Upload 4 files

.gitattributes

@@ -57,3 +57,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+runscripts/attnfiles/flash_attn-2.8.3+cu128torch2.8.0cxx11abiFALSE-cp312-cp312-win_amd64.whl filter=lfs diff=lfs merge=lfs -text
+runscripts/attnfiles/sageattention-2.2.0+cu128torch2.9.0.post3-cp39-abi3-win_amd64.whl filter=lfs diff=lfs merge=lfs -text

runscripts/attnfiles/flash_attn-2.8.3+cu128torch2.8.0cxx11abiFALSE-cp312-cp312-win_amd64.whl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:709d6663d74f96bb0304b5ba2ffee84a4dafbcb56d73d35bc90b7ca8cd44261b
+size 127588011

runscripts/attnfiles/python_3.12.9_include_libs.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59b94a36903a6822c2bb7adf551e33e3aa85b079e0ca3a76ff205a907b7799e6
+size 467075

runscripts/attnfiles/sageattention-2.2.0+cu128torch2.9.0.post3-cp39-abi3-win_amd64.whl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb555644b49eacb26f336bbdf0701779a24c5639a620c963f48a14ab65c2c29c
+size 11963869

runscripts/attnfiles/test_sageattn.py

@@ -0,0 +1,48 @@
+#!/usr/bin/env python3
+
+import torch
+import torch.nn.functional as F
+from sageattention import sageattn
+from torch.nn.attention import SDPBackend, sdpa_kernel
+
+
+def get_rtol_atol(actual, expect):
+    actual = actual.float()
+    expect = expect.float()
+    diff = (actual - expect).abs()
+    eps = torch.tensor(
+        torch.finfo(actual.dtype).eps, device=actual.device, dtype=actual.dtype
+    )
+    rdiff = diff / torch.maximum(torch.maximum(actual.abs(), expect.abs()), eps)
+    return (
+        f"mean_rtol={rdiff.mean().item():.3g} "
+        f"max_rtol={rdiff.max().item():.3g} "
+        f"mean_atol={diff.max().item():.3g} "
+        f"max_atol={diff.max().item():.3g}"
+    )
+
+
+def main():
+    batch_size = 4
+    head_num = 32
+    seq_len = 64
+    head_dim = 128
+    dtype = torch.float16
+
+    q = torch.randn(batch_size, head_num, seq_len, head_dim, device="cuda", dtype=dtype)
+    k = torch.randn_like(q)
+    v = torch.randn_like(q)
+    print("q", tuple(q.shape), q.device, q.dtype)
+
+    # 'Mathematically correct' implementation
+    torch.backends.cuda.enable_math_sdp(True)
+    with sdpa_kernel(SDPBackend.MATH):
+        out_math = F.scaled_dot_product_attention(q, k, v)
+
+    out_sage = sageattn(q, k, v)
+    print("sage vs math:", get_rtol_atol(out_sage, out_math))
+    print("The above (except max_rtol) should be < 0.05 (on RTX 20xx/30xx) or < 0.1 (on RTX 40xx/50xx)")
+
+
+if __name__ == "__main__":
+    main()