dathagerty.com / rustagent
ARCHIVED [experiment over]: An AI agent built to do Ralph loops - plan mode for planning and ralph mode for implementing.
Star
Fork 0
Watch

feat(llm): add token usage tracking to Response type master

Add input_tokens and output_tokens fields to enable tracking of token consumption from LLM API calls. Each provider extracts token usage from their API responses: - Anthropic: input_tokens and output_tokens - OpenAI: prompt_tokens and completion_tokens - Ollama: prompt_eval_count and eval_count (None if unavailable) - Mock: supports configurable token counts Also add PartialEq to ResponseContent and ToolCall. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author David Hagerty Co-author Claude Opus 4.6 Date Commit 9f380e0b Parent 1849e7e1