SentenceTransformer based on Qwen/Qwen3-Embedding-0.6B

This is a sentence-transformers model finetuned from Qwen/Qwen3-Embedding-0.6B. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: Qwen/Qwen3-Embedding-0.6B
Maximum Sequence Length: 32768 tokens
Output Dimensionality: 1024 dimensions
Similarity Function: Cosine Similarity

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 32768, 'do_lower_case': False, 'architecture': 'Qwen3Model'})
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': True, 'include_prompt': True})
  (2): Normalize()
)

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 SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
queries = [
    "hyperswitch router utility functions error handling",
]
documents = [
    '// PATH: data/code_corpus_hyperswitch/crates__router__src__utils.rs\n// MODULE: data::code_corpus_hyperswitch::crates__router__src__utils.rs\n// SYMBOL: get_error_response\n    fn get_error_response(self) -> RouterResult<types::Response> {\n        self.map_err(|error| error.change_context(errors::ApiErrorResponse::InternalServerError))\n            .attach_printable("Error while receiving response")\n            .and_then(|inner| match inner {\n                Ok(res) => {\n                    logger::error!(response=?res);\n                    Err(errors::ApiErrorResponse::InternalServerError).attach_printable(format!(\n                        "Expecting error response, received response: {res:?}"\n                    ))\n                }\n                Err(err_res) => Ok(err_res),\n            })\n    }',
    '// PATH: data/code_corpus_hyperswitch/crates__hyperswitch_domain_models__src__router_data.rs\n// MODULE: data::code_corpus_hyperswitch::crates__hyperswitch_domain_models__src__router_data.rs\n// SYMBOL: CustomerInfo\npub struct CustomerInfo {\n    pub customer_id: Option<id_type::CustomerId>,\n    pub customer_email: Option<common_utils::pii::Email>,\n    pub customer_name: Option<Secret<String>>,\n    pub customer_phone_number: Option<Secret<String>>,\n    pub customer_phone_country_code: Option<String>,\n}\n#[derive(Debug, Clone, Serialize, Deserialize)]',
    '// PATH: data/code_corpus_hyperswitch/crates__hyperswitch_connectors__src__connectors__zift.rs\n// MODULE: data::code_corpus_hyperswitch::crates__hyperswitch_connectors__src__connectors__zift.rs\n// SYMBOL: build_request\n    fn build_request(\n        &self,\n        req: &SetupMandateRouterData,\n        connectors: &Connectors,\n    ) -> CustomResult<Option<Request>, errors::ConnectorError> {\n        Ok(Some(\n            RequestBuilder::new()\n                .method(Method::Post)\n                .url(&types::SetupMandateType::get_url(self, req, connectors)?)\n                .attach_default_headers()\n                .headers(types::SetupMandateType::get_headers(self, req, connectors)?)\n                .set_body(types::SetupMandateType::get_request_body(\n                    self, req, connectors,\n                )?)\n                .build(),\n        ))\n    }',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 1024] [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[ 0.4434, -0.0422, -0.0713]], dtype=torch.bfloat16)

Training Details

Training Dataset

Unnamed Dataset

Size: 1,736 training samples
Columns: sentence_0 and sentence_1
Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1
type string string
details
min: 6 tokens
mean: 10.85 tokens
max: 18 tokens

min: 100 tokens
mean: 225.13 tokens
max: 854 tokens

	sentence_0	sentence_1
type	string	string
details	min: 6 tokens mean: 10.85 tokens max: 18 tokens	min: 100 tokens mean: 225.13 tokens max: 854 tokens

Samples:

