PyTokenCounter

PyTokenCounter is a Python library designed to simplify text tokenization and token counting. It supports various encoding schemes, with a focus on those used by Large Language Models (LLMs), particularly those developed by OpenAI. Leveraging the tiktoken library for efficient processing, PyTokenCounter facilitates seamless integration with LLM workflows. This project is based on the tiktoken library created by OpenAI.

Table of Contents

• Background
• Install
• Usage
  • CLI
• API
  • Utility Functions
  • String Tokenization and Counting
  • File and Directory Tokenization and Counting
  • Token Mapping
• Ignored Files
• Maintainers
• Acknowledgements
• Contributing
• License

Background

The development of PyTokenCounter was driven by the need for a user-friendly and efficient way to handle text tokenization in Python, particularly for applications that interact with Large Language Models (LLMs) like OpenAI's language models. LLMs process text by breaking it down into tokens, which are the fundamental units of input and output for these models. Tokenization, the process of converting text into a sequence of tokens, is a fundamental step in natural language processing and essential for optimizing interactions with LLMs.

Understanding and managing token counts is crucial when working with LLMs because it directly impacts aspects such as API usage costs, prompt length limitations, and response generation. PyTokenCounter addresses these needs by providing an intuitive interface for tokenizing strings, files, and directories, as well as counting the number of tokens based on different encoding schemes. With support for various OpenAI models and their associated encodings, PyTokenCounter is versatile enough to be used in a wide range of applications involving LLMs, such as prompt engineering, cost estimation, and monitoring usage.

Install

Install PyTokenCounter using pip:

pip install PyTokenCounter

To run the project's test suite, install the optional test dependencies:

pip install "PyTokenCounter[test]"

Usage

Here are a few examples to get you started with PyTokenCounter, especially in the context of LLMs:

from pathlib import Path
from collections import OrderedDict

import PyTokenCounter as tc
import tiktoken

# Count tokens in a string for an LLM model
numTokens = tc.GetNumTokenStr(
    string="This is a test string.", model="gpt-4o"
)
print(f"Number of tokens: {numTokens}")

# Count tokens in a file intended for LLM processing
filePath = Path("./TestFile.txt")
numTokensFile = tc.GetNumTokenFile(filePath=filePath, model="gpt-4o")
print(f"Number of tokens in file: {numTokensFile}")

# Count tokens in a directory of documents for batch processing with an LLM
dirPath = Path("./TestDir")
numTokensDir = tc.GetNumTokenDir(dirPath=dirPath, model="gpt-4o", recursive=True)
print(f"Number of tokens in directory: {numTokensDir}")

# Get the encoding for a specific LLM model
encoding = tc.GetEncoding(model="gpt-4o")

# Tokenize a string using a specific encoding for LLM input
tokens = tc.TokenizeStr(string="This is another test.", encoding=encoding)
print(f"Token IDs: {tokens}")

# Map tokens to their decoded strings
mappedTokens = tc.MapTokens(tokens=tokens, encoding=encoding)
print(f"Mapped tokens: {mappedTokens}")

# Count tokens in a string using the default model
numTokens = tc.GetNumTokenStr(string="This is a test string.")
print(f"Number of tokens: {numTokens}")

# Count tokens in a file using the default model
filePath = Path("./TestFile.txt")
numTokensFile = tc.GetNumTokenFile(filePath=filePath)
print(f"Number of tokens in file: {numTokensFile}")

# Tokenize a string using the default model
tokens = tc.TokenizeStr(string="This is another test.")
print(f"Token IDs: {tokens}")

# Tokenize a string and map tokens to strings using the default model
mappedTokensResult = tc.TokenizeStr(string="This is another test.", mapTokens=True)
print(f"Mapped tokens result: {mappedTokensResult}")

# Map tokens to their decoded strings using the default model
mappedTokens = tc.MapTokens(tokens=tokens)
print(f"Mapped tokens: {mappedTokens}")

# Tokenize a directory and get mapped tokens with counts
dirPath = Path("./TestDir")
mappedDirTokens = tc.TokenizeDir(dirPath=dirPath, recursive=True, mapTokens=True)
print(f"Mapped directory tokens: {mappedDirTokens}")

# Count tokens in a directory and get mapped counts
mappedDirCounts = tc.GetNumTokenDir(dirPath=dirPath, recursive=True, mapTokens=True)
print(f"Mapped directory counts: {mappedDirCounts}")

CLI

PyTokenCounter can also be used as a command-line tool. Install the package and invoke the pytokencount entry point:

# Tokenize a string for an LLM
pytokencount tokenize-str "Hello, world!" --model gpt-4o

# Tokenize a string using the default model
pytokencount tokenize-str "Hello, world!"

# Tokenize a file for an LLM
pytokencount tokenize-file TestFile.txt --model gpt-4o

# Tokenize multiple files using a wildcard pattern
pytokencount tokenize-files "*.txt"

# Tokenize a directory of files for an LLM (non-recursive)
pytokencount tokenize-files MyDirectory --model gpt-4o --no-recursive

# Count tokens in a file for an LLM
pytokencount count-file TestFile.txt --model gpt-4o

# Count tokens in a directory (non-recursive)
pytokencount count-dir TestDir --model gpt-4o --no-recursive

# Get the encoding associated with a model
pytokencount get-encoding gpt-4o

# Map tokens to strings using the default model
pytokencount map-tokens 123,456,789

# Include binary files and hidden files while tokenizing
pytokencount tokenize-files MyDirectory --model gpt-4o -b -H

CLI Usage Details:

The pytokencount CLI provides several subcommands for tokenizing and counting tokens in strings, files, and directories, tailored for use with LLMs.

