NAME
App::Greple::xlate::gpt5 - GPT-5 translation engine for greple xlate module
SYNOPSIS
greple -Mxlate --xlate-engine=gpt5 --xlate=ja file.txtDESCRIPTION
This module provides GPT-5 translation support for the App::Greple::xlate module. GPT-5 is OpenAI's latest language model released in 2025, offering enhanced reasoning capabilities and improved translation quality.
GPT-5 API SPECIFICATIONS
Model Variants
GPT-5 is available in three sizes:
gpt-5 - Full model with maximum capabilities
gpt-5-mini - Smaller, faster variant
gpt-5-nano - Minimal variant for lightweight tasks
Token Limits
Input limit: 272,000 tokens
Output limit: 128,000 tokens (including reasoning tokens)
Total context window: 400,000 tokens (272,000 input + 128,000 output)
Input/Output Support
Input: Text and images
Output: Text only
New API Parameters
GPT-5 introduces several new parameters for fine-grained control:
reasoning_effort
Controls the model's thinking time and reasoning depth:
minimal - Minimal reasoning for fast responses, suitable for deterministic tasks
low - Low reasoning effort, prioritizes speed
medium - Balanced reasoning (default)
high - Maximum reasoning effort, prioritizes quality
verbosity
Controls the length and detail of responses:
low - Minimal, terse responses
medium - Balanced detail level
high - Comprehensive, verbose responses
max_completion_tokens
Specifies the maximum number of completion tokens in the response. Unlike the legacy max_tokens parameter, this specifically controls output tokens and is the recommended approach for GPT-5.
Enhanced Features
Reduced Hallucinations
GPT-5 is significantly less likely to hallucinate compared to previous models:
45% fewer factual errors compared to GPT-4o
80% fewer factual errors compared to OpenAI o3 when reasoning is enabled
Custom Tools Support
GPT-5 supports custom tools that can receive plaintext payloads instead of JSON, enabling more flexible integration with external systems.
Context-Free Grammar (CFG)
Allows strict output constraints to match predefined syntax rules, useful for ensuring valid format generation.
Pricing (2025)
GPT-5 Series
gpt-5: $1.25/1M input tokens, $10/1M output tokens
gpt-5-mini: $0.25/1M input tokens, $2/1M output tokens
gpt-5-nano: $0.05/1M input tokens, $0.40/1M output tokens
Cached input: $0.125/1M tokens (90% discount on cached input tokens)
GPT-4 Series (for comparison)
gpt-4.1: $2.00/1M input tokens, $8.00/1M output tokens (1M token context)
gpt-4.1-mini: $0.40/1M input tokens, $1.60/1M output tokens (1M token context)
gpt-4.1-nano: $0.10/1M input tokens, $0.40/1M output tokens (1M token context)
gpt-4o: $3.00/1M input tokens, $10.00/1M output tokens
gpt-4o-mini: $0.15/1M input tokens, $0.60/1M output tokens
gpt-4o with audio: $5/1M input tokens, $20/1M output tokens (text), $100/1M input tokens, $200/1M output tokens (audio)
Note: GPT-4.1 models feature 1,000,000 token context window and prompt caching (25% input cost for cached prefixes).
GPT-4.1 is approximately 26% cheaper than GPT-4o for median queries.
GPT-4o represents an 83% price drop for output tokens and 90% drop for input tokens compared to original GPT-4.
CONFIGURATION
This module uses the following default parameters:
engine: gpt-5
temperature: 1 (fixed for GPT-5)
max_length: 3000 characters per batch
reasoning_effort: minimal (for translation tasks)
max_completion_tokens: 4000
ENVIRONMENT VARIABLES
OPENAI_API_KEY - Required OpenAI API key
XLATE_DEBUG - Enable debug output
XLATE_MAXLEN - Override maximum batch length
DEPENDENCIES
App::cdif::Command - For gpty command execution
JSON - For JSON array processing
SEE ALSO
OpenAI GPT-5 Documentation: https://openai.com/gpt-5/
AUTHOR
Kazumasa Utashiro
LICENSE
Copyright 2024-2025 Kazumasa Utashiro.
This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself.