Model reference · open weights
GTE-ModernColBERT is an open-weight embedding model from lightonai, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
GTE-ModernColBERT-v1 Multi-vector embedding model based on Alibaba-NLP/gte-modernbert-base This is a multi-vector (ColBERT-style late interaction) embedding model trained on the ms-marco-en-bge-gemma dataset. It maps sentences & paragraphs to sequences of 128-dimensional dense vectors and can be used for semantic textual similarity using the MaxSim operator. Model Details Model Description - Model Type: Multi-vector embedding model - Base model: Alibaba-NLP/gte-modernbert-base <!-- at revision bc02f0a92d1b6dd82108036f6cb4b7b423fb7434 -- - Document Length: 300 tokens - Query Length: 32 tokens - Output Dimensionality: 128 dimensions - Similarity Function: MaxSim - Training Dataset: - ms-marco-en-bge-gemma - Language: English - License: Apache 2.0 Document length GTE-ModernColBERT has been trained with knowledge distillation on MS MARCO with a document length of 300 tokens, explaining its default value for documents length. However, as illustrated in the ModernBERT paper, ColBERT models can generalize to documents lengths way beyond their training length and GTE-ModernColBERT actually yields results way above SOTA in long-context embedding benchmarks, see LongEmbed results. Simply change adapt the document length parameter to your needs when loading the model: ModernBERT itself has only been trained on 8K context length, but it seems that GTE-ModernColBERT can generalize to even bigger context sizes, though it is not guaranteed so please perform your own benches! Model Sources - Documentation: PyLate Documentation - Repository: PyLate on GitHub - Hugging Face: PyLate models on Hugging Face Full Model Architecture Usage Sentence Transformers This model can be used with Sentence Transformers as a multi-vector (ColBERT-style late interaction) retriever via the MultiVectorEncoder: PyLate First install the PyLate library: Retrieval PyLate provides a streamlined interface to index and retrieve documents using ColBERT models. The index leverages the Voyager HNSW index to efficiently handle document embeddings and enable fast retrieval. Indexing documents First, load the ColBERT model and initialize the Voyager index, then encode and index your documents: Note that you do
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | lightonai |
|---|---|
| Type | Embedding models |
| Parameters (lead) | 149M |
| Context | 8k tokens |
| Variants | 1 |
| Runs with | PyLate |
| Based on | Alibaba-NLP/gte-modernbert-base |
| Released | 2025-04-30 |
| Popularity | 361k downloads / month |
| Likes | 176 |
| Licence | Open weights |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| GTE-ModernColBERT-v1 | 149M | BF16 | ~0.3 GB | ✓ | Weights ↗ |
Benchmarks
As published on the model card — the maker's own numbers, not measured by AxForge.
| Task | Dataset | Metric | Score |
|---|---|---|---|
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Accuracy@1 | 0.36 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Accuracy@3 | 0.62 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Accuracy@5 | 0.78 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Accuracy@10 | 0.86 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Precision@1 | 0.36 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Precision@3 | 0.233 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Precision@5 | 0.208 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Precision@10 | 0.128 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Recall@1 | 0.183 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Recall@3 | 0.289 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Recall@5 | 0.416 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Recall@10 | 0.496 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Ndcg@10 | 0.415 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Mrr@10 | 0.527 |
| Py Late Information Retrieval | NanoClimateFEVER | Maxsim Map@100 | 0.335 |
| Py Late Information Retrieval | NanoDBPedia | Maxsim Accuracy@1 | 0.88 |
| Py Late Information Retrieval | NanoDBPedia | Maxsim Accuracy@3 | 0.94 |
| Py Late Information Retrieval | NanoDBPedia | Maxsim Accuracy@5 | 0.96 |
| Py Late Information Retrieval | NanoDBPedia | Maxsim Accuracy@10 | 0.98 |
| Py Late Information Retrieval | NanoDBPedia | Maxsim Precision@1 | 0.88 |
| Py Late Information Retrieval | NanoDBPedia | Maxsim Precision@3 | 0.713 |
| Py Late Information Retrieval | NanoDBPedia | Maxsim Precision@5 | 0.656 |
| Py Late Information Retrieval | NanoDBPedia | Maxsim Precision@10 | 0.572 |
| Py Late Information Retrieval | NanoDBPedia | Maxsim Recall@1 | 0.118 |
Using it via the API
Once AxForge deploys gte-moderncolbert for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (gte-moderncolbert below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/embeddings \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"gte-moderncolbert","input":"text to embed"}'
Licence
Open weights under apache-2.0 — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗
Explore