Subcommands:

• tokenize-str: Tokenizes a provided string.
  • pytokencount tokenize-str "Your string here" --model gpt-4o
• tokenize-file: Tokenizes the contents of a file.
  • pytokencount tokenize-file Path/To/Your/File.txt --model gpt-4o
  • Supports wildcard patterns, e.g., pytokencount tokenize-file "*.txt"
• tokenize-files: Tokenizes the contents of multiple specified files or all files within a directory.
  • pytokencount tokenize-files Path/To/Your/File1.txt Path/To/Your/File2.txt --model gpt-4o
  • pytokencount tokenize-files Path/To/Your/Directory --model gpt-4o --no-recursive
  • Wildcards are allowed, e.g., pytokencount tokenize-files "*.txt"
• tokenize-dir: Tokenizes all files within a specified directory into lists of token IDs.
  • pytokencount tokenize-dir Path/To/Your/Directory --model gpt-4o --no-recursive
• count-str: Counts the number of tokens in a provided string.
  • pytokencount count-str "Your string here" --model gpt-4o
• count-file: Counts the number of tokens in a file.
  • pytokencount count-file Path/To/Your/File.txt --model gpt-4o
  • Supports wildcard patterns, e.g., pytokencount count-file "*.txt"
• count-files: Counts the number of tokens in multiple specified files or all files within a directory.
  • pytokencount count-files Path/To/Your/File1.txt Path/To/Your/File2.txt --model gpt-4o
  • pytokencount count-files Path/To/Your/Directory --model gpt-4o --no-recursive
  • Wildcards are allowed, e.g., pytokencount count-files "*.txt"
• count-dir: Counts the total number of tokens across all files in a specified directory.
  • pytokencount count-dir Path/To/Your/Directory --model gpt-4o --no-recursive
• get-model: Retrieves the model name from the provided encoding.
  • pytokencount get-model cl100k_base
• get-encoding: Retrieves the encoding name from the provided model.
  • pytokencount get-encoding gpt-4o
• map-tokens: Maps a list of token integers to their decoded strings.
  • pytokencount map-tokens 123,456,789 --model gpt-4o

Options:

• -m, --model: Specifies the model to use for encoding. Default: gpt-4o
• -e, --encoding: Specifies the encoding to use directly.
• -nr, --no-recursive: When used with tokenize-files, tokenize-dir, count-files, or count-dir for a directory, it prevents the tool from processing subdirectories recursively.
• -q, --quiet: When used with any of the above commands, it prevents the tool from showing progress bars and minimizes output.
• -M, --mapTokens: When specified, the output will be in a mapped (nested) format. For tokenize commands, this outputs a nested OrderedDict mapping decoded strings to their token IDs. For count commands, this outputs a nested OrderedDict with token counts, including keys such as "numTokens" and "tokens".
• -o, --output: When used with any of the commands, specifies an output JSON file to save the results to.
• -b, --include-binary: Include binary files in processing. (Default: binary files are excluded.)
• -H, --include-hidden: Include hidden files and directories. (Default: hidden files and directories are skipped.)

API

Here's a detailed look at the PyTokenCounter API, designed to integrate seamlessly with LLM workflows:

Utility Functions

GetModelMappings() -> dict

Retrieves the mappings between models and their corresponding encodings, essential for selecting the correct tokenization strategy for different LLMs.

Returns:

• dict: A dictionary where keys are model names and values are their corresponding encodings.

Example:

import PyTokenCounter as tc

modelMappings = tc.GetModelMappings()
print(modelMappings)

---

GetValidModels() -> list[str]

Returns a list of valid model names supported by PyTokenCounter, primarily focusing on LLMs.

Returns:

• list[str]: A list of valid model names.

Example:

import PyTokenCounter as tc

validModels = tc.GetValidModels()
print(validModels)

---

GetValidEncodings() -> list[str]

Returns a list of valid encoding names, ensuring compatibility with various LLMs.

Returns:

• list[str]: A list of valid encoding names.

Example:

import PyTokenCounter as tc

validEncodings = tc.GetValidEncodings()
print(validEncodings)

---

GetModelForEncoding(encoding: tiktoken.Encoding) -> list[str] | str

Determines the model name(s) associated with a given encoding, facilitating the selection of appropriate LLMs.

Parameters:

• encoding (tiktoken.Encoding): The encoding to get the model for.

Returns:

• str: The model name or a list of models corresponding to the given encoding.

Raises:

• ValueError: If the encoding name is not valid.

