Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+flash_attn-2.8.3.post1-cp312-cp312-linux_x86_64.whl filter=lfs diff=lfs merge=lfs -text
+sageattention-2.2.0-cp312-cp312-linux_x86_64.whl filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,10 @@
+This repository redistributes binary wheels of two open-source projects.
+
+- SageAttention 2.2 — licensed under Apache License 2.0
+  https://github.com/thu-ml/SageAttention/blob/main/LICENSE
+- FlashAttention 2.8.3 — licensed under BSD 3-Clause License
+  https://github.com/Dao-AILab/flash-attention/blob/main/LICENSE
+
+Each wheel contains its own LICENSE file inside its dist-info/ directory.
+Redistribution of these wheels is permitted under both licenses; users
+are responsible for complying with the respective terms.

README.md

@@ -0,0 +1,150 @@
+---
+license: apache-2.0
+tags:
+  - sageattention
+  - flash-attention
+  - flash-attn
+  - blackwell
+  - sm_120
+  - rtx-pro-6000
+  - rtx-5090
+  - cuda-13
+  - pytorch-2.11
+  - python-3.12
+  - wheels
+  - prebuilt
+  - comfyui
+language:
+  - en
+---
+
+# SageAttention 2.2 + FlashAttention 2.8.3 — Linux wheels for Blackwell (sm_120) on PyTorch 2.11 / CUDA 13.0
+
+Community-built `.whl` files for a combo that, as of June 2026, isn't shipped by upstream:
+
+- **SageAttention 2.2.0** — only Windows wheels in upstream releases; Linux requires source build
+- **FlashAttention 2.8.3.post1** — Dao-AILab releases don't cover cu130 + torch 2.11 yet
+
+If your hardware/software matches the table below, these install in seconds instead of 5–40 minutes of source compilation.
+
+---
+
+## Compatibility
+
+| Requirement | Value | Why it matters |
+|---|---|---|
+| Python | **3.12.x** (built on 3.12.3) | `cp312` ABI tag — won't load on 3.11 or 3.13 |
+| PyTorch | **2.11.0+cu130** | C++ ABI must match — won't load with `cu126` torch |
+| CUDA toolkit | 13.0 (driver compatible with cu130) | required for runtime |
+| GPU compute capability | **12.0 (Blackwell)** — RTX PRO 6000 Blackwell, RTX 5090/5080/5070 Ti, B100/B200, etc. | wheels compiled with `arch=compute_120,code=sm_120` natively (no PTX-JIT fallback) |
+| OS | Linux x86_64 | `manylinux` not applied — built on Ubuntu, untested elsewhere |
+
+If **any** row doesn't match — these wheels probably won't load and you'll need to build from source.
+
+---
+
+## Files & hashes
+
+| File | Size | SHA256 |
+|---|---|---|
+| `sageattention-2.2.0-cp312-cp312-linux_x86_64.whl` | 14.8 MB | `4a47a462059bf09b7d10be68f93a0a462dd6184068298fc6ff10e94751fa6dd4` |
+| `flash_attn-2.8.3.post1-cp312-cp312-linux_x86_64.whl` | 65.8 MB | `a77fe499d1c7244e0549d9d2700def0c1c82a1b629d993204e9be78debd96cb8` |
+
+Verify before installing:
+```bash
+sha256sum sageattention-2.2.0-cp312-cp312-linux_x86_64.whl
+sha256sum flash_attn-2.8.3.post1-cp312-cp312-linux_x86_64.whl
+```
+
+---
+
+## Install
+
+```bash
+pip install sageattention-2.2.0-cp312-cp312-linux_x86_64.whl
+pip install flash_attn-2.8.3.post1-cp312-cp312-linux_x86_64.whl
+```
+
+Smoke test:
+```python
+import torch
+import sageattention
+from sageattention import sageattn
+from flash_attn import flash_attn_func
+
+q = torch.randn(2, 8, 256, 64, dtype=torch.float16, device='cuda')
+k = torch.randn(2, 8, 256, 64, dtype=torch.float16, device='cuda')
+v = torch.randn(2, 8, 256, 64, dtype=torch.float16, device='cuda')
+
