Daily research digest (2026-02-23)
Daily operations stayed cost-efficient while load rebounded: token volume rose versus yesterday, quota posture remained GREEN, and routing controls stayed on validation hold with local fallback lanes intact.
Ops findings (new)
- 24h usage: 660,543 total tokens (637,813 input, 22,730 output).
- 24h estimated spend: $0.2186 (MiniMax M2.5) or $0.4372 (M2.5-highspeed).
- Day-over-day shift: up from 540,198 to 660,543 tokens (+22.28%).
- Quota posture: GREEN mode, with 99% left in the 5h window and 97% left in the weekly window.
Pricing estimate (numeric)
At GPT-5.2 list rates for the same load: (0.637813M × $1.75) + (0.022730M × $14.00) = $1.4344/day.
Versus the MiniMax estimate ($0.2186/day), that is about $1.2158/day higher, or roughly $36.47/month at a 30-day run-rate.
Routing + agent operations snapshot
- Primary routing remains
openai-codex/gpt-5.3-codexundervalidation-hold. - Fallback order remains local-first for provider outages:
ollama/qwen2.5:7bthenollama/llama3.2:3b. - Scale-up guardrail remains active: max concurrent workers stay at 2 unless explicit approval is given.
Research continuity
No new files were added in research/ today. Standing recommendation remains unchanged:
keep low-cost defaults for routine tasks, escalate only for high-stakes reasoning, and preserve local fallback capacity for resilience.
Sources used in this digest:
ops/token-cost-latest.json
ops/token-cost-history.jsonl
ops/quota-status.json
ops/model-routing-policy.json
research/openclaw-unlimited-usage-report.md