FlashRWKV2 0.1.0a1__tar.gz

flashrwkv2-0.1.0a1/FlashRWKV2.egg-info/PKG-INFO

@@ -0,0 +1,102 @@
+Metadata-Version: 2.4
+Name: FlashRWKV2
+Version: 0.1.0a1
+Summary: High-performance CUDA operator library for RWKV-7
+License-Expression: MIT
+Project-URL: Repository, https://github.com/rwkv-rs/FlashRWKV2
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+License-File: NOTICE
+Requires-Dist: torch==2.11.0
+Dynamic: license-file
+
+# FlashRWKV2
+
+FlashRWKV2 is a high-performance CUDA operator library for RWKV-7. It provides
+composable inference and training operators; complete models, schedulers, and
+training frameworks remain the responsibility of downstream projects.
+
+## Requirements
+
+- Python 3.10 or newer
+- An NVIDIA CUDA build environment and a supported CUDA device
+- `uv`, using the repository-local `./.venv`
+- SM120 or newer for the current native build
+
+The runtime and reproducible source-build contracts currently pin PyTorch
+2.11.0. When the build cannot detect a local GPU, set
+`TORCH_CUDA_ARCH_LIST=12.0` explicitly.
+
+## Installation
+
+The current alpha is distributed as source and builds the CUDA extension on the
+target machine:
+
+```bash
+TORCH_CUDA_ARCH_LIST=12.0 \
+  python -m pip install --pre FlashRWKV2
+```
+
+For development from a checkout:
+
+```bash
+git clone https://github.com/rwkv-rs/FlashRWKV2.git
+cd FlashRWKV2
+uv sync
+TORCH_CUDA_ARCH_LIST=12.0 \
+  ./.venv/bin/python -m pip install -v --no-build-isolation -e .
+```
+
+## Kernel API
+
+See the [Kernel API reference](https://github.com/rwkv-rs/FlashRWKV2/blob/main/docs/kernel_api.md)
+for the complete public operator surface and tensor contracts.
+
+## Tests
+
+```bash
+./.venv/bin/python -m pytest -q
+```
+
+CUDA tests require a successfully built `flashrwkv2._C` extension and a
+supported GPU.
+
+## Benchmarks
+
+Operator benchmarks live in [`benchmarks/`](benchmarks/). For example, run the
+WKV7 recurrent correctness benchmark with:
+
+```bash
+./.venv/bin/python -m benchmarks.tmix.wkv7.bench \
+  --shapes h32d64 \
+  --dtype bfloat16 \
+  --correctness-only \
+  --output /tmp/flashrwkv2-wkv7-correctness.json
+```
+
+These benchmarks measure individual operators. They do not report or infer
+model-level latency.
+
+For Albatross-compatible TMix low-rank inference, `varlen` means that the
+operator consumes packed token rows; it does not mean that one fused kernel is
+used for every row count.  The public composite callers automatically use the
+canonical fused rank-in window at `M<=7`, the fused rank-out/value-residual
+window at `M<=4`, and the canonical large-row linear dispatcher otherwise.
+Callers may provide both the original checkpoint layout and a runtime layout.
+Both layouts must be prepared outside the timed forward region and retained for
+the lifetime of the inference weights; FlashRWKV2 never transposes or copies a
+missing layout during dispatch.
+
+The existing low-rank operator benchmark reports correctness and latency as
+JSON without adding a model-level benchmark:
+
+```bash
+./.venv/bin/python benchmarks/tmix/linear/bench.py \
+  --operator projection-group --layout both --channels 4096
+```
+
+## License
+
+FlashRWKV2 is distributed under the
+[MIT License](https://github.com/rwkv-rs/FlashRWKV2/blob/main/LICENSE).

