FFM-Llama2-v2 - 7B, 13B, 70B (incl. FFM-Llama2)
FFM-Llama2 series models are enhanced Traditional Chinese versions of large language models based on the open-source large language model Meta Llama 2 launched by Meta.
FFM-Llama2-v2 introduces an expanded vocabulary list, significantly enhancing the comprehension of Chinese articles, improving the quality of English poetry translation, and increasing the efficiency of model inference and response speed, thereby saving computational resources.
When processing the same Chinese text, FFM-Llama2-v2 generates approximately 50% fewer tokens per character compared to FFM-Llama2, improving the token calculation efficiency by approximately 2 times.
For the characteristics and specifications of the FFM-Llama2-v2 series models, please refer to this document.
By using the FFM-Llama2-v2 and FFM-Llama2 series models, you agree to comply with the Meta-Llama2 licensing terms.
The FFM-Llama2-v2 series models can be fine-tuned in AFS Platform and deployed in AFS Cloud, AFS ModelSpace public mode and private mode.
- For instructions on AFS Platform, please refer to this document.
- For instructions on AFS Cloud, please refer to this document.
- For instructions on AFS ModelSpace public mode, please refer to this document.
- For instructions on AFS ModelSpace private mode, please refer to this document.
- For instructions on Chat and Playground Interfaces in AFS ModelSpace private mode, please refer to this document.
- From April 17, 2024, the FFM-Llama2-v2 series models deployed in AFS MS and AFS Cloud now support Function Calling. For more instructions on Function Calling, please refer to this document.