deepseek-llm

An advanced language model crafted with 2 trillion bilingual tokens.

14.7K Pulls Updated 4 months ago

64 Tags

5 months ago

9aab369a853b · 4.0GB

model
llama
·
7B
·
4-bit
params
{"num_ctx":4096}
template
{{ .System }} User: {{ .Prompt }} Assistant:

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).

References

GitHub

HuggingFace