++ b/mteb/models/uae_models.py
@@ -83,4 +83,5 @@ def encode(
         "SNLI": [],
     },
     public_training_code=None,
+    public_training_data=None,
 )
diff --git a/mteb/models/voyage_models.py b/mteb/models/voyage_models.py
index a98bc041bc..a637dee36a 100644
--- a/mteb/models/voyage_models.py
+++ b/mteb/models/voyage_models.py
@@ -12,6 +12,11 @@
 
 from .wrapper import Wrapper
 
+VOYAGE_TRAINING_DATA = {
+    # Self-reported (message from VoyageAI member)
+    # synthetic data
+}
+
 
 def token_limit(max_tpm: int, interval: int = 60):
     limit_interval_start_ts = time.time()
@@ -156,8 +161,9 @@ def _batched_encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    training_datasets=None,
+    training_datasets=VOYAGE_TRAINING_DATA,
     public_training_code=None,
+    public_training_data=None,
 )
 
 voyage_finance_2 = ModelMeta(
@@ -179,8 +185,9 @@ def _batched_encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    training_datasets=None,
+    training_datasets=VOYAGE_TRAINING_DATA,
     public_training_code=None,
+    public_training_data=None,
 )
 
 voyage_law_2 = ModelMeta(
@@ -202,8 +209,9 @@ def _batched_encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    training_datasets=None,
+    training_datasets=VOYAGE_TRAINING_DATA,
     public_training_code=None,
+    public_training_data=None,
 )
 
 voyage_code_2 = ModelMeta(
@@ -225,8 +233,9 @@ def _batched_encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    training_datasets=None,
+    training_datasets=VOYAGE_TRAINING_DATA,
     public_training_code=None,
+    public_training_data=None,
 )
 
 voyage_large_2 = ModelMeta(
@@ -248,8 +257,9 @@ def _batched_encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    training_datasets=None,
+    training_datasets=VOYAGE_TRAINING_DATA,
     public_training_code=None,
+    public_training_data=None,
 )
 
 voyage_2 = ModelMeta(
@@ -271,8 +281,9 @@ def _batched_encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    training_datasets=None,
+    training_datasets=VOYAGE_TRAINING_DATA,
     public_training_code=None,
+    public_training_data=None,
 )
 voyage_multilingual_2 = ModelMeta(
     name="voyageai/voyage-multilingual-2",
@@ -293,8 +304,9 @@ def _batched_encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    training_datasets=None,
+    training_datasets=VOYAGE_TRAINING_DATA,
     public_training_code=None,
+    public_training_data=None,
 )
 
 voyage_3 = ModelMeta(
@@ -316,8 +328,9 @@ def _batched_encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    training_datasets=None,
+    training_datasets=VOYAGE_TRAINING_DATA,
     public_training_code=None,
+    public_training_data=None,
 )
 
 voyage_3_lite = ModelMeta(
@@ -339,6 +352,79 @@ def _batched_encode(
     similarity_fn_name="cosine",
     framework=["API"],
     use_instructions=True,
-    training_datasets=None,
+    training_datasets=VOYAGE_TRAINING_DATA,
+    public_training_code=None,
+    public_training_data=None,
+)
+
+
+voyage_3_exp = ModelMeta(
+    name="voyageai/voyage-3-m-exp",
+    revision="1",
+    release_date=None,  # not released
+    languages=None,  # supported languages not specified
+    loader=partial(
+        VoyageWrapper,
+        model_name="voyage-3-m-exp",
+        model_prompts=model_prompts,
+    ),
+    max_tokens=32000,
+    embed_dim=512,
+    open_weights=False,
+    n_parameters=None,
+    license=None,
+    reference="https://huggingface.co/voyageai/voyage-3-m-exp",
+    similarity_fn_name="cosine",
+    framework=["API"],
+    use_instructions=True,
+    training_datasets={
+        # MTEB(eng, classic) training data:
+        "ArguAna": ["train"],
+        "ArguAna-PL": ["train"],
+        "NanoArguAnaRetrieval": ["train"],
+        "HotpotQA": ["train"],
+        "HotpotQA-PL": ["train"],  # translation not trained on
+        "HotpotQAHardNegatives": ["train"],
+        "MSMARCO": ["train"],
+        "MSMARCOHardNegatives": ["train"],
+        "NanoMSMARCORetrieval": ["train"],
+        "MSMARCO-PL": ["train"],  # translation not trained on
+        "NQ": ["train"],
+        "NQHardNegatives": ["train"],
+        "NanoNQRetrieval": ["train"],
+        "NQ-PL": ["train"],  # translation not trained on
+        "FEVER": ["train"],
+        "FEVERHardNegatives": ["train"],
+        "NanoFEVERRetrieval": ["train"],
+        "FiQA2018": ["train"],
+        "FiQA2018-PL": ["train"],  # translation not trained on
+        "STS12": ["train"],
+        "STS22": ["train"],
+        "AmazonReviewsClassification": ["train"],
+        "AmazonCounterfactualClassification": ["train"],
+        "Banking77Classification": ["train"],
+        "EmotionClassification": ["train"],
+        "ImdbClassification": ["train"],
+        "MTOPIntentClassification": ["train"],
+        "ToxicConversationsClassification": ["train"],
+        "TweetSentimentExtractionClassification": ["train"],
+        "ArxivClusteringP2P": ["train"],
+        "ArxivClusteringP2P.v2": ["train"],
+        "ArxivClusteringS2S": ["train"],
+        "ArxivClusteringS2S.v2": ["train"],
+        "BiorxivClusteringP2P": ["train"],
+        "BiorxivClusteringP2P.v2": ["train"],
+        "BiorxivClusteringS2S": ["train"],
+        "BiorxivClusteringS2S.v2": ["train"],
+        "MedrxivClusteringP2P": ["train"],
+        "MedrxivClusteringP2P.v2": ["train"],
+        "MedrxivClusteringS2S": ["train"],
+        "MedrxivClusteringS2S.v2": ["train"],
+        "TwentyNewsgroupsClustering": ["train"],
+        "TwentyNewsgroupsClustering.v2": ["train"],
+        "STSBenchmark": ["train"],
+        "STSBenchmarkMultilingualSTS": ["train"],  # translated, not trained on
+    },
     public_training_code=None,
+    public_training_data=None,
 )
diff --git a/scripts/generate_metadata.py b/scripts/generate_metadata.py
index a192fa1341..4ae87fdbca 100644
--- a/scripts/generate_metadata.py
+++ b/scripts/generate_metadata.py
@@ -242,6 +242,7 @@ def model_meta_from_hf_hub(model_name: str) -> ModelMeta:
             license=None,
             open_weights=True,
             public_training_code=None,
+            public_training_data=None,
             similarity_fn_name=None,
             use_instructions=None,
             training_datasets=None,
diff --git a/tests/test_tasks/test_mteb_rerank.py b/tests/test_tasks/test_mteb_rerank.py
index fb0cf6cf5a..dc65dae905 100644
--- a/tests/test_tasks/test_mteb_rerank.py
+++ b/tests/test_tasks/test_mteb_rerank.py
@@ -377,6 +377,7 @@ def test_reranker_same_ndcg1():
         embed_dim=None,
         license=None,
         public_training_code=None,
+        public_training_data=None,
         reference=None,
         similarity_fn_name=None,
         use_instructions=None,