/ Модель

Характеристики

Контекст
8 192 токена
Параметры
149 млн
Лицензия
Apache 2.0
Архитектура
modernbert · dense
Базовая модель
ModernBERT-base
Библиотека
sentence-transformers
Опубликована
14 окт 2026

modernbert-base-reranker-msmarco524k-1e-4-20261014

Ранжирование текстаТекст → score

Англоязычный кросс-энкодер на базе ModernBERT оценивает релевантность пар «запрос — текст» для ранжирования результатов поиска.

80скачиваний за 30 дней0лайковвеса обновлены14 окт 2026

Бенчмарки

Как модель справилась с задачами пользователей ailist и что о ней публикуют открытые лидерборды

Тесты пользователейпока нет

Никто ещё не опубликовал тест этой модели. Соберите бенчмарк из своих задач — результат появится здесь.

  1. Добавьте модель в бенчмарк

    Вместе с другими моделями категории «Ранжирование» — сравнение идёт внутри одной категории.

  2. Опишите свои задачи

    Напишите запросы, при необходимости приложите файлы и эталонные ответы.

  3. Запустите и опубликуйте

    Модели ответят на каждую задачу, судья оценит ответы. Опубликованный результат появится на этой странице.

Публичные оценки

Из открытых лидербордов. Сравнивайте результаты одного теста при одинаковых условиях — отсутствие оценки не значит, что модель хуже.

В подключённых источниках результатов для этой модели пока нет.

Что умеет модель

Описание составлено по карточке модели и метаданным Hugging Face

ModernBERT-base reranker MS MARCO pilot — это англоязычный кросс-энкодер, дообученный на базе answerdotai/ModernBERT-base с помощью библиотеки sentence-transformers. Он оценивает пары «запрос — текст» и предназначен для ранжирования результатов поиска и семантического поиска. Модель содержит 149 млн параметров и принимает текст; в README максимальная длина последовательности указана как 291 токен, а метаданные сообщают нативный контекст 8192 токена. Выход — оценка релевантности пары.

Для обучения использован набор данных parquet размером около 1,05 млн примеров с запросами, фрагментами текстов и метками. Модель можно применять для повторного упорядочивания кандидатов, найденных поисковой системой, и оценки соответствия текста запросу. Лицензия — Apache-2.0. Ограничение, указанное в карточке: модель обучена для английского языка; длина последовательности в README существенно меньше контекста, заявленного в метаданных.

Сценарии применения

  • переранжирование документов по запросу
  • оценка релевантности поисковых результатов
  • ранжирование passages в системе вопросно-ответного поиска
  • улучшение семантического поиска повторной сортировкой кандидатов

Ограничения

Модель рассчитана на английский язык. README указывает максимальную длину последовательности 291 токен, хотя метаданные сообщают нативный контекст 8192 токена; практический предел следует проверять для используемой конфигурации.

Карточка модели

README от автора на Hugging Face, как он его опубликовал

Открыть на Hugging Face

ModernBERT-base reranker MS MARCO pilot

This is a Cross Encoder model finetuned from answerdotai/ModernBERT-base on the parquet dataset using the sentence-transformers library. It computes scores for pairs of texts, which can be used for text reranking and semantic search.

Model Details

Model Description

  • Model Type: Cross Encoder
  • Base model: answerdotai/ModernBERT-base
  • Maximum Sequence Length: 291 tokens
  • Number of Output Labels: 1 label
  • Supported Modality: Text
  • Training Dataset:
    • parquet
  • Language: en
  • License: apache-2.0

