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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("JacobLinCool/Qwen3-Embedding-0.6B-GIR-1")
# Run inference
queries = [
    "Generates samples of text from the provided vocabulary.\n\n  Args:\n    plain_vocab: vocabulary.\n    distribution: distribution.\n    train_samples: samples for training.\n    length: length.\n\n  Returns:\n    train_indices (np.array of Integers): random integers for training.\n      shape = [num_samples, length]\n    test_indices (np.array of Integers): random integers for testing.\n      shape = [num_samples, length]\n    plain_vocab   (list of Integers): unique vocabularies.",
]
documents = [
    'def generate_plaintext_random(plain_vocab, distribution, train_samples,\n                              length):\n  \n  if distribution is not None:\n    assert len(distribution) == len(plain_vocab)\n\n  train_indices = np.random.choice(\n      range(len(plain_vocab)), (train_samples, length), p=distribution)\n\n  return train_indices',
    'def switch(self, name):\n        \n        try:\n            switch = self.storage[self.__namespaced(name)]\n        except KeyError:\n            if not self.autocreate:\n                raise ValueError("No switch named  registered in " % (name, self.namespace))\n\n            switch = self.__create_and_register_disabled_switch(name)\n\n        switch.manager = self\n        return switch',
    'def late_filling(target, pressure=,\n                 Pc_star=,\n                 Swp_star=0.2, eta=3):\n    r\n    element = pressure.split()[0]\n    network = target.project.network\n    phase = target.project.find_phase(target)\n    pc_star = phase[Pc_star]\n    Pc = phase[pressure]\n    \n        Ts = network.map_throats(throats=target.Ts, origin=target)\n        values = values[Ts]\n    else:\n        Ps = network.map_pores(pores=target.Ps, origin=target)\n        values = values[Ps]\n    return values',
]
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([[ 0.8344, -0.0822,  0.0233]])

Evaluation

Metrics

Information Retrieval

Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.99
cosine_accuracy@5	1.0
cosine_accuracy@10	1.0
cosine_precision@1	0.99
cosine_precision@3	0.3333
cosine_precision@5	0.2
cosine_precision@10	0.1
cosine_recall@1	0.99
cosine_recall@3	1.0
cosine_recall@5	1.0
cosine_recall@10	1.0
cosine_ndcg@1	0.99
cosine_ndcg@5	0.9963
cosine_ndcg@10	0.9963
cosine_mrr@1	0.99
cosine_mrr@5	0.995
cosine_mrr@10	0.995
cosine_map@100	0.995

Training Details

Training Dataset

Unnamed Dataset

Size: 400 training samples
Columns: query and code
Approximate statistics based on the first 400 samples:
query code
type string string
details
min: 2 tokens
mean: 67.12 tokens
max: 3156 tokens
min: 24 tokens
mean: 126.98 tokens
max: 1236 tokens

Samples:

query	code
`For memory actions, get a list of addresses it operates on.[object Object][object Object] :param SimAction action: The action object to work with.[object Object] :return: A list of addresses that are accessed with that action.[object Object] :rtype: list`	`def _get_actual_addrs(action, state):[object Object] [object Object][object Object] if action.actual_addrs is None:[object Object] [object Object] addr_list = {0x60000000} [object Object] else:[object Object] addr_list = set(action.actual_addrs)[object Object][object Object] return addr_list`
`Construct the input file of the calculation.`	`def make_input(self, with_header=False):[object Object] [object Object] s = str(self.input)[object Object] if with_header: s = str(self) + "\n" + s[object Object] return s`
`Check worker status route`	def check_worker_status():[object Object] [object Object] if not in request.args:[object Object] resp = {"status": "bad request"}[object Object] return jsonify(**resp)[object Object] else:[object Object] worker_id = request.args[][object Object] assignment_id = request.args[][object Object] allow_repeats = CONFIG.getboolean(, )[object Object] if allow_repeats: [object Object] try:[object Object] part = Participant.query.<br> filter(Participant.workerid == worker_id).<br> filter(Participant.assignmentid == assignment_id).one()[object Object] status = part.status[object Object] except exc.SQLAlchemyError:[object Object] status = NOT_ACCEPTED[object Object] else: [object Object] try:[object Object] matches = Participant.query.<br> filter(Participant.workerid == worker_id).all()[object Object] numrecs = len(matches)[object Object] if numrecs==0: [object Object] status = NOT_ACCEPTED[object Object] else:[object Object] status = max([record.status for record in matches])[object Object] except exc.SQLAlchemyError:[object Object] ...

