> For longer queries or requests that rely on AI agents or reasoning, the footprint is higher. Epoch AI estimated that a long query (7,500 words of input) could consume about 2.5 Wh, and a very long one (75,000 words) could consume 40 Wh. Those are far higher than the simple text query.

One thing that's missing from almost all of these energy estimates is the impact of "agentic" queries, which were rare a year ago and increasingly common today.

I've asked questions of GPT-5.6 Pro that take 19 minutes to return an answer, as the model runs hundreds of search and site navigation tool calls (and a bit of executed Python code too) trying to get to a credible answer.

I'm confident that 19 minutes involves a lot more than 75,000 words worth of tokens.

With coding agents we can at least get an accurate token count. I'm already at 204,000 tokens of GPT-5.6 Sol today and I haven't started work yet.

With the new Ultracode modes in Claude Code and Codex it's been taken to the next level. I mean, multi agent systems have been available for a long time, but the newest models seem to be much more RL'd for them.

Both GPT-5.6 and Fable will happily run 3+ hours straight off a relatively simple goal in that mode, burn through millions of tokens in the process.

Not saying it's great bang for your buck, but just that those modes are there and being pushed by the companies.

Where did the tokens go?