Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Cloud computing platform for running Python on GPUs and serverless infrastructure. Use when deploying AI/ML models, running GPU-accelerated workloads, serving web endpoints, scheduling batch jobs, or scaling Python code to the cloud. Use this skill whenever the user mentions Modal, serverless GPU compute, deploying ML models to the cloud, serving inference endpoints, running batch processing in the cloud, or needs to scale Python workloads beyond their local machine. Also use when the user wants
.claude/skills/mkurman-modal/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | 71% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -28% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 36% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -42% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -3% | 0% |
---|-------------| | modal setup | Authenticate with Modal | | modal run script.py | Run a script's local entrypoint | | modal serve script.py | Dev server with hot reload | | modal deploy script.py | Deploy to production | | modal volume ls <name> | List files in a volume | | modal volume put <name> <file> | Upload file to volume | | modal volume get <name> <file> | Download file from volume | | modal secret create <name> K=V | Create a secret | | modal secret list | List secrets | | modal app list | List deployed apps | | modal app stop <name> | Stop a deployed app |
Detailed documentation for each topic:
references/getting-started.md — Installation, authentication, first appreferences/functions.md — Functions, classes, lifecycle hooks, remote executionreferences/images.md — Container images, package installation, cachingreferences/gpu.md — GPU types, selection, multi-GPU, trainingreferences/volumes.md — Persistent storage, file management, v2 volumesreferences/secrets.md — Credentials, environment variables, dotenvreferences/web-endpoints.md — FastAPI, ASGI/WSGI, streaming, auth, WebSocketsreferences/scheduled-jobs.md — Cron, periodic schedules, managementreferences/scaling.md — Autoscaling, concurrency, .map(), limitsreferences/resources.md — CPU, memory, disk, timeout configurationreferences/examples.md — Common use cases and patternsreferences/api_reference.md — Key API classes and methodsRead these files when detailed information is needed beyond this overview.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 6,311 | 2,098 | -67% | 1 | 1 | 0% | 1,049 | 752 | -28% | 0 | 0 | — |
case-02 | pass→pass | 3,215 | 1,569 | -51% | 1 | 1 | 0% | 455 | 619 | +36% | 0 | 0 | — |
case-03 | pass→pass | 8,467 | 5,351 | -37% | 1 | 1 | 0% | 1,479 | 861 | -42% | 0 | 0 | — |
case-04 | pass→pass | 4,105 | 1,648 | -60% | 1 | 1 | 0% | 684 | 663 | -3% | 0 | 0 | — |
case-05 | pass→pass | 5,277 | 2,482 | -53% | 1 | 1 | 0% | 989 | 850 | -14% | 0 | 0 | — |
case-06 | pass→pass | 2,745 | 2,180 | -21% | 1 | 1 | 0% | 461 | 797 | +73% | 0 | 0 | — |
case-07 | pass→pass | 4,018 | 2,013 | -50% | 1 | 1 | 0% | 682 | 773 | +13% | 0 | 0 | — |
case-08 | pass→pass | 2,896 | 1,588 | -45% | 1 | 1 | 0% | 435 | 622 | +43% | 0 | 0 | — |
case-09 | pass→pass | 5,374 | 1,392 | -74% | 1 | 1 | 0% | 855 | 595 | -30% | 0 | 0 | — |
case-10 | pass→pass | 3,351 | 1,613 | -52% | 1 | 1 | 0% | 509 | 589 | +16% | 0 | 0 | — |
case-11 | pass→pass | 4,734 | 3,330 | -30% | 1 | 1 | 0% | 795 | 951 | +20% | 0 | 0 | — |
case-12 | pass→pass | 4,780 | 3,863 | -19% | 1 | 1 | 0% | 817 | 1,014 | +24% | 0 | 0 | — |
case-13 | pass→pass | 3,475 | 2,282 | -34% | 1 | 1 | 0% | 525 | 763 | +45% | 0 | 0 | — |
case-14 | pass→pass | 6,685 | 3,592 | -46% | 1 | 1 | 0% | 1,293 | 1,065 | -18% | 0 | 0 | — |
case-15 | pass→pass | 5,557 | 3,466 | -38% | 1 | 1 | 0% | 1,050 | 1,019 | -3% | 0 | 0 | — |
case-16 | pass→pass | 5,584 | 3,196 | -43% | 1 | 1 | 0% | 1,035 | 921 | -11% | 0 | 0 | — |
case-17 | fail→pass | 3,714 | 3,874 | +4% | 1 | 1 | 0% | 594 | 1,015 | +71% | 0 | 0 | — |
case-18 | pass→pass | 3,755 | 2,934 | -22% | 1 | 1 | 0% | 579 | 900 | +55% | 0 | 0 | — |
case-19 | pass→pass | 2,529 | 1,938 | -23% | 1 | 1 | 0% | 416 | 730 | +75% | 0 | 0 | — |
case-20 | pass→pass | 2,166 | 2,126 | -2% | 1 | 1 | 0% | 350 | 743 | +112% | 0 | 0 | — |
case-21 | pass→pass | 2,770 | 2,670 | -4% | 1 | 1 | 0% | 420 | 787 | +87% | 0 | 0 | — |
case-22 | pass→pass | 7,032 | 2,929 | -58% | 1 | 1 | 0% | 1,300 | 908 | -30% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +5 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.