Model Sources

Full Model Architecture

CrossEncoder(
  (0): Transformer({'transformer_task': 'sequence-classification', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'logits'}}, 'module_output_name': 'scores', 'architecture': 'ModernBertForSequenceClassification'})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("tomaarsen/modernbert-base-reranker-msmarco524k-1e-4-20261014")
# Get scores for pairs of inputs
pairs = [
    ['how much does david cameron earn', "OK-So definitively David Cameron who is the current British Prime Minister earns £142,500 per year. Recently the Civil Service disclosed the salary's of all of the Cabinet members who earn over £150,000. David Cameron's £142,500 was used as a bench mark for this."],
    ['how much does david cameron earn', 'Celebrities wealth, income, and salary in 2016, 2017. Cameron Dallas. Handsome Vine star Cameron Dallas earned his first million before he was 20 years old. His current net worth is approximately $1.2 million as of 2017 and although he never admitted having any official girlfriend, we know the names of last two girls he was dating.'],
    ['distance from the island of palms to downtown charleston, sc', 'Distance, Gas Consumption and Emission Notes. Distance from Isle of Palms, SC to Charleston, SC is 18Miles or 29 Km. You can get this distance about 39 mins. If you want to planning travel with plane for 13 Miles or 20 Km, You can get this distance about 31 mins . A car with an average MPG will needs 0.83 gallons of gas to get the route between these points. The estimated cost of gas to get between Isle of Palms, SC and Charleston, SC is $1.88. During the route, an average car will release 16.26 pounds of CO 2 to the atmosphere. Your carbon footprint is 0.9 pounds of CO 2 per mile.'],
    ['distance from the island of palms to downtown charleston, sc', 'Once youâ\x80\x99ve done your obligatory sunbathing and beachcombing at the Isle of Palms County Park Beach, head to nearby Charleston, a 30-minute drive away, which has historic sites, landmarks, shopping districts, golf courses and family attractions like The South Carolina Aquarium and Fountain Walk, a new dining, entertainment and shopping complex nearby.'],
    ['how much does nicu cost per day', 'According to a 2010 article in Managed Care Magazine, the average cost for infants in NICU is around $3,000 per day. This average does not include surgeries or helicopter transports.'],
]
scores = model.predict(pairs)
print(scores)
# [10.6875 -5.6562  5.1875 -2.      1.125 ]

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'how much does david cameron earn',
    [
        "OK-So definitively David Cameron who is the current British Prime Minister earns £142,500 per year. Recently the Civil Service disclosed the salary's of all of the Cabinet members who earn over £150,000. David Cameron's £142,500 was used as a bench mark for this.",
        'Celebrities wealth, income, and salary in 2016, 2017. Cameron Dallas. Handsome Vine star Cameron Dallas earned his first million before he was 20 years old. His current net worth is approximately $1.2 million as of 2017 and although he never admitted having any official girlfriend, we know the names of last two girls he was dating.',
        'Distance, Gas Consumption and Emission Notes. Distance from Isle of Palms, SC to Charleston, SC is 18Miles or 29 Km. You can get this distance about 39 mins. If you want to planning travel with plane for 13 Miles or 20 Km, You can get this distance about 31 mins . A car with an average MPG will needs 0.83 gallons of gas to get the route between these points. The estimated cost of gas to get between Isle of Palms, SC and Charleston, SC is $1.88. During the route, an average car will release 16.26 pounds of CO 2 to the atmosphere. Your carbon footprint is 0.9 pounds of CO 2 per mile.',
        'Once youâ\x80\x99ve done your obligatory sunbathing and beachcombing at the Isle of Palms County Park Beach, head to nearby Charleston, a 30-minute drive away, which has historic sites, landmarks, shopping districts, golf courses and family attractions like The South Carolina Aquarium and Fountain Walk, a new dining, entertainment and shopping complex nearby.',
        'According to a 2010 article in Managed Care Magazine, the average cost for infants in NICU is around $3,000 per day. This average does not include surgeries or helicopter transports.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Training Details