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "gather_across_devices": false 5}

Evaluation Dataset

Unnamed Dataset

Size: 100 evaluation samples
Columns: query and code
Approximate statistics based on the first 100 samples:
query code
type string string
details
min: 5 tokens
mean: 66.56 tokens
max: 548 tokens
min: 24 tokens
mean: 142.11 tokens
max: 901 tokens

Samples:

query	code
Return the value of the android prefixed attribute in a specific tag.[object Object][object Object] This function will always try to get the attribute with a android: prefix first,[object Object] and will try to return the attribute without the prefix, if the attribute could not be found.[object Object] This is useful for some broken AndroidManifest.xml, where no android namespace is set,[object Object] but could also indicate malicious activity (i.e. wrongly repackaged files).[object Object] A warning is printed if the attribute is found without a namespace prefix.[object Object][object Object] If you require to get the exact result you need to query the tag directly:[object Object][object Object] example::[object Object] >>> from lxml.etree import Element[object Object] >>> tag = Element('bar', nsmap={'android': '[object Object])[object Object] >>> tag.set('{[object Object]', 'barfoo')[object Object] >>> tag.set('name', 'baz')[object Object] # Assume that [object Object] is some APK object[object Object] >>> a.get_value_from_tag(tag, 'name'...	def get_value_from_tag(self, tag, attribute):[object Object] [object Object][object Object] [object Object] [object Object] value = tag.get(self._ns(attribute))[object Object] if value is None:[object Object] value = tag.get(attribute)[object Object][object Object] if value:[object Object] [object Object] log.warning("Failed to get the attribute on tag with namespace. "[object Object] "But found the same attribute without namespace!".format(attribute, tag.tag))[object Object] return value
`Get information about this object as a dictionary. Used by WebSocket interface to pass some[object Object] relevant information to client applications.`	`def get_as_datadict(self):[object Object] [object Object] return dict(type=self.[object Object].[object Object], tags=list(self.tags))`
Makes forecast with the estimated model[object Object][object Object] Parameters[object Object] ----------[object Object] h : int (default : 5)[object Object] How many steps ahead would you like to forecast?[object Object][object Object] past_values : int (default : 20)[object Object] How many past observations to show on the forecast graph?[object Object][object Object] intervals : Boolean[object Object] Would you like to show 95% prediction intervals for the forecast?[object Object][object Object] Returns[object Object] ----------[object Object] - Plot of the forecast	def plot_predict(self,h=5,past_values=20,intervals=True,*kwargs): [object Object] [object Object] import matplotlib.pyplot as plt[object Object] import seaborn as sns[object Object][object Object] figsize = kwargs.get(,(10,7))[object Object][object Object] if self.latent_variables.estimated is False:[object Object] raise Exception("No latent variables estimated!")[object Object] else:[object Object] [object Object] scale, shape, skewness = self._get_scale_and_shape(self.latent_variables.get_z_values(transformed=True))[object Object] previous_value = self.data[-1] [object Object] forecasted_values = np.ones(h)self.states[-1] [object Object] date_index = self.shift_dates(h)[object Object] simulations = 10000[object Object] sim_vector = np.zeros([simulations,h])[object Object] t_params = self.transform_z()[object Object][object Object] for n in range(0,simulations): [object Object] rnd_q = np.random.normal(0,np.sqrt(self.latent_variables.get_z_values(transformed=True)[0]),h) [object Object] exp = forecasted_values.copy()[object Object][object Object] for t in range(0,h):[object Object] if t == 0:...

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "gather_across_devices": false 5}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: epoch
per_device_train_batch_size: 64
per_device_eval_batch_size: 64
num_train_epochs: 1
warmup_ratio: 0.1
seed: 2025
bf16: True
load_best_model_at_end: True
optim: adamw_torch
push_to_hub: True
hub_model_id: JacobLinCool/Qwen3-Embedding-0.6B-GIR-1
hub_private_repo: False
gradient_checkpointing: True
eval_on_start: True
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: epoch
prediction_loss_only: True
per_device_train_batch_size: 64
per_device_eval_batch_size: 64
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 5e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 1
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 2025
data_seed: None
jit_mode_eval: False
use_ipex: False
bf16: True
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: True
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
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
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: True
resume_from_checkpoint: None
hub_model_id: JacobLinCool/Qwen3-Embedding-0.6B-GIR-1
hub_strategy: every_save
hub_private_repo: False
hub_always_push: False
hub_revision: None
gradient_checkpointing: True
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: True
use_liger_kernel: False
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Validation Loss	cosine_ndcg@10
0	0	0.0616	0.9926
1.0	7	0.0358	0.9963
-1	-1	-	0.9963

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.11.11
Sentence Transformers: 5.1.1
Transformers: 4.56.2
PyTorch: 2.8.0+cu128
Accelerate: 1.10.1
Datasets: 4.1.1
Tokenizers: 0.22.1

Citation

BibTeX

Sentence Transformers

bibtex

@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

bibtex

@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}
}

	query	code
type	string	string
details	min: 2 tokens mean: 67.12 tokens max: 3156 tokens	min: 24 tokens mean: 126.98 tokens max: 1236 tokens