An advanced language model crafted with 2 trillion bilingual tokens.
7B
67B
81.5K Pulls Updated 10 months ago
Updated 10 months ago
10 months ago
e825785a2f93 · 28GB
model
archllama
·
parameters67.4B
·
quantizationQ2_K
28GB
params
{"num_ctx":4096}
17B
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).