flashrwkv2-0.1.0a1/FlashRWKV2.egg-info/SOURCES.txt

@@ -0,0 +1,109 @@
+LICENSE
+MANIFEST.in
+NOTICE
+README.md
+pyproject.toml
+setup.py
+FlashRWKV2.egg-info/PKG-INFO
+FlashRWKV2.egg-info/SOURCES.txt
+FlashRWKV2.egg-info/dependency_links.txt
+FlashRWKV2.egg-info/not-zip-safe
+FlashRWKV2.egg-info/requires.txt
+FlashRWKV2.egg-info/top_level.txt
+LICENSES/Apache-2.0.txt
+csrc/bindings.cpp
+csrc/bindings.h
+csrc/registration.cpp
+csrc/validation.cpp
+csrc/validation.h
+csrc/sm120/cmix/mix/infer_fp16_forward_varlen.cpp
+csrc/sm120/cmix/mix/infer_fp16_forward_varlen.cu
+csrc/sm120/cmix/sparse/infer_fp16_forward_varlen.cpp
+csrc/sm120/cmix/sparse/infer_fp16_forward_varlen.cu
+csrc/sm120/embedding/infer_fp16_forward_varlen.cpp
+csrc/sm120/embedding/infer_fp16_forward_varlen.cu
+csrc/sm120/head/linear/infer_fp16_forward_varlen.cpp
+csrc/sm120/head/linear/infer_fp16_forward_varlen.cu
+csrc/sm120/tmix/kk_a_gate/infer_fp16_forward_varlen.cpp
+csrc/sm120/tmix/kk_a_gate/infer_fp16_forward_varlen.cu
+csrc/sm120/tmix/linear/infer_fp16_forward_varlen.cpp
+csrc/sm120/tmix/linear/infer_fp16_forward_varlen.cu
+csrc/sm120/tmix/lnx_rkvres_xg/infer_fp16_forward_varlen.cpp
+csrc/sm120/tmix/lnx_rkvres_xg/infer_fp16_forward_varlen.cu
+csrc/sm120/tmix/mix6/infer_fp16_forward_varlen.cpp
+csrc/sm120/tmix/mix6/infer_fp16_forward_varlen.cu
+csrc/sm120/tmix/normalization/infer_fp16_forward_varlen.cpp
+csrc/sm120/tmix/normalization/infer_fp16_forward_varlen.cu
+csrc/sm120/tmix/vres_gate/infer_fp16_forward_varlen.cpp
+csrc/sm120/tmix/vres_gate/infer_fp16_forward_varlen.cu
+csrc/sm120/tmix/wkv7/infer_chunk_bf16_forward_varlen.cpp
+csrc/sm120/tmix/wkv7/infer_chunk_bf16_forward_varlen.cu
+csrc/sm120/tmix/wkv7/infer_recurrent_fp16_forward_varlen.cpp
+csrc/sm120/tmix/wkv7/infer_recurrent_fp16_forward_varlen.cu
+csrc/sm120/tmix/wkv7/infer_recurrent_fp32io16_forward_varlen.cpp
+csrc/sm120/tmix/wkv7/infer_recurrent_fp32io16_forward_varlen.cu
+csrc/sm120/tmix/wkv7/recurrent_decay.cuh
+csrc/sm90/cmix/mix/pretrain_bf16_backward.cpp
+csrc/sm90/cmix/mix/pretrain_bf16_backward.cu
+csrc/sm90/cmix/mix/pretrain_bf16_forward.cpp
+csrc/sm90/cmix/mix/pretrain_bf16_forward.cu
+csrc/sm90/head/l2wrap_ce/pretrain_bf16_forward.cpp
+csrc/sm90/head/l2wrap_ce/pretrain_bf16_forward.cu
+csrc/sm90/loss/l2wrap_ce/pretrain_bf16_backward.cpp
+csrc/sm90/loss/l2wrap_ce/pretrain_bf16_backward.cu
+csrc/sm90/loss/l2wrap_ce/pretrain_bf16_forward.cpp
+csrc/sm90/loss/l2wrap_ce/pretrain_bf16_forward.cu
