mistrallite:7b-v0.1-q3_K

mistrallite:7b-v0.1-q3_K_L

158.2K Downloads Updated 2 years ago

MistralLite is a fine-tuned model based on Mistral with enhanced capabilities of processing long contexts.

ollama run mistrallite:7b-v0.1-q3_K_L

curl http://localhost:11434/api/chat \
  -d '{
    "model": "mistrallite:7b-v0.1-q3_K_L",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='mistrallite:7b-v0.1-q3_K_L',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'mistrallite:7b-v0.1-q3_K_L',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 2 years ago

2 years ago

bb7562f0532f · 3.8GB ·

model

archllama

parameters7.24B

quantizationQ3_K_L

3.8GB

params

{ "num_ctx": 32768, "stop": [ "<|prompter|>", "<|assistant|>", "</s>

94B

template

<|prompter|>{{ .System }} {{ .Prompt }}</s><|assistant|>

57B

Readme

Mistrallite is a fine-tuned model based on Mistral, with enhanced capabilities of processing long context (up to 32K tokens). This model is able to perform significantly better on several long context retrieve and answering tasks.

Reference

Hugging Face