Tasks and inputs
BAAI/bge-reranker-v2-m3 is listed for Text ranking. It accepts text and produces label. Use the publisher’s input format and preprocessing for your chosen deployment.
SyncDev model directory · Profile updated Oct 13, 2026
BAAI/bge-reranker-v2-m3 is a source-linked Hugging Face repository assigned to Text classification. Its recorded task interface accepts text and produces label.
The specification score summarizes recorded features and limits. It is not a benchmark result.
Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.
Accepts
Produces
BAAI/bge-reranker-v2-m3 is a Hugging Face repository returned by the public Text classification task feed. Its recorded download count is 19,219,519 and it has 1,119 likes when the repository snapshot was collected. Popularity is useful for discovery, but it is not a performance ranking or a guarantee of suitability.
This record is categorized for Text classification. A task category describes the intended machine-learning problem; it does not prove that every version, checkpoint, quantization, or deployment method produces the same quality. Teams should read the model card and test representative inputs before choosing it.
The public repository lists sentence-transformers as its library metadata. Its recorded tags include sentence-transformers, safetensors, xlm-roberta, text-classification, transformers, text-embeddings-inference, multilingual, arxiv:2312.15503. These source details are retained so readers can trace this directory entry back to the repository instead of relying on copied descriptions.
Use a held-out set from your own domain. Check per-class precision and recall, ambiguous examples, class imbalance and the meaning of every returned label; an overall accuracy score can hide the errors that matter most.
The Hugging Face task assignment records text as input and label as output for BAAI/bge-reranker-v2-m3. This interface summary comes from the repository’s recorded Text classification task classification. It helps readers distinguish a text, image, audio, video, document, embedding, or prediction workflow without inferring undocumented API behavior. Model wrappers can expose different preprocessing and response formats, so verify the upstream model card and the exact runtime before deployment.
Explore BAAI models and the text classification collection. Use the comparison tool to inspect a second candidate.
Compare the model on the work you need to complete.
BAAI/bge-reranker-v2-m3 is listed for Text ranking. It accepts text and produces label. Use the publisher’s input format and preprocessing for your chosen deployment.
Documented limits: 8.19K context and 1 maximum output. The listed route charges $0.01 input and $0.00 output per million tokens. Measure a completed task using the prompts, documents and tools your application will use. Include reasoning and retries in the result.
The weights are downloadable under the publisher’s model licence. Check runtime support and size the complete model and working context for your expected concurrency.
These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.
Qwen
Qwen/Qwen3-Reranker-4B is a source-linked Hugging Face repository assigned to Text ranking. Its recorded task interface accepts text and produces score.
Qwen
Qwen/Qwen3-Reranker-8B is a source-linked Hugging Face repository assigned to Text ranking. Its recorded task interface accepts text and produces score.
Alibaba-NLP
Alibaba-NLP/gte-multilingual-reranker-base is a source-linked Hugging Face repository assigned to Text ranking. Its recorded task interface accepts text and produces score.
Alibaba-NLP
Alibaba-NLP/gte-reranker-modernbert-base is a source-linked Hugging Face repository assigned to Text ranking. Its recorded task interface accepts text and produces score.
antoinelouis
Hugging Face repository for Text ranking.
cl-nagoya
cl-nagoya/ruri-v3-reranker-310m is a source-linked Hugging Face repository assigned to Text ranking. Its recorded task interface accepts text and produces score.