Feature Extraction
sentence-transformers
Safetensors
Transformers
qwen3_pseudo_moe
sentence-similarity
custom_code
Instructions to use geevec-ai/geevec-embeddings-1.0-lite with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use geevec-ai/geevec-embeddings-1.0-lite with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("geevec-ai/geevec-embeddings-1.0-lite", trust_remote_code=True) sentences = [ "The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] - Transformers
How to use geevec-ai/geevec-embeddings-1.0-lite with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="geevec-ai/geevec-embeddings-1.0-lite", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("geevec-ai/geevec-embeddings-1.0-lite", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update modeling_qwen3_pseudo_moe.py
#2
by ctranslate2-4you - opened
Fix loading under Transformers 5 (_tied_weights_keys)
Loading fails on Transformers 5.x in post_init() with AttributeError: 'list' object has no attribute 'keys'. The cause is that _tied_weights_keys = [] uses the old list format while config.json sets tie_word_embeddings: true. This sets it to None, the PreTrainedModel default; nothing is tied, since the model has no LM head. It also replaces the deprecated config.use_return_dict with an equivalent expression. Tested with Transformers 4.51.3, 4.57.1 and 5.18.0: all 475 weights load and the embeddings are identical to the original code's.