Example:

import PyTokenCounter as tc
import tiktoken

encoding = tiktoken.get_encoding('cl100k_base')
model = tc.GetModelForEncoding(encoding=encoding)
print(model)

---

GetModelForEncodingName(encodingName: str) -> str

Determines the model name associated with a given encoding name, facilitating the selection of appropriate LLMs.

Parameters:

• encodingName (str): The name of the encoding.

Returns:

• str: The model name or a list of models corresponding to the given encoding.

Raises:

• ValueError: If the encoding name is not valid.

Example:

import PyTokenCounter as tc

modelName = tc.GetModelForEncodingName(encodingName="cl100k_base")
print(modelName)

---

GetEncodingForModel(modelName: str) -> tiktoken.Encoding

Retrieves the encoding associated with a given model name, ensuring accurate tokenization for the selected LLM.

Parameters:

• modelName (str): The name of the model.

Returns:

• tiktoken.Encoding: The encoding corresponding to the given model name.

Raises:

• ValueError: If the model name is not valid.

Example:

import PyTokenCounter as tc

encoding = tc.GetEncodingForModel(modelName="gpt-4o")
print(encoding)

---

GetEncodingNameForModel(modelName: str) -> str

Retrieves the encoding name associated with a given model name, ensuring accurate tokenization for the selected LLM.

Parameters:

• modelName (str): The name of the model.

Returns:

• str: The encoding name corresponding to the given model name.

Raises:

• ValueError: If the model name is not valid.

Example:

import PyTokenCounter as tc

encodingName = tc.GetEncodingNameForModel(modelName="gpt-4o")
print(encodingName)

---

GetEncoding(model: str | None = None, encodingName: str | None = None) -> tiktoken.Encoding

Obtains the tiktoken encoding based on the specified model or encoding name, tailored for LLM usage. If neither model nor encodingName is provided, it defaults to the encoding associated with the "gpt-4o" model.

Parameters:

• model (str, optional): The name of the model.
• encodingName (str, optional): The name of the encoding.

Returns:

• tiktoken.Encoding: The tiktoken encoding object.

Raises:

• ValueError: If neither model nor encoding is provided, or if the provided model or encoding is invalid.

Example:

import PyTokenCounter as tc
import tiktoken

encoding = tc.GetEncoding(model="gpt-4o")
print(encoding)
encoding = tc.GetEncoding(encodingName="p50k_base")
print(encoding)
encoding = tc.GetEncoding()
print(encoding)

---

String Tokenization and Counting

TokenizeStr(string: str, model: str | None = "gpt-4o", encodingName: str | None = None, encoding: tiktoken.Encoding | None = None, quiet: bool = False, mapTokens: bool = False) -> list[int] | OrderedDict[str, int]

Tokenizes a string into a list of token IDs or a mapping of decoded strings to tokens, preparing text for input into an LLM.

Parameters:

• string (str): The string to tokenize.
• model (str, optional): The name of the model. Default: "gpt-4o"
• encodingName (str, optional): The name of the encoding.
• encoding (tiktoken.Encoding, optional): A tiktoken encoding object.
• quiet (bool, optional): If True, suppresses progress updates.
• mapTokens (bool, optional): If True, outputs an OrderedDict mapping decoded strings to their token IDs. Default: False

Returns:

• list[int]: A list of token IDs if mapTokens is False.
• OrderedDict[str, int]: An OrderedDict mapping decoded strings to token IDs if mapTokens is True.

Raises:

• ValueError: If the provided model or encoding is invalid.

Example:

import PyTokenCounter as tc
from collections import OrderedDict

tokens = tc.TokenizeStr(string="Hail to the Victors!", model="gpt-4o")
print(tokens)

tokens = tc.TokenizeStr(string="Hail to the Victors!")
print(tokens)


import tiktoken
encoding = tiktoken.get_encoding("cl100k_base")
mappedTokens = tc.TokenizeStr(string="2024 National Champions", encoding=encoding, mapTokens=True)
print(mappedTokens)

mappedTokens = tc.TokenizeStr(string="2024 National Champions", mapTokens=True)
print(mappedTokens)

---

GetNumTokenStr(string: str, model: str | None = "gpt-4o", encodingName: str | None = None, encoding: tiktoken.Encoding | None = None, quiet: bool = False, mapTokens: bool = False) -> int | OrderedDict[str, int]

Counts the number of tokens in a string.

Parameters:

• string (str): The string to count tokens in.
• model (str, optional): The name of the model. Default: "gpt-4o"
• encodingName (str, optional): The name of the encoding.
• encoding (tiktoken.Encoding, optional): A tiktoken.Encoding object to use for tokenization.
• quiet (bool, optional): If True, suppresses progress updates.
• mapTokens (bool, optional): If True, outputs an OrderedDict mapping decoded strings to token counts (which are always 1 for strings). Primarily for consistency with other functions. Default: False

Returns:

• int: The number of tokens in the string if mapTokens is False.
• OrderedDict[str, int]: An OrderedDict mapping decoded strings to token counts if mapTokens is True.

Raises:

• ValueError: If the provided model or encoding is invalid.

