Upload README.md with huggingface_hub

README.md

@@ -18,15 +18,18 @@ Prebuilt Python 3.11 wheels.
 |---|---|---|---|---|---|
 | `ace_step-1.6.0-py3-none-any.whl` | any | — | — | built by us | Pure-Python, cross-platform |
 | `qwen_tts-0.1.1+deapi3-py3-none-any.whl` | any | — | — | built by us | Pure-Python, patched fork (`+deapi3`) |
-| `image_server_kernels-0.4.2+cu128torch2.8-cp311-cp311-win_amd64.whl` | Windows x64 | 12.8 | 2.8 | built by us | **Current.** Adds Sol-Attn sparse attention, W4A8 linear, and in-tree llama.cpp GGUF CUDA kernels (`_C_gguf`), on top of: FP8 dense + grouped GEMM, NVFP4 W4A4, INT8 W8A8 + ConvRot, fused QK-norm+RoPE. SM89 + SM120 |
-| `image_server_kernels-0.3.0-cp311-cp311-win_amd64.whl` | Windows x64 | 12.8 | 2.8 | built by us | Previous build, kept as a fallback. No Sol-Attn, no W4A8, no bundled GGUF kernels |
-| `image_server_kernels-0.3.0+cu129torch2.8-cp311-cp311-linux_x86_64.whl` | Linux x86_64 | 12.9 | 2.8 | built by us | Linux build is still at 0.3.0 — there is no Linux 0.4.2 yet |
+| `image_server_kernels-0.4.3+cu128torch2.8-cp311-cp311-win_amd64.whl` | Windows x64 | 12.8 | 2.8 | built by us | **Current.** FP8 dense + grouped GEMM, NVFP4 W4A4, INT8 W8A8 + ConvRot, W4A8, GGUF, fused QK-norm+RoPE, Sol-Attn. SM89 + SM120 |
+| `image_server_kernels-0.4.3+cu129torch2.8-cp311-cp311-linux_x86_64.whl` | Linux x86_64 | 12.9 | 2.8 | built by us | **Current.** Same op set |
+| `image_server_kernels-0.4.2+cu128torch2.8-cp311-cp311-win_amd64.whl` | Windows x64 | 12.8 | 2.8 | built by us | Previous build, kept as a rollback |
+| `image_server_kernels-0.4.2+cu129torch2.8-cp311-cp311-linux_x86_64.whl` | Linux x86_64 | 12.9 | 2.8 | built by us | Previous build, kept as a rollback |
+| `image_server_kernels-0.3.0-cp311-cp311-win_amd64.whl` | Windows x64 | 12.8 | 2.8 | built by us | Legacy. No Sol-Attn, no W4A8, no bundled GGUF kernels |
+| `image_server_kernels-0.3.0+cu129torch2.8-cp311-cp311-linux_x86_64.whl` | Linux x86_64 | 12.9 | 2.8 | built by us | Legacy. Same omissions as the Windows 0.3.0 |
 | `block_sparse_attn-0.0.2-cp311-cp311-win_amd64.whl` | Windows x64 | 12.8 | 2.8 | built by us | Used by video pipeline |
 | `block_sparse_attn-0.0.2-cp311-cp311-linux_x86_64.whl` | Linux x86_64 | 12.8 | 2.8 | built by us | Used by video pipeline |
 | `q8_kernels-0.0.5-cp311-cp311-win_amd64.whl` | Windows x64 | 12.8 | 2.8 | built by us | Used by LTX video |
 | `q8_kernels-0.0.5-cp311-cp311-linux_x86_64.whl` | Linux x86_64 | 12.8 | 2.8 | built by us | Used by LTX video |
 | `sageattention-2.2.0+cu128torch2.8.0-cp311-cp311-win_amd64.whl` | Windows x64 | 12.8 | 2.8 | [woct0rdho/SageAttention](https://github.com/woct0rdho/SageAttention/releases/tag/v2.2.0-windows) | Mirror of upstream release |
-| `sageattention-2.2.0+cu129torch2.8-cp311-cp311-linux_x86_64.whl` | Linux x86_64 | 12.9 | 2.8 | provenance unconfirmed — see Credits | |
+| `sageattention-2.2.0+cu129torch2.8-cp311-cp311-linux_x86_64.whl` | Linux x86_64 | 12.9 | 2.8 | — | |
 | `flash_attn-2.8.2+cu128torch2.8-cp311-cp311-win_amd64.whl` | Windows x64 | 12.8 | 2.8 | [mjun0812/flash-attention-prebuild-wheels](https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/tag/v0.4.10) | Mirror of upstream release |
 | `flash_attn-2.8.3+cu130torch2.10-cp311-cp311-win_amd64.whl` | Windows x64 | 13.0 | 2.10 | [mjun0812/flash-attention-prebuild-wheels](https://github.com/mjun0812/flash-attention-prebuild-wheels) | Mirror of upstream release |
 | `flash_attn-2.8.3+cu128torch2.8-cp311-cp311-linux_x86_64.whl` | Linux x86_64 | 12.8 | 2.8 | [mjun0812/flash-attention-prebuild-wheels](https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/tag/v0.7.16) | Mirror of upstream release |
@@ -37,7 +40,7 @@ Prebuilt Python 3.11 wheels.
 BASE=https://huggingface.co/deAPI-ai/image-server-wheels/resolve/main
 