Training Dataset

parquet

  • Dataset: parquet
  • Size: 1,048,576 training samples
  • Columns: query, passage, and label
  • Approximate statistics based on the first 100 samples:
    querypassagelabel
    typestringstringfloat
    modalitytexttext
    detailsmin: 5 tokensmean: 9.27 tokensmax: 15 tokensmin: 27 tokensmean: 84.48 tokensmax: 175 tokensmin: 0.0mean: 0.5max: 1.0
  • Samples:
    querypassagelabel
    how much does david cameron earnOK-So definitively David Cameron who is the current British Prime Minister earns £142,500 per year. Recently the Civil Service disclosed the salary's of all of the Cabinet members who earn over £150,000. David Cameron's £142,500 was used as a bench mark for this.1.0
    how much does david cameron earnCelebrities wealth, income, and salary in 2016, 2017. Cameron Dallas. Handsome Vine star Cameron Dallas earned his first million before he was 20 years old. His current net worth is approximately $1.2 million as of 2017 and although he never admitted having any official girlfriend, we know the names of last two girls he was dating.0.0
    distance from the island of palms to downtown charleston, scDistance, Gas Consumption and Emission Notes. Distance from Isle of Palms, SC to Charleston, SC is 18Miles or 29 Km. You can get this distance about 39 mins. If you want to planning travel with plane for 13 Miles or 20 Km, You can get this distance about 31 mins . A car with an average MPG will needs 0.83 gallons of gas to get the route between these points. The estimated cost of gas to get between Isle of Palms, SC and Charleston, SC is $1.88. During the route, an average car will release 16.26 pounds of CO 2 to the atmosphere. Your carbon footprint is 0.9 pounds of CO 2 per mile.1.0
  • Loss: BinaryCrossEntropyLoss with these parameters:
    {
        "activation_fn": "torch.nn.modules.linear.Identity",
        "pos_weight": 1.0
    }
    

Training Hyperparameters

Non-Default Hyperparameters

  • per_device_train_batch_size: 32
  • num_train_epochs: 1
  • learning_rate: 0.0001
  • warmup_steps: 0.05
  • optim: adamw_torch
  • weight_decay: 0.01
  • gradient_accumulation_steps: 4
  • bf16: True
  • tf32: True
  • logging_nan_inf_filter: False
  • disable_tqdm: True
  • per_device_eval_batch_size: 32
  • push_to_hub: True
  • hub_private_repo: False
  • hub_model_id: tomaarsen/modernbert-base-reranker-msmarco524k-1e-4-20261014
  • load_best_model_at_end: True
  • seed: 12
  • data_seed: 12
  • dataloader_num_workers: 2

All Hyperparameters

  • per_device_train_batch_size: 32
  • num_train_epochs: 1
  • max_steps: -1
  • learning_rate: 0.0001
  • lr_scheduler_type: linear
  • lr_scheduler_kwargs: None
  • warmup_steps: 0.05
  • optim: adamw_torch
  • optim_args: None
  • weight_decay: 0.01
  • adam_beta1: 0.9
  • adam_beta2: 0.999
  • adam_epsilon: 1e-08
  • optim_target_modules: None
  • gradient_accumulation_steps: 4
  • average_tokens_across_devices: True
  • max_grad_norm: 1.0
  • label_smoothing_factor: 0.0
  • bf16: True
  • fp16: False
  • bf16_full_eval: False
  • fp16_full_eval: False
  • tf32: True
  • gradient_checkpointing: False
  • gradient_checkpointing_kwargs: None
  • torch_compile: False
  • torch_compile_backend: None
  • torch_compile_mode: None
  • use_liger_kernel: False
  • liger_kernel_config: None
  • use_cache: False
  • neftune_noise_alpha: None
  • torch_empty_cache_steps: None
  • auto_find_batch_size: False
  • log_on_each_node: True
  • logging_nan_inf_filter: False
  • include_num_input_tokens_seen: no
  • log_level: passive
  • log_level_replica: warning
  • disable_tqdm: True
  • project: huggingface
  • trackio_space_id: None
  • trackio_bucket_id: None
  • trackio_static_space_id: None
  • per_device_eval_batch_size: 32
  • prediction_loss_only: True
  • eval_on_start: False
  • eval_do_concat_batches: True
  • eval_use_gather_object: False
  • eval_accumulation_steps: None
  • include_for_metrics: []
  • batch_eval_metrics: False
  • save_only_model: False
  • save_on_each_node: False
  • enable_jit_checkpoint: False
  • push_to_hub: True
  • hub_private_repo: False
  • hub_model_id: tomaarsen/modernbert-base-reranker-msmarco524k-1e-4-20261014
  • hub_strategy: every_save
  • hub_always_push: False
  • hub_revision: None
  • load_best_model_at_end: True
  • ignore_data_skip: False
  • restore_callback_states_from_checkpoint: False
  • full_determinism: False
  • seed: 12
  • data_seed: 12
  • use_cpu: False
  • accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
  • parallelism_config: None
  • dataloader_drop_last: False
  • dataloader_num_workers: 2
  • dataloader_pin_memory: True
  • dataloader_persistent_workers: False
  • dataloader_prefetch_factor: None
  • dataloader_multiprocessing_context: None
  • dataloader_in_order: True
  • remove_unused_columns: True
  • label_names: None
  • train_sampling_strategy: random
  • length_column_name: length
  • ddp_find_unused_parameters: None
  • ddp_bucket_cap_mb: None
  • ddp_broadcast_buffers: False
  • ddp_static_graph: None
  • ddp_backend: None
  • ddp_timeout: 1800
  • fsdp: None
  • fsdp_config: None
  • deepspeed: None
  • debug: []
  • skip_memory_metrics: True
  • do_predict: False
  • resume_from_checkpoint: None
  • local_rank: -1
  • prompts: None
  • batch_sampler: batch_sampler
  • multi_dataset_batch_sampler: proportional
  • router_mapping: {}
  • learning_rate_mapping: {}
  • warmup_ratio: None

