Shipping with AI agents around the clock -- human hours for thinking, machine hours for doing.
Stats auto-updated by aidevops.
| Metric | Yesterday | Prior 7 Days | Prior 28 Days | Prior 365 Days |
|---|---|---|---|---|
| Screen time (Mac) | 24h | 106.7h | 331.4h | ~3413h* |
| Input activity (Mac) | unavailable | unavailable | unavailable | unavailable |
| Interactive human attention | 3.0h | 31.8h | 138.1h | 472.3h |
| Interactive AI generation | 3.4h | 98.1h | 325.7h | 1060.7h |
| Worker-classified human attention | 0.0h | 0.1h | 3.3h | 128.8h |
| Worker/headless AI generation | 1.3h | 18.8h | 92.3h | 317.7h |
| Additive observed work | 7.7h | 148.7h | 557.0h | 1,936.6h |
| Interactive sessions | 8 | 1,116 | 1,279 | 1,994 |
| Worker sessions | 185 | 1,023 | 4,136 | 9,644 |
Screen time from screen-time-history:daily-observations; collection status: ok. *365-day estimate uses observed calendar coverage.
Input activity from macos-input-activity (time from each keyboard, mouse or trackpad input until 5 idle minutes); collection status: unavailable (no-input-activity-samples).
Periods are completed local calendar days ending at midnight; today is excluded.
Human attention is unioned wall-clock time, so overlapping sessions are not double-counted. AI generation is additive machine work across sessions; it is not wall-clock concurrency.
AI session 365-day totals cover 220 days of local assistant session history (not extrapolated).
| Model | Requests | Input | Output | Cache read | Cache write | Cache Hit-Rate % | Session Count | Session Hours |
|---|---|---|---|---|---|---|---|---|
| claude-opus-5-5 | 33,076 | 70K | 13.9M | 5,224.0M | 258.8M | 95.3% | 489 | 187.4h |
| gpt-6.1-sol | 14,937 | 97.4M | 2.8M | 1,393.5M | 0 | 93.5% | 1,196 | 62.9h |
| gpt-6.1-sol-fast | 13,063 | 44.3M | 3.8M | 1,910.7M | 0 | 97.7% | 61 | 106.9h |
| gpt-6-luna | 7,136 | 157.6M | 2.3M | 490.3M | 0 | 75.7% | 2,351 | 26.1h |
| gpt-6-astra | 5,215 | 17.8M | 1.5M | 796.3M | 0 | 97.8% | 114 | 43.3h |
| claude-sonnet-5-5 | 1,020 | 18K | 1.0M | 42.8M | 21.9M | 66.1% | 556 | 11.5h |
| studio-glm-5.3-flash | 490 | 1.9M | 105K | 34.2M | 0 | 94.5% | 17 | 1.0h |
| claude-haiku-5-5 | 294 | 1K | 806K | 5.6M | 18.5M | 23.5% | 221 | 3.0h |
| claude-fable-5-2 | 65 | 140 | 76K | 9.7M | 983K | 90.8% | 2 | 0.4h |
| gpt-5.6-luna | 43 | 947K | 7K | 5.6M | 0 | 85.6% | 5 | 0.1h |
| gpt-6-sol | 30 | 214K | 2K | 45K | 0 | 17.4% | 23 | 0.1h |
| gpt-6-sol-fast | 9 | 206K | 4K | 1.6M | 0 | 88.8% | 1 | 0.0h |
| claude-sonnet-4-6 | 6 | 18 | 9K | 29K | 66K | 30.9% | 6 | 0.1h |
| Total | 75,384 | 320.7M | 26.5M | 9,914.9M | 300.3M | 94.1% | 4,864 | 443.0h |
10,562.5M total tokens processed. 94.1% cache hit rate.
