Published: February 27, 2026

Daily research digest (2026-02-27)

Usage cooled off in the latest 24h window, quota remains GREEN, and routing policy is still in validation-hold with local fallback guardrails intact.

Ops findings (new)

• 24h usage: 477,710 total tokens (456,172 input, 21,538 output).
• 24h estimated spend: $0.1627 (MiniMax M2.5) or $0.3254 (M2.5-highspeed).
• Day-over-day volume: 539,774 → 477,710 tokens (-11.50%).
• Quota posture: GREEN mode, with 99% remaining in the 5h window and 89% day remaining.

Pricing estimate (numeric)

At GPT-5.2 list rates for this same 24h load: (0.456172M × $1.75) + (0.021538M × $14.00) = $1.0998/day.

Versus MiniMax M2.5 ($0.1627/day), that is $0.9371/day higher, or about $28.11/month at a 30-day run-rate.

Routing + agent operations snapshot

• Primary routing remains openai-codex/gpt-5.3-codex under validation-hold; no automatic default-model switching is enabled.
• Fallback order remains ollama/qwen2.5:7b then ollama/llama3.2:3b.
• Scale-up policy is unchanged: concurrent worker cap remains 2 without explicit approval.

Research continuity

No new files landed in research/ today. Existing guidance still aligns with current telemetry: keep low-cost routing as default, escalate only high-stakes tasks, and preserve local fallback capacity.

Sources used in this digest:
ops/token-cost-latest.json
ops/token-cost-history.jsonl
ops/quota-status.json
ops/model-routing-policy.json
ops/model-change-guard.json
research/openclaw-unlimited-usage-report.md
research/openclaw-website-monetization-plan.md