Dao-AILab/flash-attention

FlashAttention

Usage

FlashAttention-3 beta release

cd hopper
python setup.py install

export PYTHONPATH=$PWD
pytest -q -s test_flash_attn.py

from flash_attn_3 import flash_attn_interface
flash_attn_interface.flash_attn_func()

[project]
dependencies = [
    "flash-attn-3"
]

[tool.uv]
no-build-isolation = true

[tool.uv.sources]
flash-attn-3 = { git = "https://github.com/Dao-AILab/flash-attention", subdirectory = "hopper" }

FlashAttention-4 (CuTeDSL)

pip install flash-attn-4

pip install "flash-attn-4[cu13]"

from flash_attn.cute import flash_attn_func

out = flash_attn_func(q, k, v, causal=True)

Installation and features

pip install flash-attn --no-build-isolation

python setup.py install

MAX_JOBS=4 pip install flash-attn --no-build-isolation

NVIDIA CUDA Support

AMD ROCm Support

Composable Kernel Backend

Triton Backend

cd flash-attention
FLASH_ATTENTION_TRITON_AMD_ENABLE="TRUE" pip install --no-build-isolation .

cd flash-attention
FLASH_ATTENTION_TRITON_AMD_ENABLE="TRUE" AITER_USE_SYSTEM_TRITON=1 pip install --no-build-isolation .

cd flash-attention
FLASH_ATTENTION_TRITON_AMD_ENABLE="TRUE" FLASH_ATTENTION_USE_SYSTEM_AITER="TRUE" pip install --no-build-isolation .

cd flash-attention
cd third_party/aiter && git fetch origin && git checkout <commit-sha> && cd ../..
FLASH_ATTENTION_TRITON_AMD_ENABLE="TRUE" pip install --no-build-isolation .

FLASH_ATTENTION_TRITON_AMD_ENABLE="TRUE" pytest tests/test_flash_attn_triton_amd.py

FLASH_ATTENTION_FWD_TRITON_AMD_CONFIG_JSON='{"BLOCK_M":128,"BLOCK_N":64,"waves_per_eu":1,"PRE_LOAD_V":false,"num_stages":1,"num_warps":8}'

FROM rocm/pytorch:latest

WORKDIR /workspace

# build flash attention with triton backend
RUN git clone https://github.com/Dao-AILab/flash-attention &&\ 
    cd flash-attention &&\
    FLASH_ATTENTION_TRITON_AMD_ENABLE="TRUE" pip install --no-build-isolation .

# set working dir
WORKDIR /workspace/flash-attention

# set env variable to use triton backend
ENV FLASH_ATTENTION_TRITON_AMD_ENABLE="TRUE"

docker build -t flash-attn-triton .
docker run -it --network=host --user root --group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined --ipc=host --shm-size 16G --device=/dev/kfd --device=/dev/dri flash-attn-triton

How to use FlashAttention

from flash_attn import flash_attn_qkvpacked_func, flash_attn_func

flash_attn_qkvpacked_func(qkv, dropout_p=0.0, softmax_scale=None, causal=False,
                          window_size=(-1, -1), alibi_slopes=None, deterministic=False):
"""dropout_p should be set to 0.0 during evaluation
If Q, K, V are already stacked into 1 tensor, this function will be faster than
calling flash_attn_func on Q, K, V since the backward pass avoids explicit concatenation
of the gradients of Q, K, V.
If window_size != (-1, -1), implements sliding window local attention. Query at position i
will only attend to keys between [i - window_size[0], i + window_size[1]] inclusive.
Arguments:
    qkv: (batch_size, seqlen, 3, nheads, headdim)
    dropout_p: float. Dropout probability.
    softmax_scale: float. The scaling of QK^T before applying softmax.
        Default to 1 / sqrt(headdim).
    causal: bool. Whether to apply causal attention mask (e.g., for auto-regressive modeling).
    window_size: (left, right). If not (-1, -1), implements sliding window local attention.
    alibi_slopes: (nheads,) or (batch_size, nheads), fp32. A bias of (-alibi_slope * |i - j|) is added to
        the attention score of query i and key j.
    deterministic: bool. Whether to use the deterministic implementation of the backward pass,
        which is slightly slower and uses more memory. The forward pass is always deterministic.
Return:
    out: (batch_size, seqlen, nheads, headdim).
"""