 # Windows
-pip install $BASE/image_server_kernels-0.4.2+cu128torch2.8-cp311-cp311-win_amd64.whl
+pip install $BASE/image_server_kernels-0.4.3+cu128torch2.8-cp311-cp311-win_amd64.whl
 pip install $BASE/q8_kernels-0.0.5-cp311-cp311-win_amd64.whl
 pip install $BASE/block_sparse_attn-0.0.2-cp311-cp311-win_amd64.whl
 pip install $BASE/flash_attn-2.8.2+cu128torch2.8-cp311-cp311-win_amd64.whl
@@ -46,39 +49,16 @@ pip install --no-deps $BASE/ace_step-1.6.0-py3-none-any.whl
 pip install --no-deps $BASE/qwen_tts-0.1.1+deapi3-py3-none-any.whl
 
 # Linux
-pip install $BASE/image_server_kernels-0.3.0+cu129torch2.8-cp311-cp311-linux_x86_64.whl
+pip install $BASE/image_server_kernels-0.4.3+cu129torch2.8-cp311-cp311-linux_x86_64.whl
 pip install $BASE/q8_kernels-0.0.5-cp311-cp311-linux_x86_64.whl
 pip install $BASE/block_sparse_attn-0.0.2-cp311-cp311-linux_x86_64.whl
 pip install $BASE/sageattention-2.2.0+cu129torch2.8-cp311-cp311-linux_x86_64.whl
 pip install --no-deps $BASE/flash_attn-2.8.3+cu128torch2.8-cp311-cp311-linux_x86_64.whl
 ```
 
-`sageattention` is required by the MiniMax-H3 video model, which imports it
-unconditionally. Every other model that can use it falls back to SDPA when it is
-absent, so on a box that does not serve H3 the wheel is optional.
-
-> `image_server_kernels` is at **0.4.2 on Windows but only 0.3.0 on Linux** — the
-> Linux build of 0.4.2 has not been made yet. Sol-Attn, W4A8 and the bundled GGUF
-> kernels are therefore Windows-only for now.
-
-## What is verified in `image_server_kernels-0.4.2`
-
-Built and tested on Windows 11 / RTX 5090 Laptop (SM120), CUDA 12.8, torch 2.8.0+cu128.
-
-- `scripts/smoke_test_rebuild.py` — ALL PASS: `fp8_dense_gemm` (per-tensor cos 0.99965,
-  per-channel+bias 0.99972), `dynamic_per_token_scaled_fp8_quant` 0.99965,
-  `fp8_dense_gemm_pre_quantized` 1.00000, `fp8_dense_gemm_sm120` 1.00000 via native
-  CUTLASS 3.x, `rmsnorm_forward` 0.999999.
-- `scripts/validate_gguf_kernels.py` against real GGUF tensors — ALL PASS for
-  Q4_K, Q5_K, Q6_K, Q5_0, Q5_1, Q8_0 (cos 0.99984–1.00000), which covers the bf16
-  K-quant dequant path the in-tree patch exists for.
-- `sol_attn` on real MiniMax-H3 activations (S=15,479, 5 capture points):
-  **2.22x vs SageAttention2 at cos 0.9805** at `tau=1.0`. Both speed and accuracy
-  rise with sequence length (4k → 1.68x/0.965, 15.5k → 2.90x/0.980 in one run),
-  so longer clips do better than this figure.
-
-Not exercised on this box: `sol_attn` on Ada SM89 (compiles and links, but no Ada
-GPU here), `w4a8_linear`, `nvfp4_*`, `fp8_grouped_gemm`, `w8a16_dense_gemm`.
+Wheels are built for **Python 3.11** and the CUDA/torch pair named in each filename.
+Installing one against a different torch build is the first thing to suspect if the
+import fails.
 
 ## Credits
 
@@ -86,26 +66,17 @@ GPU here), `w4a8_linear`, `nvfp4_*`, `fp8_grouped_gemm`, `w8a16_dense_gemm`.
 [mjun0812/flash-attention-prebuild-wheels](https://github.com/mjun0812/flash-attention-prebuild-wheels)
 — all credit for those builds goes to the upstream author.
 
-`sageattention` is **a mirror, not our build**. The library is
-[thu-ml/SageAttention](https://github.com/thu-ml/SageAttention) (Apache-2.0); the
-Windows wheel is built and published by
+The Windows `sageattention` wheel is **a mirror, not our build**. The library is
+[thu-ml/SageAttention](https://github.com/thu-ml/SageAttention) (Apache-2.0); the wheel
+is built and published by
 [woct0rdho](https://github.com/woct0rdho/SageAttention/releases), and we found it
 through [wildminder/AI-windows-whl](https://github.com/wildminder/AI-windows-whl),
 which indexes Windows CUDA wheels. All credit for the library and the build goes to
-them. The Windows file here is byte-identical to the upstream release
+them. The file here is byte-identical to the upstream release
 (`sha256:4379951403809dfcd5b1e10d35e287abf42afafb7b27615c09d6062cfbdf230a`).
 
-> The **Linux** `sageattention-2.2.0+cu129torch2.8` wheel's provenance is not
-> recorded — we have not confirmed whether it is our own build or a mirror. Fill
-> this in before relying on the attribution.
-
 We mirror third-party wheels so the install scripts have a single source of truth and
 do not break if upstream release URLs change.
 
-`ace_step`, `qwen_tts`, `image_server_kernels`, `block_sparse_attn` and `q8_kernels`
-were built in-house.
-
-Sol-Attn in `image_server_kernels` wraps CUDA kernels from comfy-kitchen
-(Apache-2.0); see `third_party/comfy_kitchen_sol/PROVENANCE.md` in the
-`image-server-kernels` repo. The bundled GGUF kernels are llama.cpp's `ggml-cuda`
-(MIT), vendored per `third_party/llamacpp_gguf/PROVENANCE.md`.
+The remaining wheels (`ace_step`, `qwen_tts`, `image_server_kernels`,
+`block_sparse_attn`, `q8_kernels`) were built in-house.