tokyotech-llm/Swallow-7b-NVE-instruct-hf
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kLicense:llama2Architecture:Transformer0.0K Open Weights Cold

The Swallow-7b-NVE-instruct-hf model by tokyotech-llm is a 7 billion parameter instruction-tuned causal language model, continually pre-trained from the Llama 2 family with a focus on Japanese language data. This specific variant, NVE (No Vocabulary Expansion), uses a standard tokenizer without broadened vocabulary, distinguishing it from other Swallow models. It is designed for instruction-following tasks in both Japanese and English, demonstrating strong performance in Japanese benchmarks.

Loading preview...