flash_attn_func(q, k, v, dropout_p=0.0, softmax_scale=None, causal=False,
                window_size=(-1, -1), alibi_slopes=None, deterministic=False):
"""dropout_p should be set to 0.0 during evaluation
Supports multi-query and grouped-query attention (MQA/GQA) by passing in KV with fewer heads
than Q. Note that the number of heads in Q must be divisible by the number of heads in KV.
For example, if Q has 6 heads and K, V have 2 heads, head 0, 1, 2 of Q will attention to head
0 of K, V, and head 3, 4, 5 of Q will attention to head 1 of K, V.
If window_size != (-1, -1), implements sliding window local attention. Query at position i
will only attend to keys between
[i + seqlen_k - seqlen_q - window_size[0], i + seqlen_k - seqlen_q + window_size[1]] inclusive.

Arguments:
    q: (batch_size, seqlen, nheads, headdim)
    k: (batch_size, seqlen, nheads_k, headdim)
    v: (batch_size, seqlen, nheads_k, headdim)
    dropout_p: float. Dropout probability.
    softmax_scale: float. The scaling of QK^T before applying softmax.
        Default to 1 / sqrt(headdim).
    causal: bool. Whether to apply causal attention mask (e.g., for auto-regressive modeling).
    window_size: (left, right). If not (-1, -1), implements sliding window local attention.
    alibi_slopes: (nheads,) or (batch_size, nheads), fp32. A bias of
        (-alibi_slope * |i + seqlen_k - seqlen_q - j|)
        is added to the attention score of query i and key j.
    deterministic: bool. Whether to use the deterministic implementation of the backward pass,
        which is slightly slower and uses more memory. The forward pass is always deterministic.
Return:
    out: (batch_size, seqlen, nheads, headdim).
"""

def flash_attn_with_kvcache(
    q,
    k_cache,
    v_cache,
    k=None,
    v=None,
    rotary_cos=None,
    rotary_sin=None,
    cache_seqlens: Optional[Union[(int, torch.Tensor)]] = None,
    cache_batch_idx: Optional[torch.Tensor] = None,
    block_table: Optional[torch.Tensor] = None,
    softmax_scale=None,
    causal=False,
    window_size=(-1, -1),  # -1 means infinite context window
    rotary_interleaved=True,
    alibi_slopes=None,
):
    """
    If k and v are not None, k_cache and v_cache will be updated *inplace* with the new values from
    k and v. This is useful for incremental decoding: you can pass in the cached keys/values from
    the previous step, and update them with the new keys/values from the current step, and do
    attention with the updated cache, all in 1 kernel.

    If you pass in k / v, you must make sure that the cache is large enough to hold the new values.
    For example, the KV cache could be pre-allocated with the max sequence length, and you can use
    cache_seqlens to keep track of the current sequence lengths of each sequence in the batch.

    Also apply rotary embedding if rotary_cos and rotary_sin are passed in. The key @k will be
    rotated by rotary_cos and rotary_sin at indices cache_seqlens, cache_seqlens + 1, etc.
    If causal or local (i.e., window_size != (-1, -1)), the query @q will be rotated by rotary_cos
    and rotary_sin at indices cache_seqlens, cache_seqlens + 1, etc.
    If not causal and not local, the query @q will be rotated by rotary_cos and rotary_sin at
    indices cache_seqlens only (i.e. we consider all tokens in @q to be at position cache_seqlens).

    See tests/test_flash_attn.py::test_flash_attn_kvcache for examples of how to use this function.

    Supports multi-query and grouped-query attention (MQA/GQA) by passing in KV with fewer heads
    than Q. Note that the number of heads in Q must be divisible by the number of heads in KV.
    For example, if Q has 6 heads and K, V have 2 heads, head 0, 1, 2 of Q will attention to head
    0 of K, V, and head 3, 4, 5 of Q will attention to head 1 of K, V.

    If causal=True, the causal mask is aligned to the bottom right corner of the attention matrix.
    For example, if seqlen_q = 2 and seqlen_k = 5, the causal mask (1 = keep, 0 = masked out) is:
        1 1 1 1 0
        1 1 1 1 1
    If seqlen_q = 5 and seqlen_k = 2, the causal mask is:
        0 0
        0 0
        0 0
        1 0
        1 1
    If the row of the mask is all zero, the output will be zero.

    If window_size != (-1, -1), implements sliding window local attention. Query at position i
    will only attend to keys between
    [i + seqlen_k - seqlen_q - window_size[0], i + seqlen_k - seqlen_q + window_size[1]] inclusive.

