How AI describes GTE-Qwen2-7B-instruct

gte-Qwen2-7B-instruct-Q8_0 is a General Text Embeddings (GTE) model in the gte family, built on the Qwen2-7B base and fine-tuned for instruction-following to produce high-quality embeddings. It is quantized to 8-bit GGUF format (q8_0) and can be used with llama.cpp to compute embeddings on CPU or GPU, with options for CLI and server deployment. The model has 7B parameters, 3584-dimensional embeddings, a 131072-token context, multilingual training, and ranked No.1 in English and Chinese MTEB evaluations as of June 16, 2024.