Model reference · open weights
gte-large is an open-weight embedding model from thenlper, 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-large General Text Embeddings (GTE) model. Towards General Text Embeddings with Multi-stage Contrastive Learning The GTE models are trained by Alibaba DAMO Academy. They are mainly based on the BERT framework and currently offer three different sizes of models, including GTE-large, GTE-base, and GTE-small. The GTE models are trained on a large-scale corpus of relevance text pairs, covering a wide range of domains and scenarios. This enables the GTE models to be applied to various downstream tasks of text embeddings, including information retrieval, semantic textual similarity, text reranking, etc. Metrics We compared the performance of the GTE models with other popular text embedding models on the MTEB benchmark. For more detailed comparison results, please refer to the MTEB leaderboard. Usage Code example Use with sentence-transformers: Limitation This model exclusively caters to English texts, and any lengthy texts will be truncated to a maximum of 512 tokens. Citation If you find our paper or models helpful, please consider citing them as follows:
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | thenlper |
|---|---|
| Type | Embedding models |
| Parameters (lead) | 335M |
| Context | 512 tokens |
| Variants | 1 |
| Runs with | sentence-transformers |
| Released | 2023-07-27 |
| Popularity | 704k downloads / month |
| Likes | 306 |
| 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-large | 335M | BF16 | ~0.8 GB | ✓ | Weights ↗ |
Benchmarks
As published on the model card — the maker's own numbers, not measured by AxForge.
| Task | Dataset | Metric | Score |
|---|---|---|---|
| Classification | MTEB AmazonCounterfactualClassification (en) | accuracy | 72.627 |
| Classification | MTEB AmazonCounterfactualClassification (en) | ap | 34.469 |
| Classification | MTEB AmazonCounterfactualClassification (en) | f1 | 66.237 |
| Classification | MTEB AmazonPolarityClassification | accuracy | 92.518 |
| Classification | MTEB AmazonPolarityClassification | ap | 89.498 |
| Classification | MTEB AmazonPolarityClassification | f1 | 92.511 |
| Classification | MTEB AmazonReviewsClassification (en) | accuracy | 49.074 |
| Classification | MTEB AmazonReviewsClassification (en) | f1 | 48.448 |
| Retrieval | MTEB ArguAna | map_at_1 | 32.077 |
| Retrieval | MTEB ArguAna | map_at_10 | 48.153 |
| Retrieval | MTEB ArguAna | map_at_100 | 48.963 |
| Retrieval | MTEB ArguAna | map_at_1000 | 48.966 |
| Retrieval | MTEB ArguAna | map_at_3 | 43.184 |
| Retrieval | MTEB ArguAna | map_at_5 | 46.072 |
| Retrieval | MTEB ArguAna | mrr_at_1 | 33.073 |
| Retrieval | MTEB ArguAna | mrr_at_10 | 48.54 |
| Retrieval | MTEB ArguAna | mrr_at_100 | 49.335 |
| Retrieval | MTEB ArguAna | mrr_at_1000 | 49.338 |
| Retrieval | MTEB ArguAna | mrr_at_3 | 43.563 |
| Retrieval | MTEB ArguAna | mrr_at_5 | 46.383 |
| Retrieval | MTEB ArguAna | ndcg_at_1 | 32.077 |
| Retrieval | MTEB ArguAna | ndcg_at_10 | 57.158 |
| Retrieval | MTEB ArguAna | ndcg_at_100 | 60.325 |
| Retrieval | MTEB ArguAna | ndcg_at_1000 | 60.402 |
Using it via the API
Once AxForge deploys gte-large for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (gte-large 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-large","input":"text to embed"}'
Licence
Open weights under mit — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗
Explore