Example:

import PyTokenCounter as tc
from collections import OrderedDict
import tiktoken

numTokens = tc.GetNumTokenStr(string="Hail to the Victors!", model="gpt-4o")
print(numTokens)

numTokens = tc.GetNumTokenStr(string="Hail to the Victors!")
print(numTokens)

numTokens = tc.GetNumTokenStr(string="Corum 4 Heisman", encoding=tiktoken.get_encoding("cl100k_base"))
print(numTokens)

numTokens = tc.GetNumTokenStr(string="Corum 4 Heisman")
print(numTokens)

mappedCounts = tc.GetNumTokenStr(string="Mapped count example", mapTokens=True)
print(mappedCounts)

---

File and Directory Tokenization and Counting

TokenizeFile(filePath: Path | str, model: str | None = "gpt-4o", encodingName: str | None = None, encoding: tiktoken.Encoding | None = None, quiet: bool = False, mapTokens: bool = False) -> list[int] | OrderedDict[str, OrderedDict[str, int | list[int]]]

Tokenizes the contents of a file into a list of token IDs or a nested OrderedDict structure.

Parameters:

• filePath (Path | str): The path to the file to tokenize.
• model (str, optional): The name of the model to use for encoding. Default: "gpt-4o"
• encodingName (str, optional): The name of the encoding to use.
• encoding (tiktoken.Encoding, optional): An existing tiktoken.Encoding object to use for tokenization.
• quiet (bool, optional): If True, suppresses progress updates.
• mapTokens (bool, optional): If True, outputs an OrderedDict where the key is the filename and the value is another OrderedDict with keys "tokens" (the list of token IDs) and "numTokens" (the total token count). If False, returns just the list of token IDs. Default: False

Returns:

• list[int]: A list of token IDs representing the tokenized file contents if mapTokens is False.
• OrderedDict[str, OrderedDict[str, int | list[int]]]: An OrderedDict as described above if mapTokens is True.

Raises:

• TypeError: If the types of input parameters are incorrect.
• ValueError: If the provided model or encoding is invalid.
• UnsupportedEncodingError: If the file encoding is not supported.
• FileNotFoundError: If the file does not exist.

Example:

from pathlib import Path
import PyTokenCounter as tc
from collections import OrderedDict
import tiktoken

filePath = Path("TestFile1.txt")
tokens = tc.TokenizeFile(filePath=filePath, model="gpt-4o")
print(tokens)

filePath = Path("TestFile1.txt")
mappedTokensFile = tc.TokenizeFile(filePath=filePath, model="gpt-4o", mapTokens=True)
print(mappedTokensFile)

filePath = Path("TestFile1.txt")
tokens = tc.TokenizeFile(filePath=filePath)
print(tokens)

import tiktoken
encoding = tiktoken.get_encoding("p50k_base")
filePath = Path("TestFile2.txt")
mappedTokensFileEncoding = tc.TokenizeFile(filePath=filePath, encoding=encoding, mapTokens=True)
print(mappedTokensFileEncoding)

filePath = Path("TestFile2.txt")
mappedTokensFileDefault = tc.TokenizeFile(filePath=filePath, mapTokens=True)
print(mappedTokensFileDefault)

---

GetNumTokenFile(filePath: Path | str, model: str | None = "gpt-4o", encodingName: str | None = None, encoding: tiktoken.Encoding | None = None, quiet: bool = False, mapTokens: bool = False) -> int | OrderedDict[str, int]

Counts the number of tokens in a file based on the specified model or encoding.

Parameters:

• filePath (Path | str): The path to the file to count tokens for.
• model (str, optional): The name of the model to use for encoding. Default: "gpt-4o"
• encodingName (str, optional): The name of the encoding to use.
• encoding (tiktoken.Encoding, optional): An existing tiktoken.Encoding object to use for tokenization.
• quiet (bool, optional): If True, suppresses progress updates.
• mapTokens (bool, optional): If True, outputs an OrderedDict where the key is the filename and the value is the token count. If False, returns just the token count as an integer. Default: False

Returns:

• int: The number of tokens in the file if mapTokens is False.
• OrderedDict[str, int]: An OrderedDict mapping the filename to its token count if mapTokens is True.

Raises:

• TypeError: If the types of filePath, model, encodingName, or encoding are incorrect.
• ValueError: If the provided model or encodingName is invalid, or if there is a mismatch between the model and encoding name, or between the provided encoding and the derived encoding.
• UnsupportedEncodingError: If the file's encoding cannot be determined.
• FileNotFoundError: If the file does not exist.