+csrc/sm90/rl_infctx/wkv7/rl_infctx_chunk_fp32io16_backward.cpp
+csrc/sm90/rl_infctx/wkv7/rl_infctx_chunk_fp32io16_backward.cu
+csrc/sm90/rl_infctx/wkv7/rl_infctx_chunk_fp32io16_forward.cpp
+csrc/sm90/rl_infctx/wkv7/rl_infctx_chunk_fp32io16_forward.cu
+csrc/sm90/rl_infctx/wkv7/rl_infctx_chunk_fp32io16_replay.cuh
+csrc/sm90/tmix/a_gate/pretrain_bf16_backward.cpp
+csrc/sm90/tmix/a_gate/pretrain_bf16_backward.cu
+csrc/sm90/tmix/a_gate/pretrain_bf16_forward.cpp
+csrc/sm90/tmix/a_gate/pretrain_bf16_forward.cu
+csrc/sm90/tmix/kk_pre/pretrain_bf16_backward.cpp
+csrc/sm90/tmix/kk_pre/pretrain_bf16_backward.cu
+csrc/sm90/tmix/kk_pre/pretrain_bf16_forward.cpp
+csrc/sm90/tmix/kk_pre/pretrain_bf16_forward.cu
+csrc/sm90/tmix/lnx_rkvres_xg/pretrain_bf16_backward.cpp
+csrc/sm90/tmix/lnx_rkvres_xg/pretrain_bf16_backward.cu
+csrc/sm90/tmix/lnx_rkvres_xg/pretrain_bf16_forward.cpp
+csrc/sm90/tmix/lnx_rkvres_xg/pretrain_bf16_forward.cu
+csrc/sm90/tmix/mix6/pretrain_bf16_backward.cpp
+csrc/sm90/tmix/mix6/pretrain_bf16_backward.cu
+csrc/sm90/tmix/mix6/pretrain_bf16_forward.cpp
+csrc/sm90/tmix/mix6/pretrain_bf16_forward.cu
+csrc/sm90/tmix/vres_gate/pretrain_bf16_backward.cpp
+csrc/sm90/tmix/vres_gate/pretrain_bf16_backward.cu
+csrc/sm90/tmix/vres_gate/pretrain_bf16_forward.cpp
+csrc/sm90/tmix/vres_gate/pretrain_bf16_forward.cu
+csrc/sm90/tmix/wkv7/pretrain_recurrent_bf16_forward.cpp
+csrc/sm90/tmix/wkv7/pretrain_recurrent_bf16_forward.cu
+csrc/sm90/tmix/wkv7/statetune_recurrent_fp32io16_backward.cpp
+csrc/sm90/tmix/wkv7/statetune_recurrent_fp32io16_backward.cu
+csrc/sm90/tmix/wkv7/statetune_recurrent_fp32io16_forward.cpp
+csrc/sm90/tmix/wkv7/statetune_recurrent_fp32io16_forward.cu
+csrc/validation/recurrent_metadata.cu
+flashrwkv2/__init__.py
+flashrwkv2/cmix/mix/__init__.py
+flashrwkv2/cmix/sparse/__init__.py
+flashrwkv2/embedding/__init__.py
+flashrwkv2/head/l2wrap_ce/__init__.py
+flashrwkv2/head/linear/__init__.py
+flashrwkv2/loss/l2wrap_ce/__init__.py
+flashrwkv2/rl_infctx/__init__.py
+flashrwkv2/rl_infctx/wkv7/__init__.py
+flashrwkv2/tmix/a_gate/__init__.py
+flashrwkv2/tmix/kk_a_gate/__init__.py
+flashrwkv2/tmix/kk_pre/__init__.py
+flashrwkv2/tmix/linear/__init__.py
+flashrwkv2/tmix/lnx_rkvres_xg/__init__.py
+flashrwkv2/tmix/lnx_rkvres_xg/pretrain.py
+flashrwkv2/tmix/mix6/__init__.py
+flashrwkv2/tmix/normalization/__init__.py
+flashrwkv2/tmix/vres_gate/__init__.py
+flashrwkv2/tmix/wkv7/__init__.py
+flashrwkv2/tmix/wkv7/chunk.py
+flashrwkv2/tmix/wkv7/pretrain.py
+flashrwkv2/tmix/wkv7/statetune.py

