# Optimizing Attention Flash

Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory...

- Repository: https://github.com/Orchestra-Research/AI-Research-SKILLs (14176 stars)
- Category: Engineering
- License: MIT
- Last updated: 2026-10-11
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add Orchestra-Research/AI-Research-SKILLs --skill optimizing-attention-flash -g -a claude-code -y`
Codex: `npx skills add Orchestra-Research/AI-Research-SKILLs --skill optimizing-attention-flash -g -a codex -y`
Cursor: `npx skills add Orchestra-Research/AI-Research-SKILLs --skill optimizing-attention-flash -g -a cursor -y`
Gemini CLI: `npx skills add Orchestra-Research/AI-Research-SKILLs --skill optimizing-attention-flash -g -a gemini-cli -y`
Claude Code plugin marketplace: `/plugin marketplace add Orchestra-Research/AI-Research-SKILLs`

## Description

Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Use when training/running transformers with long sequences (>512 tokens), encountering GPU memory issues with attention, or need faster inference. Supports PyTorch native SDPA, flash-attn library, H100 FP8, and sliding window attention.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# Flash Attention - Fast Memory-Efficient Attention

## Quick start

Flash Attention provides 2-4x speedup and 10-20x memory reduction for transformer attention through IO-aware tiling and recomputation.

**PyTorch native (easiest, PyTorch 2.2+)**:
```python
import torch
import torch.nn.functional as F

q = torch.randn(2, 8, 512, 64, device='cuda', dtype=torch.float16)  # [batch, heads, seq, dim]
k = torch.randn(2, 8, 512, 64, device='cuda', dtype=torch.float16)
v = torch.randn(2, 8, 512, 64, device='cuda', dtype=torch.float16)

# Automatically uses Flash Attention if available
out = F.scaled_dot_product_attention(q, k, v)
```

**flash-attn library (more features)**:
```bash
pip install flash-attn --no-build-isolation
```

```python
from flash_attn import flash_attn_func

# q, k, v: [batch, seqlen, nheads, headdim]
out = flash_attn_func(q, k, v, dropout_p=0.0, causal=True)
```

## Common workflows

### Workflow 1: Enable in existing PyTorch model

Copy this checklist:

```
Flash Attention Integration:
- [ ] Step 1: Check PyTorch version (≥2.2)
- [ ] Step 2: Enable Flash Attention backend
- [ ] Step 3: Verify speedup with profiling
- [ ] Step 4: Test accuracy matches baseline
```

**Step 1: Check PyTorch version**

```bash
python -c "import torch; print(torch.__version__)"
# Should be ≥2.2.0
```

If <2.2, upgrade:
```bash
pip install --upgrade torch
```

**Step 2: Enable Flash Attention backend**

Replace standard attention:
```python
# Before (standard attention)
attn_weights = torch.softmax(q @ k.transpose(-2, -1) / math.sqrt(d_k), dim=-1)
out = attn_weights @ v

# After (Flash Attention)
import torch.nn.functional as F
out = F.scaled_dot_product_attention(q, k, v, attn_mask=mask)
```

Force Flash Attention backend:
```python
with torch.backends.cuda.sdp_kernel(
    enable_flash=True,
    enable_math=False,
    enable_mem_efficient=False
):
    out = F.scaled_dot_product_attention(q, k, v)
```

**Step 3: Verify speedup with profiling**

The rest: https://github.com/Orchestra-Research/AI-Research-SKILLs/blob/HEAD/10-optimization/flash-attention/SKILL.md