- Cloudflare-hosted
- Deprecated
Meta 推出的 80 亿参数量化(int4)生成式文本模型。
| 模型信息 | |
|---|---|
| 已弃用 | 2026/5/30 |
| 上下文窗口 ↗ | 8,192 个 token |
| 条款和许可证 | 链接 ↗ |
| 单位定价 | $0.12 per M input tokens, $0.27 per M output tokens |
试用环境
使用 Workers AI LLM Playground 试用此模型。无需设置或身份验证,即可直接在浏览器中预览和测试模型。
启动 LLM Playground用法
export interface Env {
AI: Ai;
}
export default {
async fetch(request, env): Promise<Response> {
const messages = [
{ role: "system", content: "You are a friendly assistant" },
{
role: "user",
content: "What is the origin of the phrase Hello, World",
},
];
const stream = await env.AI.run("@cf/meta/llama-3.1-8b-instruct-awq", {
messages,
stream: true,
});
return new Response(stream, {
headers: { "content-type": "text/event-stream" },
});
},
} satisfies ExportedHandler<Env>;
export interface Env {
AI: Ai;
}
export default {
async fetch(request, env): Promise<Response> {
const messages = [
{ role: "system", content: "You are a friendly assistant" },
{
role: "user",
content: "What is the origin of the phrase Hello, World",
},
];
const response = await env.AI.run("@cf/meta/llama-3.1-8b-instruct-awq", { messages });
return Response.json(response);
},
} satisfies ExportedHandler<Env>;
import os
import requests
ACCOUNT_ID = "your-account-id"
AUTH_TOKEN = os.environ.get("CLOUDFLARE_AUTH_TOKEN")
prompt = "Tell me all about PEP-8"
response = requests.post(
f"https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai/run/@cf/meta/llama-3.1-8b-instruct-awq",
headers={"Authorization": f"Bearer {AUTH_TOKEN}"},
json={
"messages": [
{"role": "system", "content": "You are a friendly assistant"},
{"role": "user", "content": prompt}
]
}
)
result = response.json()
print(result)
curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run/@cf/meta/llama-3.1-8b-instruct-awq \
-X POST \
-H "Authorization: Bearer $CLOUDFLARE_AUTH_TOKEN" \
-d '{ "messages": [{ "role": "system", "content": "You are a friendly assistant" }, { "role": "user", "content": "Why is pizza so good" }]}'参数
输入
prompt
stringrequiredminLength: 1供模型生成响应的输入文本提示词。lora
string用于微调基础模型的 LoRA(Low-Rank Adaptation,低秩适配)模型名称。▶response_format{}
objectraw
booleandefault: false若为 true,则不应用聊天模板,您必须遵循特定模型所要求的格式。stream
booleandefault: false若为 true,响应将通过 SSE(Server Sent Events,服务器发送事件)以增量方式流式返回。max_tokens
integerdefault: 256响应中生成的最大 token 数。temperature
numberdefault: 0.6minimum: 0maximum: 5控制输出的随机性;值越高,结果越随机。top_p
numberminimum: 0.001maximum: 1通过控制 AI 考虑的候选词数量来调整其响应的创造性。值越低,输出越可预测;值越高,响应越多样、越有创意。top_k
integerminimum: 1maximum: 50限制 AI 仅从概率最高的前 'k' 个词中选择。值越低,响应越集中;值越高,响应越多样,也可能带来更多意外结果。seed
integerminimum: 1maximum: 9999999999用于复现生成结果的随机种子。repetition_penalty
numberminimum: 0maximum: 2对重复 token 的惩罚;值越高,越能抑制重复。frequency_penalty
numberminimum: -2maximum: 2降低模型逐字重复相同内容的可能性。presence_penalty
numberminimum: -2maximum: 2提高模型引入新话题的可能性。输出
Synchronous — Send a request and receive a complete response
response
string模型生成的文本响应▶usage{}
object推理请求的用量统计▶tool_calls[]
array生成响应期间发出的工具调用请求数组Streaming — Send a request with `stream: true` and receive server-sent events
type
stringformat
binary