    Note: Does not support backward pass.

    Arguments:
        q: (batch_size, seqlen, nheads, headdim)
        k_cache: (batch_size_cache, seqlen_cache, nheads_k, headdim) if there's no block_table,
            or (num_blocks, page_block_size, nheads_k, headdim) if there's a block_table (i.e. paged KV cache)
            page_block_size must be a multiple of 256.
        v_cache: (batch_size_cache, seqlen_cache, nheads_k, headdim) if there's no block_table,
            or (num_blocks, page_block_size, nheads_k, headdim) if there's a block_table (i.e. paged KV cache)
        k [optional]: (batch_size, seqlen_new, nheads_k, headdim). If not None, we concatenate
            k with k_cache, starting at the indices specified by cache_seqlens.
        v [optional]: (batch_size, seqlen_new, nheads_k, headdim). Similar to k.
        rotary_cos [optional]: (seqlen_ro, rotary_dim / 2). If not None, we apply rotary embedding
            to k and q. Only applicable if k and v are passed in. rotary_dim must be divisible by 16.
        rotary_sin [optional]: (seqlen_ro, rotary_dim / 2). Similar to rotary_cos.
        cache_seqlens: int, or (batch_size,), dtype torch.int32. The sequence lengths of the
            KV cache.
        block_table [optional]: (batch_size, max_num_blocks_per_seq), dtype torch.int32.
        cache_batch_idx: (batch_size,), dtype torch.int32. The indices used to index into the KV cache.
            If None, we assume that the batch indices are [0, 1, 2, ..., batch_size - 1].
            If the indices are not distinct, and k and v are provided, the values updated in the cache
                 might come from any of the duplicate indices.
        softmax_scale: float. The scaling of QK^T before applying softmax.
            Default to 1 / sqrt(headdim).
        causal: bool. Whether to apply causal attention mask (e.g., for auto-regressive modeling).
        window_size: (left, right). If not (-1, -1), implements sliding window local attention.
        rotary_interleaved: bool. Only applicable if rotary_cos and rotary_sin are passed in.
            If True, rotary embedding will combine dimensions 0 & 1, 2 & 3, etc. If False,
            rotary embedding will combine dimensions 0 & rotary_dim / 2, 1 & rotary_dim / 2 + 1
            (i.e. GPT-NeoX style).
        alibi_slopes: (nheads,) or (batch_size, nheads), fp32. A bias of
            (-alibi_slope * |i + seqlen_k - seqlen_q - j|)
            is added to the attention score of query i and key j.

    Return:
        out: (batch_size, seqlen, nheads, headdim).
    """

Using with 🤗 Kernels

# pip install kernels

from kernels import get_kernel

# FA2
fa_module = get_kernel("kernels-community/flash-attn2", version=1)
flash_attn_func = fa_module.flash_attn_func

# FA3
fa3_module = get_kernel("kernels-community/flash-attn3", version=1)
flash_attn_func = fa3_module.flash_attn_func

Changelog

2.0: Complete rewrite, 2x faster

flash_attn_qkvpacked_func(qkv, dropout_p=0.0, softmax_scale=None, causal=False)

flash_attn_func(q, k, v, dropout_p=0.0, softmax_scale=None, causal=False)

2.1: Change behavior of causal flag

2.2: Optimize for inference

2.3: Local (i.e., sliding window) attention

2.4: ALiBi (attention with linear bias), deterministic backward pass.

2.5: Paged KV cache.

2.6: Softcapping.

2.7: Compatibility with torch compile

Performance

A100

Speedup

Memory

H100

Full model code and training script

Triton implementation of FlashAttention

Tests

pytest -q -s tests/test_flash_attn.py

When you encounter issues

Tests

pytest tests/test_flash_attn_ck.py

Citation

@inproceedings{dao2022flashattention,
  title={Flash{A}ttention: Fast and Memory-Efficient Exact Attention with {IO}-Awareness},
  author={Dao, Tri and Fu, Daniel Y. and Ermon, Stefano and Rudra, Atri and R{\'e}, Christopher},
  booktitle={Advances in Neural Information Processing Systems (NeurIPS)},
  year={2022}
}
@inproceedings{dao2023flashattention2,
  title={Flash{A}ttention-2: Faster Attention with Better Parallelism and Work Partitioning},
  author={Dao, Tri},
  booktitle={International Conference on Learning Representations (ICLR)},
  year={2024}
}