Alibaba (Tongyi Lab) - Open weights - verified
GTE ModernBERT Base embeddings
GTE ModernBERT Base is a current embedding model from Alibaba (Tongyi Lab), with open weights on Hugging Face at Alibaba-NLP/gte-modernbert-base. It returns 768-dimension vectors by default and accepts up to 8,192 input tokens, for text input. There is no API price: the weights download from Hugging Face under Apache 2.0, and you pay for the compute you run them on.
CurrentModel id: Alibaba-NLP/gte-modernbert-base- weights: Alibaba-NLP/gte-modernbert-base
- Default dims
- 768
- Max input
- 8,192 tok
- Price / 1M
- Self-hosted
- Inputs
- Text
Dimensions
- 768default
Default output: 768 dimensions.
Pricing
No API price: the weights are on Hugging Face and you run them on your own hardware or host.
Open weights
- Hugging Face repo
- Alibaba-NLP/gte-modernbert-base
- License
- Apache 2.0
apache-2.0 - Commercial use
- Commercial use allowedCommercial use is allowed, per the license.
- Parameters
- 149M (149,014,272)
Worth knowing
English-only 149M ModernBERT embedding model with CLS pooling and an 8192-token input, also servable through Text Embeddings Inference with an OpenAI-compatible route.
Released . Multilingual: no.
Usage
Copied verbatim from the page linked below. Placeholders such as project ids and API keys are the source's own.
# Requires transformers>=4.48.0
from sentence_transformers import SentenceTransformer
from sentence_transformers.util import cos_sim
input_texts = [
"what is the capital of China?",
"how to implement quick sort in python?",
"Beijing",
"sorting algorithms"
]
model = SentenceTransformer("Alibaba-NLP/gte-modernbert-base")
embeddings = model.encode(input_texts)
print(embeddings.shape)
# (4, 768)
similarities = cos_sim(embeddings[0], embeddings[1:])FAQ
How many dimensions does GTE ModernBERT Base output?
Default output: 768 dimensions.
What is the max input length of GTE ModernBERT Base?
8,192 tokens per input, per the model card. Longer text has to be chunked before you embed it.
Can I use GTE ModernBERT Base commercially?
GTE ModernBERT Base is released under Apache 2.0, and commercial use is allowed. The license text is at https://www.apache.org/licenses/LICENSE-2.0.
What can GTE ModernBERT Base embed?
Text input, per the provider. It is not documented as multilingual.
Sources
Last verified from the pages below.
- Hugging Face model card - huggingface.co
- Hugging Face API (safetensors, cardData) - huggingface.co
Other open-weight embedding models
Default dimensions and max input tokens for the other models you can run yourself.
- Nemotron 3.5 Embed 8B4,096 dims - 32,768 tokens - Self-hosted
- Qwen3-Embedding-8B4,096 dims - 32,768 tokens - Self-hosted
- Qwen3-Embedding-4B2,560 dims - 32,768 tokens - Self-hosted
- EmbeddingGemma 2768 dims - 8,192 tokens - Self-hosted
- Qwen3-Embedding-0.6B1,024 dims - 32,768 tokens - Self-hosted
- Nomic Embed Text v2 MoE768 dims - 512 tokens - Self-hosted
- Voyage 4 Nano1,024 dims - 32,000 tokens - Self-hosted
- Granite Embedding 311M Multilingual R2768 dims - 32,768 tokens - Self-hosted
- BGE-M31,024 dims - 8,192 tokens - Self-hosted
All embedding models - Open-weight embedding models
Newsletter
New models, every week
One email per week. New model releases, embedding and retrieval updates and what they mean for what you build.
Keep exploring
Model API pricing and specs
First-party prices per 1M tokens for chat and reasoning models.
Local models
Open-weight LLMs you can download and run, with sizes and licenses.
What is RAG?
Where embeddings fit in a retrieval pipeline, end to end.
Vector databases compared
Where the vectors go once you have picked a model.