+# sageattention expects (B, H, S, D) by default with tensor_layout='HND'
+out_sage = sageattn(q, k, v, tensor_layout='HND', is_causal=False)
+
+# flash_attn_func expects (B, S, H, D)
+q2 = q.transpose(1, 2).contiguous()
+k2 = k.transpose(1, 2).contiguous()
+v2 = v.transpose(1, 2).contiguous()
+out_flash = flash_attn_func(q2, k2, v2, causal=False)
+
+print('sageattention:', out_sage.shape, '— OK')
+print('flash_attn:   ', out_flash.shape, '— OK')
+```
+
+---
+
+## Build environment (exactly what produced these wheels)
+
+```
+OS:       Ubuntu 24.04 (kernel 6.14.0-1010-oem)
+Python:   3.12.3
+PyTorch:  2.11.0+cu130
+CUDA:     13.0.r13.0 / V13.0.88 (apt cuda-toolkit-13-0)
+ninja:    1.11.1.4
+setuptools: 81.0.0
+GPU:      NVIDIA RTX PRO 6000 Blackwell Workstation Edition (sm_120)
+```
+
+Build flags:
+
+**sageattention:**
+```bash
+TORCH_CUDA_ARCH_LIST="12.0" MAX_JOBS=4 NVCC_THREADS=2 \
+  pip install . --no-build-isolation
+```
+
+**flash-attn:**
+```bash
+TORCH_CUDA_ARCH_LIST="12.0" FLASH_ATTN_CUDA_ARCHS="120" MAX_JOBS=2 NVCC_THREADS=1 \
+  pip install flash-attn==2.8.3.post1 --no-build-isolation
+```
+
+Backward kernels are **included** in flash-attn — training works, not just inference.
+
+---
+
+## Licenses
+
+Both upstream projects allow redistribution; redistributing these binary builds preserves their license terms:
+
+- **SageAttention** — [Apache License 2.0](https://github.com/thu-ml/SageAttention/blob/main/LICENSE)
+- **FlashAttention** — [BSD 3-Clause](https://github.com/Dao-AILab/flash-attention/blob/main/LICENSE)
+
+The wheels contain copies of `LICENSE` files in their `*.dist-info/licenses/` directories. Originals are at:
+- https://github.com/thu-ml/SageAttention
+- https://github.com/Dao-AILab/flash-attention
+
+---
+
+## Disclaimers (read these)
+
+- **Not official.** This is a community build. Authors of upstream packages have no involvement.
+- **Smoke-tested only.** Each wheel was verified to: (1) import without errors, (2) execute a small forward pass on a Blackwell GPU and return a tensor of expected shape and dtype. No accuracy benchmarks, no model-level integration tests were run.
+- **No warranty.** If these wheels misbehave on your system — wrong results, crashes, performance regressions — that's on you to verify against a source build before relying on them.
+- **Reproducible-ish.** Wheel SHA-256 may differ slightly between rebuilds due to embedded timestamps even when binary kernel content is identical.
+
+---
+
+## Attribution
+
+These wheels were built by **JahJedi** with assistance from **Claude Code** (Anthropic) — the diagnosis of an earlier system freeze under unbounded `ninja`, the choice of safe build parameters (`MAX_JOBS`, `NVCC_THREADS`, single-arch targeting), and this README itself were Claude-assisted. Verification commands were run by the human; final responsibility for the published artifacts is the human's.
+
+If anything in these wheels turns out to be wrong, ping the human first — they can't claim AI ignorance.

flash_attn-2.8.3.post1-cp312-cp312-linux_x86_64.whl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a77fe499d1c7244e0549d9d2700def0c1c82a1b629d993204e9be78debd96cb8
+size 65808081

sageattention-2.2.0-cp312-cp312-linux_x86_64.whl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a47a462059bf09b7d10be68f93a0a462dd6184068298fc6ff10e94751fa6dd4
+size 14856977