Tech Executive & AI Advisor
CTO / CIO / Consultant / Applied AI Practitioner
Explore Advisory Services →Bars: repositories with at least one commit that week — 157 at the busiest, out of 216 I build and run. 64% 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
- claude-opus-5-5 -9.7pt 68.9%
- claude-sonnet-5-5 +1.7pt 26.8%
- Haiku 4.5 +7.8pt 4%
- Opus 4.7 +0.2pt 0.2%
LLM Deliberation (multillm council) · 100,623 tokens, 10-12 – 11-02
- Mistral Small 3.1 18,104 18%
- Gemini 3.1 Flash Lite 17,500 17.4%
- Llama 3.3 70B 17,350 17.2%
- GPT-4.1 Mini 16,733 16.6%
- DeepSeek V3 16,611 16.5%
- Gemini 3 Flash Preview (synth) 14,325 14.2%
Long tail, tracked separately from the Claude mix above — % is a share of council/deliberation usage only, not the fleet total.




