使用模式

估算LLM与嵌入令牌数量

为了测量LLM和嵌入的令牌数量，您需要

设置 MockLLM 和 MockEmbedding 对象

from llama_index.core.llms import MockLLM
from llama_index.core import MockEmbedding

llm = MockLLM(max_tokens=256)
embed_model = MockEmbedding(embed_dim=1536)

设置 TokenCountingCallback 处理器

import tiktoken
from llama_index.core.callbacks import CallbackManager, TokenCountingHandler

token_counter = TokenCountingHandler(
    tokenizer=tiktoken.encoding_for_model("gpt-3.5-turbo").encode
)

callback_manager = CallbackManager([token_counter])

将它们添加到全局 SettingsSettings

from llama_index.core import Settings

Settings.llm = llm
Settings.embed_model = embed_model
Settings.callback_manager = callback_manager

构建索引

from llama_index.core import VectorStoreIndex, SimpleDirectoryReader

documents = SimpleDirectoryReader(
    "./docs/examples/data/paul_graham"
).load_data()

index = VectorStoreIndex.from_documents(documents)

统计数量！

print(
    "Embedding Tokens: ",
    token_counter.total_embedding_token_count,
    "\n",
    "LLM Prompt Tokens: ",
    token_counter.prompt_llm_token_count,
    "\n",
    "LLM Completion Tokens: ",
    token_counter.completion_llm_token_count,
    "\n",
    "Total LLM Token Count: ",
    token_counter.total_llm_token_count,
    "\n",
)

# reset counts
token_counter.reset_counts()

运行查询，再次测量

query_engine = index.as_query_engine()

response = query_engine.query("query")

print(
    "Embedding Tokens: ",
    token_counter.total_embedding_token_count,
    "\n",
    "LLM Prompt Tokens: ",
    token_counter.prompt_llm_token_count,
    "\n",
    "LLM Completion Tokens: ",
    token_counter.completion_llm_token_count,
    "\n",
    "Total LLM Token Count: ",
    token_counter.total_llm_token_count,
    "\n",
)