--- title: F2LLM-v2-160M canonical_url: "https://www.modelscope.cn/models/codefuse-ai/F2LLM-v2-160M" md_url: "https://www.modelscope.cn/models/codefuse-ai/F2LLM-v2-160M.md" repository: codefuse-ai/F2LLM-v2-160M last_updated: 2026-05-25 license: apache-2.0 pipeline_tag: feature-extraction tasks: - feature-extraction model_type: - qwen3 architectures: - Qwen3Model base_model: - codefuse-ai/F2LLM-v2-0.6B-Preview-Pruned-160M base_model_relation: finetune parameters: 159.2M tensor_type: - BF16 library_name: - pytorch - transformer - sentence-transformers - safetensors frameworks: - Pytorch language: - en - zh - ru - es - fr - de - ar - nl - vi - hi - ko - ja - it - id - pt - pl - tr - da - th - sv - fa - uk - cs - no - el - ca - ro - fi - bg - tl - gl - my - hy - km - ne - hu - eu - he - lo - sw - az - lv - si - sk - tg - et - lt - ms - hr - is - sl - sr - ur - bn - af - ta - ka - te - ml - mn - nn - kk - cy - mr - sq - nb - mk - jv - kn - eo - la - gu - uz - am - oc - be - mg - vo - pa - lb - ht - br - ga - xh - tt - bs - yo downloads: 39 stars: 1 tags: - sentence-transformers --- # F2LLM-v2-160M > F2LLM-v2-160M - codefuse-ai 在 ModelScope 开源的模型。F2LLM-v2 is a family of general-purpose, multilingual embedding models in 8 distinct sizes ranging from 80M to 14B. Trained on a curated composite of 60 million publicly available high-quality data, F2LLM-v2… codefuse-ai/F2LLM-v2-160M 是 ModelScope 魔搭社区上的 159.2M 参数feature-extraction模型，采用 apache-2.0 许可，基于 codefuse-ai/F2LLM-v2-0.6B-Preview-Pruned-160M 构建。 - **Repository**: codefuse-ai/F2LLM-v2-160M - **License**: apache-2.0 - **Tasks**: feature-extraction - **Parameters**: 159.2M - **Base model**: codefuse-ai/F2LLM-v2-0.6B-Preview-Pruned-160M - **Tags**: sentence-transformers - **Downloads**: 39 - **Stars**: 1 - **Last updated**: 2026-05-25 Source: https://www.modelscope.cn/models/codefuse-ai/F2LLM-v2-160M --- # F2LLM-v2-160M F2LLM-v2 is a family of general-purpose, multilingual embedding models in 8 distinct sizes ranging from 80M to 14B. Trained on a curated composite of 60 million publicly available high-quality data, F2LLM-v2 supports more than 200 languages, with a particular emphasis on previously underserved mid- and low-resource languages. F2LLM-v2 is fully open. We release base models in 5 sizes, instruct models in 8 sizes, the training data, the training code, and intermediate checkpoints. The three smallest instruct models are pruned and trained from the 0.6B base model. | Model | Base | Instruct | | ----- | --------------------------------------------------------------------------------- | ----------------------------------------------------------------------- | | 80M | | [F2LLM-v2-80M](https://modelscope.cn/models/codefuse-ai/F2LLM-v2-80M) | | 160M | | [F2LLM-v2-160M](https://modelscope.cn/models/codefuse-ai/F2LLM-v2-160M) | | 330M | | [F2LLM-v2-330M](https://modelscope.cn/models/codefuse-ai/F2LLM-v2-330M) | | 0.6B | [F2LLM-v2-0.6B-Preview](https://huggingface.co/codefuse-ai/F2LLM-v2-0.6B-Preview) | [F2LLM-v2-0.6B](https://modelscope.cn/models/codefuse-ai/F2LLM-v2-0.6B) | | 1.7B | [F2LLM-v2-1.7B-Preview](https://huggingface.co/codefuse-ai/F2LLM-v2-1.7B-Preview) | [F2LLM-v2-1.7B](https://modelscope.cn/models/codefuse-ai/F2LLM-v2-1.7B) | | 4B | [F2LLM-v2-4B-Preview](https://huggingface.co/codefuse-ai/F2LLM-v2-4B-Preview) | [F2LLM-v2-4B](https://modelscope.cn/models/codefuse-ai/F2LLM-v2-4B) | | 8B | [F2LLM-v2-8B-Preview](https://huggingface.co/codefuse-ai/F2LLM-v2-8B-Preview) | [F2LLM-v2-8B](https://modelscope.cn/models/codefuse-ai/F2LLM-v2-8B) | | 14B | [F2LLM-v2-14B-Preview](https://huggingface.co/codefuse-ai/F2LLM-v2-14B-Preview) | [F2LLM-v2-14B](https://modelscope.cn/models/codefuse-ai/F2LLM-v2-14B) | ## Usage ### With Sentence Transformers To encode