File size: 3,477 Bytes
b350db1 97dbf57 b350db1 a79f08d b350db1 a79f08d b350db1 a79f08d b350db1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 | from flask import Flask, request, jsonify
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
app = Flask(__name__)
# モデルロード（起動時1回）
torch.random.manual_seed(0)
model = AutoModelForCausalLM.from_pretrained(
"microsoft/Phi-3-mini-4k-instruct",
device_map="cpu",
torch_dtype="auto",
trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained(
"microsoft/Phi-3-mini-4k-instruct"
)
pipe = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer
)
generation_args = {
"max_new_tokens": 500,
"return_full_text": False,
"temperature": 0.0,
"do_sample": False,
}
# -----------------------
# ルートページ (HTML)
# -----------------------
@app.route("/")
def index():
return """
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<title>Local LLM Chat</title>
<style>
body{
font-family: Arial;
background:#111;
color:white;
margin:0;
}
#chat{
height:80vh;
overflow-y:auto;
padding:20px;
}
.message{
margin-bottom:12px;
}
.user{
color:#6cf;
}
.assistant{
color:#9f9;
}
#inputArea{
position:fixed;
bottom:0;
width:100%;
background:#222;
padding:10px;
}
#input{
width:80%;
padding:10px;
font-size:16px;
}
button{
padding:10px;
font-size:16px;
}
</style>
</head>
<body>
<h2 style="padding:10px;">Local Phi-3 Chat</h2>
<div id="chat"></div>
<div id="inputArea">
<input id="input" placeholder="メッセージを入力..." />
<button onclick="send()">送信</button>
</div>
<script>
let messages = [
{role:"system",content:"You are a helpful assistant."}
]
function add(role,text){
const chat=document.getElementById("chat")
const div=document.createElement("div")
div.className="message "+role
div.innerText=role+": "+text
chat.appendChild(div)
chat.scrollTop=chat.scrollHeight
}
async function send(){
const input=document.getElementById("input")
const text=input.value
if(!text) return
input.value=""
add("user",text)
messages.push({
role:"user",
content:text
})
const res=await fetch("/v1/chat/completions",{
method:"POST",
headers:{
"Content-Type":"application/json"
},
body:JSON.stringify({
messages:messages
})
})
const data=await res.json()
const reply=data.choices[0].message.content
add("assistant",reply)
messages.push({
role:"assistant",
content:reply
})
}
document.getElementById("input").addEventListener("keypress",function(e){
if(e.key==="Enter"){
send()
}
})
</script>
</body>
</html>
"""
# -----------------------
# OpenAI互換API
# -----------------------
@app.route("/v1/chat/completions", methods=["POST"])
def chat_completions():
data = request.json
messages = data.get("messages", [])
result = pipe(messages, **generation_args)
text = result[0]["generated_text"]
response = {
"id": "chatcmpl-local",
"object": "chat.completion",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": text
},
"finish_reason": "stop"
}
]
}
return jsonify(response)
if __name__ == "__main__":
app.run(host="0.0.0.0", port=7860) |