146.8K Downloads Updated 17 months ago
An advanced language model crafted with 2 trillion bilingual tokens.
Updated 17 months ago
17 months ago
9aab369a853b · 4.0GB
Readme
DeepSeek LLM is an advanced language model available in both 7 billion and 67 billion parameters. Both a chat
and base
variation are available.
Superior General Capabilities: DeepSeek LLM 67B Base outperforms Llama2 70B Base in areas such as reasoning, coding, math, and Chinese comprehension.
Proficient in Coding and Math: DeepSeek LLM 67B Chat exhibits outstanding performance in coding (using the HumanEval benchmark) and mathematics (using the GSM8K benchmark).