text with the [Sentence Transformers](https://www.sbert.net/) library: ```python from sentence_transformers import SentenceTransformer model = SentenceTransformer("codefuse-ai/F2LLM-v2-160M", device="cuda:0", model_kwargs={"torch_dtype": "bfloat16"}) # Some sample query and documents query = "What is F2LLM used for?" documents = [ 'We present F2LLM, a family of fully open embedding LLMs that achieve a strong balance between model size, training data, and embedding performance.', 'F2LLM is a model for computing text embeddings that can be used for various NLP tasks such as information retrieval, semantic search, and text classification.', 'F2LLM 是 CodeFuse 开源的系列嵌入模型。', 'F2LLM — это модель вычисления встраивания текста, которую можно использовать для различных задач НЛП, таких как поиск информации, семантический поиск и классификация текста.' ] # Encode the query and documents separately. The encode_query method uses the query prompt query_embedding = model.encode_query(query) document_embeddings = model.encode_document(documents) print(query_embedding.shape, document_embeddings.shape) # (640,) (4, 640) # Compute cosine similarity between the query and documents similarity = model.similarity(query_embedding, document_embeddings) print(similarity) # tensor([[0.6373, 0.7239, 0.6302, 0.7509]]) ``` ### With Transformers Or directly with the [Transformers](https://huggingface.co/docs/transformers/index) library: ```python from transformers import AutoModel, AutoTokenizer import torch import torch.nn.functional as F model_path = "codefuse-ai/F2LLM-v2-160M" tokenizer = AutoTokenizer.from_pretrained(model_path) model = AutoModel.from_pretrained(model_path, torch_dtype=torch.bfloat16, device_map={'': 0}) query = "What is F2LLM used for?" query_prompt = "Instruct: Given a question, retrieve passages that can help answer the question.\nQuery: " documents = [ 'We present F2LLM, a family of fully open embedding LLMs that achieve a strong balance between model size, training data, and embedding performance.', 'F2LLM is a model for computing text embeddings that can be used for various NLP tasks such as information retrieval, semantic search, and text classification.', 'F2LLM 是 CodeFuse 开源的系列嵌入模型。', 'F2LLM — это модель вычисления встраивания текста, которую можно использовать для различных задач НЛП, таких как поиск информации, семантический поиск и классификация текста.' ] def encode(sentences): batch_size = len(sentences) # the tokenizer will automatically add eos token tokenized_inputs = tokenizer(sentences, padding=True, return_tensors='pt').to(model.device) last_hidden_state = model(**tokenized_inputs).last_hidden_state eos_positions = tokenized_inputs.attention_mask.sum(dim=1) - 1 embeddings = last_hidden_state[torch.arange(batch_size, device=model.device), eos_positions] embeddings = F.normalize(embeddings, p=2, dim=1) return embeddings # Encode the query and documents query_embedding = encode([query_prompt + query]) document_embeddings = encode(documents) print(query_embedding.shape, document_embeddings.shape) # torch.Size([1, 640]) torch.Size([4, 640]) # Compute cosine similarity between the query and documents similarity = query_embedding @ document_embeddings.T print(similarity) # tensor([[0.6367, 0.7227, 0.6328, 0.7500]], device='cuda:0', # dtype=torch.bfloat16, grad_fn=) ``` ## Intermediate Checkpoints To facilitate future research, we release intermediate checkpoints in the `intermediate_checkpoints` branch. ## Citation ``` @misc{f2llm-v2, title={F2LLM-v2: Inclusive, Performant, and Efficient Embeddings for a Multilingual World}, author={Ziyin Zhang and Zihan Liao and Hang Yu and Peng Di and Rui Wang}, year={2026}, eprint={2603.19223}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2603.19223}, } ```