Training Logs

EpochStepTraining Loss
0.000110.7529
0.01221000.6871
0.02442000.4820
0.03663000.3677
0.04884000.3628
0.06105000.3112
0.0625512-
0.07326000.2978
0.08547000.2902
0.09778000.2857
0.10999000.2743
0.122110000.2729
0.1251024-
0.134311000.2552
0.146512000.2598
0.158713000.2440
0.170914000.2504
0.183115000.2442
0.18751536-
0.195316000.2395
0.207517000.2399
0.219718000.2282
0.231919000.2307
0.244120000.2263
0.252048-
0.256321000.2173
0.268622000.2197
0.280823000.2145
0.293024000.2219
0.305225000.2173
0.31252560-
0.317426000.2209
0.329627000.2149
0.341828000.2038
0.354029000.2046
0.366230000.2060
0.3753072-
0.378431000.1976
0.390632000.2014
0.402833000.2070
0.415034000.1959
0.427235000.1919
0.43753584-
0.439536000.1944
0.451737000.1903
0.463938000.1920
0.476139000.1871
0.488340000.1883
0.54096-
0.500541000.1874
0.512742000.1840
0.524943000.1826
0.537144000.1909
0.549345000.1813
0.561546000.1666
0.56254608-
0.573747000.1844
0.585948000.1708
0.598149000.1741
0.610450000.1721
0.622651000.1690
0.6255120-
0.634852000.1743
0.647053000.1754
0.659254000.1745
0.671455000.1594
0.683656000.1622
0.68755632-
0.695857000.1645
0.708058000.1597
0.720259000.1628
0.732460000.1620
0.744661000.1622
0.756144-
0.756862000.1585
0.769063000.1512
0.781264000.1513
0.793565000.1508
0.805766000.1512
0.81256656-
0.817967000.1565
0.830168000.1487
0.842369000.1478
0.854570000.1543
0.866771000.1451
0.8757168-
0.878972000.1447
0.891173000.1400
0.903374000.1409
0.915575000.1479
0.927776000.1448
0.93757680-
0.939977000.1390
0.952178000.1359
0.964479000.1427
0.976680000.1452
0.988881000.1383
1.08192-
  • The bold row denotes the saved checkpoint.

Training Time

  • Training: 44.2 minutes
  • Evaluation: 3.3 minutes
  • Total: 47.5 minutes

Framework Versions

  • Python: 3.12.12
  • Sentence Transformers: 6.2.0.dev0
  • Transformers: 5.16.1
  • PyTorch: 2.10.0+cu128
  • Accelerate: 1.15.0
  • Datasets: 5.1.0
  • Tokenizers: 0.23.3

Additional Resources

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

Retrieval scaling experiment

One epoch on 524,288 shared MS MARCO triplets, seed 12, LR 0.0001. Query content is clipped to 32 tokens, passage content to 256, combined cap 291. Checkpoint selected by the original 1,024-query validation set. Test scores do not select checkpoints. Experiment artifacts.

Как запустить

На своей GPU в ailist. Цена — по текущим предложениям провайдеров

GPU в ailist

Универсальный рантайм · sentence-transformers

Можно запустить

Подбираем GPU…

Класс архитектуры
ModernBertForSequenceClassification
Ревизия весов
a6d842a

0 моделей