Tech Executive & AI Advisor
CTO / CIO / Consultant / Applied AI Practitioner
Explore Advisory Services →Bars: repositories with at least one commit that week — 134 at the busiest, out of 161 I build and run. 67% of commits paired with a coding agent. · commits
Bars are input tokens per week (incl. cache-creation/cache-read), left axis; the line is output tokens per week, right axis, independently scaled — not dollar cost. · output tokens
- Opus 5 -57.9pt 42.4%
- Sonnet 5 -2.6pt 33.2%
- claude-opus-5-5 +53.6pt 14.6%
- Haiku 4.5 -1.4pt 6.9%
- claude-sonnet-5-5 new 1.4%
- Opus 4.8 -0.1pt 0.9%
- Opus 4.7 +0.2pt 0.4%
- claude-fable-5-1 0pt 0.1%
LLM Deliberation (multillm council) · 65,109 tokens, 09-07 – 09-28
- GPT-4.1 Mini 12,429 19.1%
- Gemini 3.1 Flash Lite 11,946 18.3%
- Gemini 3 Flash Preview (synth) 11,056 17%
- DeepSeek V3 11,049 17%
- Llama 3.3 70B 9,647 14.8%
- Mistral Small 3.1 8,982 13.8%
Long tail, tracked separately from the Claude mix above — % is a share of council/deliberation usage only, not the fleet total.