# Ivx Cf Memory Management

> Optimize memory usage in Python applications and GPU workloads. Use when debugging memory leaks, reducing memory footprint, or designing memory-efficient pipelines.

- Skill: `intelli-verse-x/ivx-cf-memory-management` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add intelli-verse-x/ivx-cf-memory-management`
- Raw SKILL.md: https://api.skillmd.com/api/skills/intelli-verse-x/ivx-cf-memory-management/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: intelli-verse-x (https://skillmd.com/u/intelli-verse-x)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/intelli-verse-x/ivx-cf-memory-management

---

# Memory Management Skill

## Purpose

Keep memory usage predictable and within limits.

## Python Techniques

1. **Generators**: Stream data instead of loading all
2. **Context Managers**: Ensure cleanup (`with` statements)
3. **Weak References**: Avoid reference cycles
4. **Object Pools**: Reuse expensive objects
5. **Profiling**: `tracemalloc`, `memory_profiler`

## GPU Techniques

1. **Gradient Checkpointing**: Trade compute for memory
2. **Mixed Precision**: FP16/FP8 reduces memory 2x
3. **Batch Size**: Tune for memory limit
4. **Clear Cache**: `torch.cuda.empty_cache()` between tasks
5. **Model Sharding**: Split across GPUs

## CF-Specific

- Don't store large blobs in `task["result"]`
- Use `output_path/` for large assets
- Stream video/audio when possible
- Clear GPU memory between pipeline kinds
- Monitor worker memory in K8s

## Targets

| Component | Limit | Alert |
|-----------|-------|-------|
| Worker pod | 80% of request | At 70% |
| GPU VRAM | 90% of capacity | At 85% |
| Redis | 80% of maxmemory | At 75% |

## Quality Gates

- [ ] No memory leaks (stable over 24h)
- [ ] Large objects streamed, not loaded
- [ ] GPU memory cleared between unrelated tasks
- [ ] Memory usage within K8s limits