flashrwkv2-0.1.0a1/FlashRWKV2.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

flashrwkv2-0.1.0a1/FlashRWKV2.egg-info/not-zip-safe

@@ -0,0 +1 @@
+

flashrwkv2-0.1.0a1/FlashRWKV2.egg-info/requires.txt

@@ -0,0 +1 @@
+torch==2.11.0

flashrwkv2-0.1.0a1/FlashRWKV2.egg-info/top_level.txt

@@ -0,0 +1 @@
+flashrwkv2

flashrwkv2-0.1.0a1/LICENSE

@@ -0,0 +1,22 @@
+MIT License
+
+Copyright (c) 2026 MoonshotAI
+Copyright (c) 2026 FlashRWKV2 contributors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

flashrwkv2-0.1.0a1/LICENSES/Apache-2.0.txt

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

flashrwkv2-0.1.0a1/MANIFEST.in

@@ -0,0 +1,7 @@
+include LICENSE
+include NOTICE
+include README.md
+include pyproject.toml
+include setup.py
+recursive-include LICENSES *.txt
+recursive-include csrc *.cpp *.cu *.cuh *.h *.hpp

flashrwkv2-0.1.0a1/NOTICE

@@ -0,0 +1,82 @@
+FlashRWKV2
+
+This repository began from MoonshotAI/FlashKDA, distributed under the MIT
+License. The original KDA attention operator, Python package, tests, and
+benchmark data-plane were removed. The following new RWKV-7 files retain the
+MIT boundary and derive their explicit K1/K2 launch separation from
+MoonshotAI/FlashKDA commit
+1ce47ea3bb22c84eb9cc665028399cf35e8ffb0b:
+
+- csrc/infer/wkv7/infer_common_chunk_bf16_forward_k1_prepare.cu
+- csrc/infer/wkv7/infer_common_chunk_bf16_forward_k2_recurrence.cu
+- csrc/infer/wkv7/infer_common_chunk_bf16_bindings.cpp
+
+They implement the canonical RWKV-7 recurrence, not the KDA attention
+mathematics.
+
+The following files contain material adapted from the vLLM project and its
+RWKV/Albatross integration at rwkv-rs/vllm-rwkv commit
+6d683f9e49a2997e405c47edc147872c8609513b:
+
+- csrc/infer/wkv7/infer_common_recurrent_varlen_bindings.cpp
+- csrc/infer/wkv7/infer_common_recurrent_fp32io16_forward_varlen.cu
+- csrc/infer/wkv7/infer_sm80_recurrent_fp16_forward_varlen.cu
+- benchmarks/infer/wkv7/benchmark_infer_recurrent_fp16_fp32io16_forward_varlen.py
+- benchmarks/benchmark_rwkv7.py
+- benchmarks/benchmark_training.py
+- benchmarks/kernel_benchmark.py
+
+The FP16 recurrent lineage includes BlinkDL/Albatross faster3a_2607 commit
+63c53f4abf2cd891dd3a18c8f44f5b2cccc8c64b.
+
+The following fused FP16 inference block operators are adapted from
+BlinkDL/Albatross faster3a_2607 commit
+ee3308f6922e59f2166c7fac3c5a192340a2b48e:
+
+- csrc/infer/tmix/infer_common_tmix_fp16_forward_registration.cpp
+- csrc/infer/tmix/infer_common_tmix_fp16_forward.cu
+- csrc/infer/cmix/infer_common_cmix_fp16_forward_registration.cpp
+- csrc/infer/cmix/infer_common_cmix_fp16_forward.cu
+- csrc/infer/elementwise/infer_common_elementwise_fp16_forward_registration.cpp
+- csrc/infer/elementwise/infer_common_elementwise_fp16_forward.cu
+
+The already-owned FP16/FP32 WKV recurrence files from that source are not
+duplicated. These derived files are licensed under the Apache License,
+Version 2.0; see LICENSES/Apache-2.0.txt.
+
+The following files contain recurrent checkpoint and reverse-recurrence
+methods adapted from BlinkDL/RWKV-LM commit
