# Command-Line Help for `wordchipper-cli`
This document contains the help content for the `wordchipper-cli` command-line program.
**Command Overview:**
* [`wordchipper-cli`↴](#wordchipper-cli)
* [`wordchipper-cli cat`↴](#wordchipper-cli-cat)
* [`wordchipper-cli lexers`↴](#wordchipper-cli-lexers)
* [`wordchipper-cli lexers list`↴](#wordchipper-cli-lexers-list)
* [`wordchipper-cli lexers stress`↴](#wordchipper-cli-lexers-stress)
* [`wordchipper-cli models`↴](#wordchipper-cli-models)
* [`wordchipper-cli models list`↴](#wordchipper-cli-models-list)
* [`wordchipper-cli train`↴](#wordchipper-cli-train)
* [`wordchipper-cli doc`↴](#wordchipper-cli-doc)
## `wordchipper-cli`
Text tokenizer multi-tool
**Usage:** `wordchipper-cli <COMMAND>`
###### **Subcommands:**
* `cat` — Act as a streaming tokenizer
* `lexers` — Lexers sub-menu
* `models` — Models sub-menu
* `train` — Train a new model
* `doc` — Generate markdown documentation
## `wordchipper-cli cat`
Act as a streaming tokenizer
**Usage:** `wordchipper-cli cat [OPTIONS] <--model <MODEL>> <--encode|--decode>`
###### **Options:**
* `--model <MODEL>` — Model to use for encoding
Default value: `openai:r50k_base`
* `--encode` — Encode from text to tokens
* `--decode` — Decode from tokens to text
* `--input <INPUT>` — Optional input file; "-" may be used to indicate stdin
* `--output <OUTPUT>` — Optional output file; "-" may be used to indicate stdout
* `--cache-dir <CACHE_DIR>` — Cache directory
## `wordchipper-cli lexers`
Lexers sub-menu
**Usage:** `wordchipper-cli lexers <COMMAND>`
###### **Subcommands:**
* `list` — List available lexers
* `stress` — Stress test a regex accelerator
## `wordchipper-cli lexers list`
List available lexers
**Usage:** `wordchipper-cli lexers list [OPTIONS]`
**Command Alias:** `ls`
###### **Options:**
* `-p`, `--patterns` — Display the patterns
## `wordchipper-cli lexers stress`
Stress test a regex accelerator
**Usage:**
`wordchipper-cli lexers stress [OPTIONS] --input-format <INPUT_FORMAT> <--lexer-model <LEXER_MODEL>|--pattern <PATTERN>> [FILES]...`
###### **Arguments:**
* `<FILES>` — Input files
###### **Options:**
* `--input-format <INPUT_FORMAT>` — The input shard file format
Possible values:
- `text`:
Simple text files
- `parquet`:
Parquet files
* `--input-batch-size <INPUT_BATCH_SIZE>` — The input batch size
Default value: `100`
* `-q`, `--quiet` — Silence log messages
* `-v`, `--verbose` — Turn debugging information on (-v, -vv, -vvv)
* `-t`, `--ts` — Enable timestamped logging
* `--lexer-model <LEXER_MODEL>` — Model name for selection
* `--pattern <PATTERN>` — Pattern for selection
* `--pre-context <PRE_CONTEXT>` — Span context before error
Default value: `8`
* `--post-context <POST_CONTEXT>` — Span context after error
Default value: `8`
## `wordchipper-cli models`
Models sub-menu
**Usage:** `wordchipper-cli models <COMMAND>`
###### **Subcommands:**
* `list` — List available models
## `wordchipper-cli models list`
List available models
**Usage:** `wordchipper-cli models list`
**Command Alias:** `ls`
## `wordchipper-cli train`
Train a new model
**Usage:**
`wordchipper-cli train [OPTIONS] --input-format <INPUT_FORMAT> <--lexer-model <LEXER_MODEL>|--pattern <PATTERN>> [FILES]...`
###### **Arguments:**
* `<FILES>` — Input files
###### **Options:**
* `--input-format <INPUT_FORMAT>` — The input shard file format
Possible values:
- `text`:
Simple text files
- `parquet`:
Parquet files
* `--input-batch-size <INPUT_BATCH_SIZE>` — The input batch size
Default value: `100`
* `-q`, `--quiet` — Silence log messages
* `-v`, `--verbose` — Turn debugging information on (-v, -vv, -vvv)
* `-t`, `--ts` — Enable timestamped logging
* `--vocab-size <VOCAB_SIZE>` — Max vocab size
Default value: `50281`
* `--lexer-model <LEXER_MODEL>` — Model name for selection
* `--pattern <PATTERN>` — Pattern for selection
* `--output <OUTPUT>` — Optional output file; "-" may be used to indicate stdout
## `wordchipper-cli doc`
Generate markdown documentation
**Usage:** `wordchipper-cli doc`
<hr/>
<small><i>
This document was generated automatically by
<a href="https://crates.io/crates/clap-markdown"><code>clap-markdown</code></a>.
</i></small>