Spaces:
Runtime error
Runtime error
File size: 9,974 Bytes
ed4d993 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 |
"""Callback Handler that prints to std out."""
import threading
from typing import Any, Dict, List
from langchain_core.callbacks import BaseCallbackHandler
from langchain_core.messages import AIMessage
from langchain_core.outputs import ChatGeneration, LLMResult
MODEL_COST_PER_1K_TOKENS = {
# GPT-4o input
"gpt-4o": 0.005,
"gpt-4o-2024-05-13": 0.005,
# GPT-4o output
"gpt-4o-completion": 0.015,
"gpt-4o-2024-05-13-completion": 0.015,
# GPT-4 input
"gpt-4": 0.03,
"gpt-4-0314": 0.03,
"gpt-4-0613": 0.03,
"gpt-4-32k": 0.06,
"gpt-4-32k-0314": 0.06,
"gpt-4-32k-0613": 0.06,
"gpt-4-vision-preview": 0.01,
"gpt-4-1106-preview": 0.01,
"gpt-4-0125-preview": 0.01,
"gpt-4-turbo-preview": 0.01,
"gpt-4-turbo": 0.01,
"gpt-4-turbo-2024-04-09": 0.01,
# GPT-4 output
"gpt-4-completion": 0.06,
"gpt-4-0314-completion": 0.06,
"gpt-4-0613-completion": 0.06,
"gpt-4-32k-completion": 0.12,
"gpt-4-32k-0314-completion": 0.12,
"gpt-4-32k-0613-completion": 0.12,
"gpt-4-vision-preview-completion": 0.03,
"gpt-4-1106-preview-completion": 0.03,
"gpt-4-0125-preview-completion": 0.03,
"gpt-4-turbo-preview-completion": 0.03,
"gpt-4-turbo-completion": 0.03,
"gpt-4-turbo-2024-04-09-completion": 0.03,
# GPT-3.5 input
# gpt-3.5-turbo points at gpt-3.5-turbo-0613 until Feb 16, 2024.
# Switches to gpt-3.5-turbo-0125 after.
"gpt-3.5-turbo": 0.0015,
"gpt-3.5-turbo-0125": 0.0005,
"gpt-3.5-turbo-0301": 0.0015,
"gpt-3.5-turbo-0613": 0.0015,
"gpt-3.5-turbo-1106": 0.001,
"gpt-3.5-turbo-instruct": 0.0015,
"gpt-3.5-turbo-16k": 0.003,
"gpt-3.5-turbo-16k-0613": 0.003,
# GPT-3.5 output
# gpt-3.5-turbo points at gpt-3.5-turbo-0613 until Feb 16, 2024.
# Switches to gpt-3.5-turbo-0125 after.
"gpt-3.5-turbo-completion": 0.002,
"gpt-3.5-turbo-0125-completion": 0.0015,
"gpt-3.5-turbo-0301-completion": 0.002,
"gpt-3.5-turbo-0613-completion": 0.002,
"gpt-3.5-turbo-1106-completion": 0.002,
"gpt-3.5-turbo-instruct-completion": 0.002,
"gpt-3.5-turbo-16k-completion": 0.004,
"gpt-3.5-turbo-16k-0613-completion": 0.004,
# Azure GPT-35 input
"gpt-35-turbo": 0.0015, # Azure OpenAI version of ChatGPT
"gpt-35-turbo-0301": 0.0015, # Azure OpenAI version of ChatGPT
"gpt-35-turbo-0613": 0.0015,
"gpt-35-turbo-instruct": 0.0015,
"gpt-35-turbo-16k": 0.003,
"gpt-35-turbo-16k-0613": 0.003,
# Azure GPT-35 output
"gpt-35-turbo-completion": 0.002, # Azure OpenAI version of ChatGPT
"gpt-35-turbo-0301-completion": 0.002, # Azure OpenAI version of ChatGPT
"gpt-35-turbo-0613-completion": 0.002,
"gpt-35-turbo-instruct-completion": 0.002,
"gpt-35-turbo-16k-completion": 0.004,
"gpt-35-turbo-16k-0613-completion": 0.004,
# Others
"text-ada-001": 0.0004,
"ada": 0.0004,
"text-babbage-001": 0.0005,
"babbage": 0.0005,
"text-curie-001": 0.002,
"curie": 0.002,
"text-davinci-003": 0.02,
"text-davinci-002": 0.02,
"code-davinci-002": 0.02,
# Fine Tuned input
"babbage-002-finetuned": 0.0016,
"davinci-002-finetuned": 0.012,
"gpt-3.5-turbo-0613-finetuned": 0.003,
"gpt-3.5-turbo-1106-finetuned": 0.003,
"gpt-3.5-turbo-0125-finetuned": 0.003,
# Fine Tuned output
"babbage-002-finetuned-completion": 0.0016,
"davinci-002-finetuned-completion": 0.012,
"gpt-3.5-turbo-0613-finetuned-completion": 0.006,
"gpt-3.5-turbo-1106-finetuned-completion": 0.006,
"gpt-3.5-turbo-0125-finetuned-completion": 0.006,
# Azure Fine Tuned input
"babbage-002-azure-finetuned": 0.0004,
"davinci-002-azure-finetuned": 0.002,
"gpt-35-turbo-0613-azure-finetuned": 0.0015,
# Azure Fine Tuned output
"babbage-002-azure-finetuned-completion": 0.0004,
"davinci-002-azure-finetuned-completion": 0.002,
"gpt-35-turbo-0613-azure-finetuned-completion": 0.002,
# Legacy fine-tuned models
"ada-finetuned-legacy": 0.0016,
"babbage-finetuned-legacy": 0.0024,
"curie-finetuned-legacy": 0.012,
"davinci-finetuned-legacy": 0.12,
}
def standardize_model_name(
model_name: str,
is_completion: bool = False,
) -> str:
"""
Standardize the model name to a format that can be used in the OpenAI API.