sentence_0	sentence_1
`how to override proxy settings in session state`	// PATH: data/code_corpus_hyperswitch/crates__router__src__core__unified_connector_service.rs // MODULE: data::code_corpus_hyperswitch::crates__router__src__core__unified_connector_service.rs // SYMBOL: create_updated_session_state_with_proxy fn create_updated_session_state_with_proxy( state: SessionState, proxy_override: &ProxyOverride, ) -> SessionState { let mut updated_state = state; // Create updated configuration with proxy overrides let mut updated_conf = (*updated_state.conf).clone(); // Update proxy URLs with overrides, falling back to existing values if let Some(ref http_url) = proxy_override.http_url { updated_conf.proxy.http_url = Some(http_url.clone()); } if let Some(ref https_url) = proxy_override.https_url { updated_conf.proxy.https_url = Some(https_url.clone()); } updated_state.conf = std::sync::Arc::new(updated_conf); updated_state }
`rust hyperswitch analytics clip_to_start time bucket function`	// PATH: data/code_corpus_hyperswitch/crates__analytics__src__query.rs // MODULE: data::code_corpus_hyperswitch::crates__analytics__src__query.rs // SYMBOL: clip_to_start fn clip_to_start( &self, value: Self::SeriesType, ) -> error_stack::Result { let clip_start = \|value: u8, modulo: u8\| -> u8 { value - value % modulo }; let clipped_time = match ( self.get_lowest_common_granularity_level(), self.get_bucket_size(), ) { (TimeGranularityLevel::Minute, i) => time::Time::MIDNIGHT .replace_second(clip_start(value.second(), i)) .and_then(\|t\| t.replace_minute(value.minute())) .and_then(\|t\| t.replace_hour(value.hour())), (TimeGranularityLevel::Hour, i) => time::Time::MIDNIGHT .replace_minute(clip_start(value.minute(), i)) .and_then(\|t\| t.replace_hour(value.hour())), (Tim...
`rust function to get payment method type from domain models`	// PATH: data/code_corpus_hyperswitch/crates__hyperswitch_domain_models__src__payment_method_data.rs // MODULE: data::code_corpus_hyperswitch::crates__hyperswitch_domain_models__src__payment_method_data.rs // SYMBOL: get_payment_method pub fn get_payment_method(&self) -> Option { match self { Self::Card() \| Self::NetworkToken() \| Self::CardDetailsForNetworkTransactionId() \| Self::NetworkTokenDetailsForNetworkTransactionId() \| Self::DecryptedWalletTokenDetailsForNetworkTransactionId() \| Self::CardWithLimitedDetails() => Some(common_enums::PaymentMethod::Card), Self::CardRedirect() => Some(common_enums::PaymentMethod::CardRedirect), Self::Wallet() => Some(common_enums::PaymentMethod::Wallet), Self::PayLater() => Some(common_enums::PaymentMethod::PayLater), Self::BankRedirect() => Some(common_enums::PaymentMethod::BankRe...

Loss: MultipleNegativesRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim",
    "gather_across_devices": false
}

Training Hyperparameters

Non-Default Hyperparameters

multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

per_device_train_batch_size: 8
num_train_epochs: 3
max_steps: -1
learning_rate: 5e-05
lr_scheduler_type: linear
lr_scheduler_kwargs: None
warmup_steps: 0
optim: adamw_torch_fused
optim_args: None
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
optim_target_modules: None
gradient_accumulation_steps: 1
average_tokens_across_devices: True
max_grad_norm: 1
label_smoothing_factor: 0.0
bf16: False
fp16: False
bf16_full_eval: False
fp16_full_eval: False
tf32: None
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: True
include_num_input_tokens_seen: no
log_level: passive
log_level_replica: warning
disable_tqdm: False
project: huggingface
trackio_space_id: trackio
eval_strategy: no
per_device_eval_batch_size: 8
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: False
hub_private_repo: None
hub_model_id: None
hub_strategy: every_save
hub_always_push: False
hub_revision: None
load_best_model_at_end: False
ignore_data_skip: False
restore_callback_states_from_checkpoint: False
full_determinism: False
seed: 42
data_seed: None
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: 0
dataloader_pin_memory: True
dataloader_persistent_workers: False
dataloader_prefetch_factor: None
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_backend: None
ddp_timeout: 1800
fsdp: []
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
deepspeed: None
debug: []
skip_memory_metrics: True
do_predict: False
resume_from_checkpoint: None
warmup_ratio: None
local_rank: -1
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
2.3041	500	0.0380

Framework Versions

Python: 3.12.3
Sentence Transformers: 5.2.3
Transformers: 5.2.0
PyTorch: 2.10.0+cu128
Accelerate: 1.12.0
Datasets: 4.5.0
Tokenizers: 0.22.2

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",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}