Example:

import PyTokenCounter as tc
from pathlib import Path
from collections import OrderedDict

filePath = Path("TestFile1.txt")
numTokens = tc.GetNumTokenFile(filePath=filePath, model="gpt-4o")
print(numTokens)

filePath = Path("TestFile1.txt")
mappedNumTokensFile = tc.GetNumTokenFile(filePath=filePath, model="gpt-4o", mapTokens=True)
print(mappedNumTokensFile)

filePath = Path("TestFile1.txt")
numTokens = tc.GetNumTokenFile(filePath=filePath)
print(numTokens)

filePath = Path("TestFile2.txt")
numTokens = tc.GetNumTokenFile(filePath=filePath, model="gpt-4o")
print(numTokens)

filePath = Path("TestFile2.txt")
numTokens = tc.GetNumTokenFile(filePath=filePath)
print(numTokens)

filePath = Path("TestFile2.txt")
mappedNumTokensFileDefault = tc.GetNumTokenFile(filePath=filePath, mapTokens=True)
print(mappedNumTokensFileDefault)

---

TokenizeFiles(inputPath: Path | str | list[Path | str], model: str | None = "gpt-4o", encodingName: str | None = None, encoding: tiktoken.Encoding | None = None, recursive: bool = True, quiet: bool = False, exitOnListError: bool = True, mapTokens: bool = False, excludeBinary: bool = True, includeHidden: bool = False) -> list[int] | OrderedDict[str, list[int] | OrderedDict]

Tokenizes multiple files or all files within a directory into lists of token IDs or a nested OrderedDict structure.

Parameters:

• inputPath (Path | str | list[Path | str]): The path to a file or directory, or a list of file paths to tokenize.
• model (str, optional): The name of the model to use for encoding. Default: "gpt-4o"
• encodingName (str, optional): The name of the encoding to use.
• encoding (tiktoken.Encoding, optional): An existing tiktoken.Encoding object to use for tokenization.
• recursive (bool, optional): If inputPath is a directory, whether to tokenize files in subdirectories recursively. Default: True
• quiet (bool, optional): If True, suppresses progress updates. Default: False
• exitOnListError (bool, optional): If True, stop processing the list upon encountering an error. If False, skip files that cause errors. Default: True
• mapTokens (bool, optional): If True, outputs a nested OrderedDict structure. For files, the value is an OrderedDict with keys "tokens" (the list of token IDs) and "numTokens" (the total token count). For directories, the output is wrapped with "tokens" and "numTokens" keys. If False, returns a list of token IDs for a single file, or a dictionary mapping filenames to token lists for multiple files. Default: False
• excludeBinary (bool, optional): Excludes any binary files by skipping over them. Default: True
• includeHidden (bool, optional): Skips over hidden files and directories, including subdirectories and files of a hidden directory. Default: False

Returns:

• list[int] | OrderedDict[str, list[int] | OrderedDict]:
  • If inputPath is a file:
    • If mapTokens is False, returns a list of token IDs for that file.
    • If mapTokens is True, returns an OrderedDict with the structure described in the mapTokens parameter description.
  • If inputPath is a list of files, returns a dictionary where each key is the file name and the value depends on mapTokens:
    • If mapTokens is False, the value is the list of token IDs for that file.
    • If mapTokens is True, the value is an OrderedDict with the structure described in the mapTokens parameter description, wrapped under the key "tokens", and the total token count under the key "numTokens" at the top level.
  • If inputPath is a directory:
    • If recursive is True, returns a nested OrderedDict where each key is a file or subdirectory name with corresponding token lists or sub-dictionaries. If mapTokens is True, the directory output is wrapped with "tokens" and "numTokens" keys.
    • If recursive is False, returns a dictionary with file names as keys and their token lists as values. If mapTokens is True, the directory output is wrapped with "tokens" and "numTokens" keys.

Raises:

• TypeError: If the types of inputPath, model, encodingName, encoding, or recursive are incorrect.
• ValueError: If any of the provided file paths in a list are not files, or if a provided directory path is not a directory.
• UnsupportedEncodingError: If any of the files to be tokenized have an unsupported encoding.
• RuntimeError: If the provided inputPath is neither a file, a directory, nor a list.

Example:

from PyTokenCounter import TokenizeFiles
from pathlib import Path
from collections import OrderedDict
import tiktoken

inputFiles = [
    Path("TestFile1.txt"),
    Path("TestFile2.txt"),
]
tokens = tc.TokenizeFiles(inputPath=inputFiles, model="gpt-4o")
print(tokens)

mappedTokensFiles = tc.TokenizeFiles(inputPath=inputFiles, model="gpt-4o", mapTokens=True)
print(mappedTokensFiles)

tokens = tc.TokenizeFiles(inputPath=inputFiles)
print(tokens)

# Tokenizing multiple files using the default model
mappedTokensFilesDefault = tc.TokenizeFiles(inputPath=inputFiles, mapTokens=True)
print(mappedTokensFilesDefault)

import tiktoken
encoding = tiktoken.get_encoding('p50k_base')
dirPath = Path("TestDir")
mappedDirTokensNonRecursiveEncoding = tc.TokenizeFiles(inputPath=dirPath, encoding=encoding, recursive=False, mapTokens=True)
print(mappedDirTokensNonRecursiveEncoding)

mappedDirTokensRecursiveModel = tc.TokenizeFiles(inputPath=dirPath, model="gpt-4o", recursive=True, mapTokens=True)
print(mappedDirTokensRecursiveModel)

mappedDirTokensRecursiveDefault = tc.TokenizeFiles(inputPath=dirPath, recursive=True, mapTokens=True)
print(mappedDirTokensRecursiveDefault)

