llm-echo 0.5 Updates Simulated Token Usage Counts

What happened

Simon Willison’s llm-echo 0.5 debug plugin updates its simulated token usage counts to account for tool calls, system prompts, and reasoning tokens. The plugin provides a “model” that repeats back prompts without involving an LLM. In the example, `uvx --with llm-echo llm -m echo 'hi there' -o thinking 1 --usage` reports 2 input and 24 output tokens.

The story so far

Simon Willison updated the llm-echo debugging plugin in version 0.5 so its simulated token usage counts now account for tool calls, system prompts, and reasoning tokens. The plugin provides a “model” that repeats prompts back without involving an LLM. In the example command, `uvx --with llm-echo llm -m echo 'hi there' -o thinking 1 --usage`, the plugin reports 2 input tokens and 24 output tokens. These are simulated counts, not usage from an actual LLM.

Keep reading

Trending now

1. 1DeepSeek Releases V4.1-Flash-1017, Now Available on Ollama Cloud With US Hosting
2. 2Anthropic Pledges 5% of Its Shares to New Public Benefit Fund
3. 3Meta Removes Business Terms Language That Raised Website Scraping and AI Training Concerns
4. 4Anthropic Says Claude Reconstructed Five NIST Elliptic-Curve Seeds Using a Timestamp-Initialized Generator