# lt-wikidata-comp-en by Dell Research Harvard: Open Model Source: https://savrn.com/models/lt-wikidata-comp-en Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve lt-wikidata-comp-en (109M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them. | Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits | | --- | --- | --- | --- | --- | --- | | 16-bit | 0.2 GB | 0.3 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 0.1 GB | 0.1 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 0.1 GB | 0.1 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the [SAVRN Index](https://savrn.com/ai-index/pricing/gpus), read Oct 23, 2026. [lt-wikidata-comp-en on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/lt-wikidata-comp-en/gpus) ## Model Card This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model- it just wraps around the class. It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more. Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well. It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Take a look at the documentation of sentence-transformers if you want to use this model for more than what we… Excerpt from the card by Dell Research Harvard. ## Configuration Architecture MPNetModel Context length (tokens) 514 Layers 12 Hidden size 768 Feed-forward size 3,072 Attention heads 12 Vocabulary size 30,527 Stored precision float32 Model type mpnet ## Identity and Version Repository dell-research-harvard/lt-wikidata-comp-en Publisher Dell Research Harvard Task Sentence similarity Modality Text Library sentence-transformers Parameters 109M parameters Languages en Revision af65d96525b7cb3e125a52b89928ae78b39c6d21 First published 2023-08-11 Last updated 2024-03-20 ## Files and Weights 19 files, 452.2 MB in total. The weights are 1 file totalling 438.0 MB in safetensors. Weights1 file · 438.0 MB Configuration7 files · 3.3 KB Tokenizer3 files · 944.1 KB Documentation1 file · 5.7 KB Other6 files · 13.3 MB Repository1 file · 472 B Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 438.0 MB | 4d88dbbf624b | | 1_Pooling/config.json | Configuration | 270 B | — | | LT_training_config.json | Configuration | 1.0 KB | — | | config.json | Configuration | 622 B | — | | config_sentence_transformers.json | Configuration | 116 B | — | | modules.json | Configuration | 229 B | — | | sentence_bert_config.json | Configuration | 53 B | — | | special_tokens_map.json | Configuration | 962 B | — | | README.md | Documentation | 5.7 KB | — | | Information-Retrieval_evaluation_eval_results.csv | Other | 582 B | — | | Information-Retrieval_evaluation_test_results.csv | Other | 575 B | — | | eval/Information-Retrieval_evaluation_eval_results.csv | Other | 58.9 KB | — | | linkage_en_aliases_train.csv | Other | 6.8 MB | d2f83e1d4ae0 | | test_data.pickle | Other | 728.9 KB | 44d740112e42 | | val_data.pickle | Other | 5.7 MB | 0b6468956478 | | .gitattributes | Repository | 472 B | — | | tokenizer.json | Tokenizer | 710.9 KB | — | | tokenizer_config.json | Tokenizer | 1.6 KB | — | | vocab.txt | Tokenizer | 231.5 KB | — | ## License and Download License Not stated by the source Access Open weights, no gate Download size 438.0 MB [Download from Dell Research Harvard](https://huggingface.co/dell-research-harvard/lt-wikidata-comp-en) Released by Dell Research Harvard through its official repository on Hugging Face. ## Built From - Described by [arXiv:2309.00789](https://savrn.com/papers/linktransformer-a-unified-package-for-record-linkage-with-transformer-language-models) ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 438.0 MB | | 16-bit | 0.2 GB | | 8-bit | 0.1 GB | | 4-bit | 0.1 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About lt-wikidata-comp-en ### How much GPU memory does lt-wikidata-comp-en need? About 0.3 GB at 16-bit and 0.1 GB at 4-bit: the weights (109M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run lt-wikidata-comp-en on? At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists. ### What is lt-wikidata-comp-en's context length? 514 tokens, from the maximum position embeddings in its published configuration. ## Similar Models Model · Sentence similarity ### [lt-un-data-fine-fine-en](https://savrn.com/models/lt-un-data-fine-fine-en) [Dell Research Harvard](https://savrn.com/model-publishers/dell-research-harvard) This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model- it just wraps around the class. It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more. Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well. It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Take a look at the documentation of sentence-transformers if you want to use this model for more than what we… Open weights 109M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/lt-un-data-fine-fine-en) Model · Sentence similarity ### [names-linking-model](https://savrn.com/models/names-linking-model) [Dell Research Harvard](https://savrn.com/model-publishers/dell-research-harvard) This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model- it just wraps around the class. It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more. Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well. It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Take a look at the documentation of sentence-transformers if you want to use this model for more than what we… Open weights 109M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/names-linking-model) Model · Sentence similarity ### [lt-patent-inventor-linking](https://savrn.com/models/lt-patent-inventor-linking) [Matthew Lee Chen](https://savrn.com/model-publishers/gbpatentdata) This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model - it just wraps around the class. This model has been fine-tuned on the model: sentence-transformers/all-mpnet-base-v2. It is pretrained for the language: en. To use this model using sentence-transformers: To use this model for clustering with LinkTransformer installed: We evaluate using the standard LinkTransformer information retrieval metrics. Our test set evaluations are available here. The model was trained with the parameters: torch.utils.data.dataloader.DataLoader of length 31 with parameters: linktransformer.modifiedsbert.losses.SupConLosswandb If you use our… Open weights 109M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/lt-patent-inventor-linking) Model · Sentence similarity ### [lt-un-data-fine-coarse-en](https://savrn.com/models/lt-un-data-fine-coarse-en) [Dell Research Harvard](https://savrn.com/model-publishers/dell-research-harvard) This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model- it just wraps around the class. It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more. Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well. It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Take a look at the documentation of sentence-transformers if you want to use this model for more than what we… Open weights 109M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/lt-un-data-fine-coarse-en) Model · Sentence similarity ### [lt-un-data-fine-industry-en](https://savrn.com/models/lt-un-data-fine-industry-en) [Dell Research Harvard](https://savrn.com/model-publishers/dell-research-harvard) This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model- it just wraps around the class. It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more. Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well. It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Take a look at the documentation of sentence-transformers if you want to use this model for more than what we… Open weights 109M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/lt-un-data-fine-industry-en) Model · Sentence similarity ### [lt-patent-inventor-linking](https://savrn.com/models/lt-patent-inventor-linking-2) [Matthew Lee Chen](https://savrn.com/model-publishers/matthewleechen) This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model - it just wraps around the class. This model has been fine-tuned on the model: sentence-transformers/all-mpnet-base-v2. It is pretrained for the language: en. To use this model using sentence-transformers: To use this model for clustering with LinkTransformer installed: We evaluate using the standard LinkTransformer information retrieval metrics. Our test set evaluations are available here. The model was trained with the parameters: torch.utils.data.dataloader.DataLoader of length 31 with parameters: linktransformer.modifiedsbert.losses.SupConLosswandb If you use our… Open weights 109M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/lt-patent-inventor-linking-2) ## Dell Research Harvard [All models and datasets](https://savrn.com/model-publishers/dell-research-harvard) ## Versions - [af65d96525b7](https://savrn.com/models/lt-wikidata-comp-en/versions/af65d96525b7) · current 2026-09-18 ## Explore More - [All sentence similarity models](https://savrn.com/models/tasks/sentence-similarity) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-21. - [Hugging Face record](https://huggingface.co/dell-research-harvard/lt-wikidata-comp-en) - [How the hub is built](https://savrn.com/model-hub/methodology)