# Tokenizing a directory using the default model
mappedDirTokensDefaultModel = tc.TokenizeFiles(inputPath=dirPath, recursive=True, mapTokens=True)
print(mappedDirTokensDefaultModel)

---

GetNumTokenFiles(inputPath: Path | str | list[Path | str], model: str | None = "gpt-4o", encodingName: str | None = None, encoding: tiktoken.Encoding | None = None, recursive: bool = True, quiet: bool = False, exitOnListError: bool = True, excludeBinary: bool = True, includeHidden: bool = False, mapTokens: bool = False) -> int | OrderedDict[str, int | OrderedDict]

Counts the number of tokens across multiple files or in all files within a directory, or returns a nested OrderedDict structure with counts.

Parameters:

• inputPath (Path | str | list[Path | str]): The path to a file or directory, or a list of file paths to count tokens for.
• model (str, optional): The name of the model to use for encoding. Default: "gpt-4o"
• encodingName (str, optional): The name of the encoding to use.
• encoding (tiktoken.Encoding, optional): An existing tiktoken.Encoding object to use for tokenization.
• recursive (bool, optional): If inputPath is a directory, whether to count tokens in files in subdirectories recursively. Default: True
• quiet (bool, optional): If True, suppresses progress updates. Default: False
• exitOnListError (bool, optional): If True, stop processing the list upon encountering an error. If False, skip files that cause errors. Default: True
• excludeBinary (bool, optional): Excludes any binary files by skipping over them. Default: True
• includeHidden (bool, optional): Skips over hidden files and directories, including subdirectories and files of a hidden directory. Default: False
• mapTokens (bool, optional): If True, outputs a nested OrderedDict structure. For files, the value is the token count. For directories, the output is wrapped with "tokens" and "numTokens" keys, where "tokens" contains the nested counts. If False, returns the total token count as an integer. Default: False

Returns:

• int: The total number of tokens in the specified files or directory if mapTokens is False.
• OrderedDict[str, int | OrderedDict]: An OrderedDict mirroring the directory structure with token counts if mapTokens is True.

Raises:

• TypeError: If the types of inputPath, model, encodingName, encoding, or recursive are incorrect.
• ValueError: If any of the provided file paths in a list are not files, or if a provided directory path is not a directory, or if the provided model or encoding is invalid.
• UnsupportedEncodingError: If any of the files to be tokenized have an unsupported encoding.
• RuntimeError: If the provided inputPath is neither a file, a directory, nor a list.

Example:

import PyTokenCounter as tc
from pathlib import Path
from collections import OrderedDict
import tiktoken

inputFiles = [
    Path("TestFile1.txt"),
    Path("TestFile2.txt"),
]
numTokens = tc.GetNumTokenFiles(inputPath=inputFiles, model='gpt-4o')
print(numTokens)

mappedNumTokensFiles = tc.GetNumTokenFiles(inputPath=inputFiles, model='gpt-4o', mapTokens=True)
print(mappedNumTokensFiles)


numTokens = tc.GetNumTokenFiles(inputPath=inputFiles)
print(numTokens)

# Counting tokens in multiple files using the default model
mappedNumTokensFilesDefault = tc.GetNumTokenFiles(inputPath=inputFiles, mapTokens=True)
print(mappedNumTokensFilesDefault)


import tiktoken
encoding = tiktoken.get_encoding('p50k_base')
dirPath = Path("TestDir")
mappedNumTokensDirNonRecursiveEncoding = tc.GetNumTokenFiles(inputPath=dirPath, encoding=encoding, recursive=False, mapTokens=True)
print(mappedNumTokensDirNonRecursiveEncoding)
numTokensDirRecursiveModel = tc.GetNumTokenFiles(inputPath=dirPath, model='gpt-4o', recursive=True)
print(numTokensDirRecursiveModel)

mappedNumTokensDirRecursiveModel = tc.GetNumTokenFiles(inputPath=dirPath, model='gpt-4o', recursive=True, mapTokens=True)
print(mappedNumTokensDirRecursiveModel)


# Counting tokens in a directory using the default model
mappedNumTokensDirRecursiveDefault = tc.GetNumTokenFiles(inputPath=dirPath, recursive=True, mapTokens=True)
print(mappedNumTokensDirRecursiveDefault)

---

TokenizeDir(dirPath: Path | str, model: str | None = "gpt-4o", encodingName: str | None = None, encoding: tiktoken.Encoding | None = None, recursive: bool = True, quiet: bool = False, mapTokens: bool = False, excludeBinary: bool = True, includeHidden: bool = False) -> OrderedDict[str, list[int] | OrderedDict]

Tokenizes all files within a directory into a nested OrderedDict structure or lists of token IDs.