Args:
model_name: Model name to standardize.
is_completion: Whether the model is used for completion or not.
Defaults to False.
Returns:
Standardized model name.
"""
model_name = model_name.lower()
if ".ft-" in model_name:
model_name = model_name.split(".ft-")[0] + "-azure-finetuned"
if ":ft-" in model_name:
model_name = model_name.split(":")[0] + "-finetuned-legacy"
if "ft:" in model_name:
model_name = model_name.split(":")[1] + "-finetuned"
if is_completion and (
model_name.startswith("gpt-4")
or model_name.startswith("gpt-3.5")
or model_name.startswith("gpt-35")
or ("finetuned" in model_name and "legacy" not in model_name)
):
return model_name + "-completion"
else:
return model_name
def get_openai_token_cost_for_model(
model_name: str, num_tokens: int, is_completion: bool = False
) -> float:
"""
Get the cost in USD for a given model and number of tokens.
Args:
model_name: Name of the model
num_tokens: Number of tokens.
is_completion: Whether the model is used for completion or not.
Defaults to False.
Returns:
Cost in USD.
"""
model_name = standardize_model_name(model_name, is_completion=is_completion)
if model_name not in MODEL_COST_PER_1K_TOKENS:
raise ValueError(
f"Unknown model: {model_name}. Please provide a valid OpenAI model name."
"Known models are: " + ", ".join(MODEL_COST_PER_1K_TOKENS.keys())
)
return MODEL_COST_PER_1K_TOKENS[model_name] * (num_tokens / 1000)
class OpenAICallbackHandler(BaseCallbackHandler):
"""Callback Handler that tracks OpenAI info."""
total_tokens: int = 0
prompt_tokens: int = 0
completion_tokens: int = 0
successful_requests: int = 0
total_cost: float = 0.0
def __init__(self) -> None:
super().__init__()
self._lock = threading.Lock()
def __repr__(self) -> str:
return (
f"Tokens Used: {self.total_tokens}\n"
f"\tPrompt Tokens: {self.prompt_tokens}\n"
f"\tCompletion Tokens: {self.completion_tokens}\n"
f"Successful Requests: {self.successful_requests}\n"
f"Total Cost (USD): ${self.total_cost}"
)
@property
def always_verbose(self) -> bool:
"""Whether to call verbose callbacks even if verbose is False."""
return True
def on_llm_start(
self, serialized: Dict[str, Any], prompts: List[str], **kwargs: Any
) -> None:
"""Print out the prompts."""
pass
def on_llm_new_token(self, token: str, **kwargs: Any) -> None:
"""Print out the token."""
pass
def on_llm_end(self, response: LLMResult, **kwargs: Any) -> None:
"""Collect token usage."""
# Check for usage_metadata (langchain-core >= 0.2.2)
try:
generation = response.generations[0][0]
except IndexError:
generation = None
if isinstance(generation, ChatGeneration):
try:
message = generation.message
if isinstance(message, AIMessage):
usage_metadata = message.usage_metadata
else:
usage_metadata = None
except AttributeError:
usage_metadata = None
else:
usage_metadata = None
if usage_metadata:
token_usage = {"total_tokens": usage_metadata["total_tokens"]}
completion_tokens = usage_metadata["output_tokens"]
prompt_tokens = usage_metadata["input_tokens"]
if response.llm_output is None:
# model name (and therefore cost) is unavailable in
# streaming responses
model_name = ""
else:
model_name = standardize_model_name(
response.llm_output.get("model_name", "")
)
else:
if response.llm_output is None:
return None
if "token_usage" not in response.llm_output:
with self._lock:
self.successful_requests += 1
return None
# compute tokens and cost for this request
token_usage = response.llm_output["token_usage"]
completion_tokens = token_usage.get("completion_tokens", 0)
prompt_tokens = token_usage.get("prompt_tokens", 0)
model_name = standardize_model_name(
response.llm_output.get("model_name", "")
)
if model_name in MODEL_COST_PER_1K_TOKENS:
completion_cost = get_openai_token_cost_for_model(
model_name, completion_tokens, is_completion=True
)
prompt_cost = get_openai_token_cost_for_model(model_name, prompt_tokens)
else:
completion_cost = 0
prompt_cost = 0
# update shared state behind lock
with self._lock:
self.total_cost += prompt_cost + completion_cost
self.total_tokens += token_usage.get("total_tokens", 0)
self.prompt_tokens += prompt_tokens
self.completion_tokens += completion_tokens
self.successful_requests += 1
def __copy__(self) -> "OpenAICallbackHandler":
"""Return a copy of the callback handler."""
return self
def __deepcopy__(self, memo: Any) -> "OpenAICallbackHandler":
"""Return a deep copy of the callback handler."""
return self
|