| Model | Requests | Input | Output | Cache read | Cache write | Cache Hit-Rate % | Session Count | Session Hours |
|---|---|---|---|---|---|---|---|---|
| gpt-5.6-sol | 54,086 | 245.8M | 12.5M | 7,425.8M | 0 | 96.8% | 519 | 440.4h |
| gpt-6.1-sol | 38,225 | 219.7M | 8.0M | 4,018.5M | 0 | 94.8% | 2,136 | 215.8h |
| claude-opus-5-5 | 33,076 | 70K | 13.9M | 5,224.0M | 258.8M | 95.3% | 489 | 187.4h |
| gpt-6.1-sol-fast | 24,102 | 83.1M | 6.9M | 3,567.3M | 0 | 97.7% | 96 | 194.5h |
| gpt-5.6-terra | 16,886 | 284.7M | 3.0M | 1,283.6M | 0 | 81.8% | 1,615 | 58.0h |
| gpt-6-luna | 13,339 | 301.1M | 3.7M | 958.3M | 0 | 76.1% | 4,112 | 48.1h |
| gpt-6-astra | 12,813 | 62.3M | 3.8M | 2,130.0M | 0 | 97.2% | 397 | 128.9h |
| gpt-5.5 | 7,076 | 45.3M | 2.0M | 717.6M | 0 | 94.1% | 127 | 33.4h |
| gpt-6-sol | 2,978 | 26.8M | 437K | 233.2M | 0 | 89.7% | 324 | 14.7h |
| gpt-5.6-luna | 2,180 | 46.4M | 530K | 186.8M | 0 | 80.1% | 471 | 31.8h |
| claude-sonnet-5-5 | 1,020 | 18K | 1.0M | 42.8M | 21.9M | 66.1% | 556 | 11.5h |
| studio-glm-5.3-flash | 490 | 1.9M | 105K | 34.2M | 0 | 94.5% | 17 | 1.0h |
| claude-haiku-5-5 | 294 | 1K | 806K | 5.6M | 18.5M | 23.5% | 221 | 3.0h |
| gemma4:26b-mlx | 82 | 4.1M | 26K | 0 | 0 | 0.0% | 5 | 1.1h |
| claude-fable-5-2 | 65 | 140 | 76K | 9.7M | 983K | 90.8% | 2 | 0.4h |
| muse-spark-1.3-contributor-free | 63 | 906K | 23K | 7.6M | 0 | 89.4% | 3 | 0.2h |
| big-pickle | 44 | 257K | 15K | 3.0M | 0 | 92.3% | 3 | 0.2h |
| qwen3.8-flash | 22 | 87K | 2K | 1.3M | 0 | 93.9% | 1 | 0.1h |
| gemma4-agent | 17 | 314K | 4K | 0 | 0 | 0.0% | 2 | 0.1h |
| gpt-6-sol-fast | 9 | 206K | 4K | 1.6M | 0 | 88.8% | 1 | 0.0h |
| Qwen3.8-27B-oQ6e-mtp:qwen38-q6-stable | 7 | 63K | 1K | 249K | 0 | 79.7% | 1 | 0.3h |
| claude-sonnet-4-6 | 6 | 18 | 9K | 29K | 66K | 30.9% | 6 | 0.1h |
| gemma4-agent-mlx:latest | 6 | 170K | 3K | 0 | 0 | 0.0% | 2 | 0.1h |
| mtplx-qwen38-27b-optimized-quality | 1 | 0 | 0 | 0 | 0 | 0.0% | 1 | 0.0h |
| nemotron-3.5-lightning-30b-a3b | 1 | 49K | 107 | 0 | 0 | 0.0% | 1 | 0.0h |
| qwen/qwen3.8-flash | 1 | 0 | 0 | 0 | 0 | 0.0% | 1 | 0.0h |
| Total | 206,889 | 1,323.8M | 57.2M | 25,852.2M | 300.3M | 94.1% | 10,728 | 1,371.3h |
27,533.6M total tokens processed. 94.1% cache hit rate.
- aidevops -- Vibe-Coding is easy. DevOps is hard. OpenCode & Git token-efficient AI agent automation for your app, business, and personal development. Opinionated tools, services, CLI & API stack for speed, security, and 24/7 results. Open-source first. SOTA everything. Try on your repos for money-making magic.
- opendesk -- The open-source support inbox + live chat and inbox collaboration for teams
- playwriter -- Chrome extension & CLI to let agents control your browser. Runs Playwright snippets in a stateful sandbox. Available as CLI or MCP
Stats auto-updated 2026-11-14 21:00 UTC by aidevops pulse.
Verify on commit-history.com · Chart data
Includes commits, issues, pull requests, reviews, repositories, and restricted contributions. Refreshed daily through the prior UTC day; commit-history.com may use a different refresh cutoff. GitHub controls link navigation—Ctrl/Cmd-click opens verification in a new tab.