deepseek-v3:671b-q8

deepseek-v3:671b-q8_0

3.3M Downloads Updated 1 year ago

A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.

671b

ollama run deepseek-v3:671b-q8_0

curl http://localhost:11434/api/chat \
  -d '{
    "model": "deepseek-v3:671b-q8_0",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='deepseek-v3:671b-q8_0',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'deepseek-v3:671b-q8_0',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 1 year ago

1 year ago

96061c74c1a5 · 713GB ·

model

archdeepseek2

parameters671B

quantizationQ8_0

713GB

license

14kB

params

{ "stop": [ "<｜begin▁of▁sentence｜>", "<｜end▁of▁sentence｜>",

148B

template

{{- range $i, $_ := .Messages }} {{- if eq .Role "user" }}<｜User｜> {{- else if eq .Role "assista

359B

Readme

Note: this model requires Ollama 0.5.5 or later.

DeepSeek-V3 achieves a significant breakthrough in inference speed over previous models. It tops the leaderboard among open-source models and rivals the most advanced closed-source models globally.

References

GitHub

Paper