Parameters:

• dirPath (Path | str): The path to the directory to tokenize.
• model (str, optional): The name of the model to use for encoding. Default: "gpt-4o"
• encodingName (str, optional): The name of the encoding to use.
• encoding (tiktoken.Encoding, optional): An existing tiktoken.Encoding object to use for tokenization.
• recursive (bool, optional): Whether to tokenize files in subdirectories recursively. Default: True
• quiet (bool, optional): If True, suppresses progress updates. Default: False
• mapTokens (bool, optional): If True, outputs a nested OrderedDict where each key is a file or subdirectory name. For files, the value is an OrderedDict with keys "tokens" (the list of token IDs) and "numTokens" (the total token count). For directories, the output is recursively structured. If False, returns a nested OrderedDict of token lists without the "numTokens" wrapper. Default: False
• excludeBinary (bool, optional): Excludes any binary files by skipping over them. Default: True
• includeHidden (bool, optional): Skips over hidden files and directories, including subdirectories and files of a hidden directory. Default: False

Returns:

• OrderedDict[str, list[int] | OrderedDict]: A nested OrderedDict where each key is a file or subdirectory name. If mapTokens is True, directory entries include "numTokens" and "tokens" keys.

Raises:

• TypeError: If the types of input parameters are incorrect.
• ValueError: If the provided path is not a directory or if the model or encoding is invalid.
• UnsupportedEncodingError: If the file's encoding cannot be determined.
• FileNotFoundError: If the directory does not exist.

Example:

import PyTokenCounter as tc
from pathlib import Path
from collections import OrderedDict

dirPath = Path("TestDir")
tokenizedDir = tc.TokenizeDir(dirPath=dirPath, model="gpt-4o", recursive=True)
print(tokenizedDir)

mappedTokenizedDir = tc.TokenizeDir(dirPath=dirPath, model="gpt-4o", recursive=True, mapTokens=True)
print(mappedTokenizedDir)

tokenizedDir = tc.TokenizeDir(dirPath=dirPath, recursive=True)
print(tokenizedDir)

# Tokenizing a directory using the default model
mappedTokenizedDirDefault = tc.TokenizeDir(dirPath=dirPath, recursive=True, mapTokens=True)
print(mappedTokenizedDirDefault)

tokenizedDirNonRecursiveModel = tc.TokenizeDir(dirPath=dirPath, model="gpt-4o", recursive=False)
print(tokenizedDirNonRecursiveModel)

mappedTokenizedDirNonRecursiveModel = tc.TokenizeDir(dirPath=dirPath, model="gpt-4o", recursive=False, mapTokens=True)
print(mappedTokenizedDirNonRecursiveModel)

tokenizedDirNonRecursiveDefault = tc.TokenizeDir(dirPath=dirPath, recursive=False)
print(tokenizedDirNonRecursiveDefault)


mappedTokenizedDirRecursiveModel = tc.TokenizeDir(dirPath=dirPath, model="gpt-4o", recursive=True, mapTokens=True)
print(mappedTokenizedDirRecursiveModel)

# Tokenizing a directory using the default model with token mapping
mappedTokenizedDirRecursiveDefaultModel = tc.TokenizeDir(dirPath=dirPath, recursive=True, mapTokens=True)
print(mappedTokenizedDirRecursiveDefaultModel)

---

GetNumTokenDir(dirPath: Path | str, model: str | None = "gpt-4o", encodingName: str | None = None, encoding: tiktoken.Encoding | None = None, recursive: bool = True, quiet: bool = False, excludeBinary: bool = True, includeHidden: bool = False, mapTokens: bool = False) -> int | OrderedDict[str, int | OrderedDict]

Counts the number of tokens in all files within a directory, or returns a nested OrderedDict structure with counts.

Parameters:

• dirPath (Path | str): The path to the directory to count tokens for.
• model (str, optional): The name of the model to use for encoding. Default: "gpt-4o"
• encodingName (str, optional): The name of the encoding to use.
• encoding (tiktoken.Encoding, optional): An existing tiktoken.Encoding object to use for tokenization.
• recursive (bool, optional): Whether to count tokens in subdirectories recursively. Default: True
• quiet (bool, optional): If True, suppresses progress updates. Default: False
• excludeBinary (bool, optional): Excludes any binary files by skipping over them. Default: True
• includeHidden (bool, optional): Skips over hidden files and directories, including subdirectories and files of a hidden directory. Default: False
• mapTokens (bool, optional): If True, outputs a nested OrderedDict structure mirroring the directory structure. For files, the value is the token count. For directories, the output is wrapped with "tokens" and "numTokens" keys, where "tokens" contains the nested counts. If False, returns the total token count as an integer. Default: False

Returns:

• int: The total number of tokens in the directory if mapTokens is False.
• OrderedDict[str, int | OrderedDict]: An OrderedDict mirroring the directory structure with token counts if mapTokens is True.

Raises:

• TypeError: If the types of input parameters are incorrect.
• ValueError: If the provided path is not a directory or if the model or encoding is invalid.
• UnsupportedEncodingError: If the file's encoding cannot be determined.
• FileNotFoundError: If the directory does not exist.

Example:

import PyTokenCounter as tc
from pathlib import Path
from collections import OrderedDict

dirPath = Path("TestDir")
numTokensDir = tc.GetNumTokenDir(dirPath=dirPath, model="gpt-4o", recursive=True)
print(numTokensDir)

mappedNumTokensDir = tc.GetNumTokenDir(dirPath=dirPath, model="gpt-4o", recursive=True, mapTokens=True)
print(mappedNumTokensDir)

# Counting tokens in a directory using the default model
mappedNumTokensDirDefault = tc.GetNumTokenDir(dirPath=dirPath, recursive=True, mapTokens=True)
print(mappedNumTokensDirDefault)

numTokensDirNonRecursiveModel = tc.GetNumTokenDir(dirPath=dirPath, model="gpt-4o", recursive=False)
print(numTokensDirNonRecursiveModel)

mappedNumTokensDirNonRecursiveModel = tc.GetNumTokenDir(dirPath=dirPath, model="gpt-4o", recursive=False, mapTokens=True)
print(mappedNumTokensDirNonRecursiveModel)

numTokensDirNonRecursiveDefault = tc.GetNumTokenDir(dirPath=dirPath, recursive=False)
print(numTokensDirNonRecursiveDefault)

---

Token Mapping

MapTokens(tokens: list[int] | OrderedDict[str, list[int] | OrderedDict], model: str | None = "gpt-4o", encodingName: str | None = None, encoding: tiktoken.Encoding | None = None) -> OrderedDict[str, str] | OrderedDict[str, OrderedDict[str, str] | OrderedDict]

Maps tokens to their corresponding decoded strings based on a specified encoding.

Parameters:

• tokens (list[int] | OrderedDict[str, list[int] | OrderedDict]): The tokens to be mapped. This can either be:
  • A list of integer tokens to decode.
  • An OrderedDict with string keys and values that are either:
    • A list of integer tokens.
    • Another nested OrderedDict with the same structure.
• model (str, optional): The model name to use for determining the encoding. Default: "gpt-4o"
• encodingName (str, optional): The name of the encoding to use.
• encoding (tiktoken.Encoding, optional): The encoding object to use.

Returns:

• OrderedDict[str, str] | OrderedDict[str, OrderedDict[str, str] | OrderedDict]: A mapping of decoded strings to their corresponding integer tokens. If tokens is a nested structure, the result will maintain the same nested structure with decoded mappings.

Raises:

• TypeError: If tokens is not a list of integers or an OrderedDict of strings mapped to tokens.
• ValueError: If an invalid model or encoding name is provided, or if the encoding does not match the model or encoding name.
• KeyError: If a token is not in the given encoding's vocabulary.
• RuntimeError: If an unexpected error occurs while validating the encoding.

Example:

import PyTokenCounter as tc
import tiktoken
from collections import OrderedDict

encoding = tiktoken.get_encoding("cl100k_base")
tokens = [123,456,789]
mapped = tc.MapTokens(tokens=tokens, encoding=encoding)
print(mapped)

mapped = tc.MapTokens(tokens=tokens, encoding=encoding)
print(mapped)

# Mapping tokens using the default model
mapped = tc.MapTokens(tokens=tokens)
print(mapped)

---

Ignored Files

When the functions are set to exclude binary files (default behavior), the following file extensions are ignored:

Category	Extensions
Image formats	.png, .jpg, .jpeg, .gif, .bmp, .pbm, .webp, .avif, .tiff, .tif, .ico, .svgz
Video formats	.mp4, .mkv, .mov, .avi, .wmv, .flv, .webm, .m4v, .mpeg, .mpg, .3gp, .3g2
Audio formats	.mp3, .wav, .flac, .ogg, .aac, .m4a, .wma, .aiff, .ape, .opus
Compressed archives	.zip, .rar, .7z, .tar, .gz, .bz2, .xz, .lz, .zst, .cab, .deb, .rpm, .pkg
Disk images	.iso, .dmg, .img, .vhd, .vmdk
Executables & Libraries	.exe, .msi, .bat, .dll, .so, .bin, .o, .a, .dylib
Fonts	.ttf, .otf, .woff, .woff2, .eot
Documents	.pdf, .ps, .eps
Design & Graphics	.psd, .ai, .indd, .sketch
3D & CAD files	.blend, .stl, .step, .iges, .fbx, .glb, .gltf, .3ds, .obj, .cad
Virtual Machines & Firmware	.qcow2, .vdi, .vhdx, .rom, .bin, .img
Miscellaneous binaries	.dat, .pak, .sav, .nes, .gba, .nds, .iso, .jar, .class, .wasm

Along with ignoring the extensions in the exclude list to quickly bypass known files that cannot be read, the code also catches decoding errors and skips files when excludeBinary is True. This approach ensures all unreadable files are handled efficiently, combining fast extension-based checks with robust decoding error handling.

---

Running Tests

After installing the package's dependencies, run the test suite with pytest:

pytest

The tests rely on the sample data located in the Tests/ directory.

Maintainers

• Kaden Gruizenga

Acknowledgements

• This project is based on the tiktoken library created by OpenAI.

Contributing

Contributions are welcome! Feel free to open an issue or submit a pull request.

License

This project is licensed under the GNU General Public License v3.0. See the LICENSE file for more details.