+952102498e9ed367ea0a59ee64106916d474d30f:
+
+- csrc/common/wkv7/recurrent_common_fp32io16.cpp
+- csrc/common/wkv7/recurrent_common_fp32io16.h
+- csrc/common/wkv7/recurrent_common_fp32io16_forward.cu
+- csrc/common/wkv7/recurrent_common_fp32io16_backward.cu
+- csrc/pretrain/wkv7/pretrain_common_recurrent_fp32io16_bindings.cpp
+- csrc/statetune/wkv7/statetune_common_recurrent_fp32io16_bindings.cpp
+- csrc/pretrain/wkv7/pretrain_common_head_l2wrap_ce_bf16_forward_registration.cpp
+- csrc/pretrain/wkv7/pretrain_common_head_l2wrap_ce_bf16_forward.cu
+- csrc/pretrain/wkv7/pretrain_common_cmix_bf16_forward_backward_registration.cpp
+- csrc/pretrain/wkv7/pretrain_sm90_cmix_bf16_forward_backward.cu
+- csrc/pretrain/wkv7/pretrain_common_l2wrap_ce_bf16_forward_backward_registration.cpp
+- csrc/pretrain/wkv7/pretrain_common_l2wrap_ce_bf16_forward_backward.cu
+- csrc/pretrain/wkv7/pretrain_common_tmix_a_gate_bf16_forward_backward_registration.cpp
+- csrc/pretrain/wkv7/pretrain_common_tmix_a_gate_bf16_forward_backward.cu
+- csrc/pretrain/wkv7/pretrain_common_tmix_kk_pre_bf16_forward_backward_registration.cpp
+- csrc/pretrain/wkv7/pretrain_sm90_tmix_kk_pre_bf16_forward_backward.cu
+- csrc/pretrain/wkv7/pretrain_common_tmix_lnx_rkvres_xg_bf16_forward_backward_registration.cpp
+- csrc/pretrain/wkv7/pretrain_common_tmix_lnx_rkvres_xg_bf16_forward_backward.cu
+- csrc/pretrain/wkv7/pretrain_common_tmix_mix6_bf16_forward_backward_registration.cpp
+- csrc/pretrain/wkv7/pretrain_sm90_tmix_mix6_bf16_forward_backward.cu
+- csrc/pretrain/wkv7/pretrain_common_tmix_vres_gate_bf16_forward_backward_registration.cpp
+- csrc/pretrain/wkv7/pretrain_common_tmix_vres_gate_bf16_forward_backward.cu
+
+Those derived files are licensed under the Apache License, Version 2.0 and
+retain SPDX/source-revision headers. See LICENSES/Apache-2.0.txt.
+
+The FLA chunk_rwkv7 and fused_recurrent_rwkv7 implementations are used only
+through an external correctness and benchmark dependency. No FLA
+implementation source is bundled here.
+
+All other original FlashRWKV2 work is licensed under the repository's MIT
+license unless a file states otherwise.

flashrwkv2-0.1.0a1/PKG-INFO

@@ -0,0 +1,102 @@
+Metadata-Version: 2.4
+Name: FlashRWKV2
+Version: 0.1.0a1
+Summary: High-performance CUDA operator library for RWKV-7
+License-Expression: MIT
+Project-URL: Repository, https://github.com/rwkv-rs/FlashRWKV2
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+License-File: NOTICE
+Requires-Dist: torch==2.11.0
+Dynamic: license-file
+
+# FlashRWKV2
+
+FlashRWKV2 is a high-performance CUDA operator library for RWKV-7. It provides
+composable inference and training operators; complete models, schedulers, and
+training frameworks remain the responsibility of downstream projects.
+
+## Requirements
+
+- Python 3.10 or newer
+- An NVIDIA CUDA build environment and a supported CUDA device
+- `uv`, using the repository-local `./.venv`
+- SM120 or newer for the current native build
+
+The runtime and reproducible source-build contracts currently pin PyTorch
+2.11.0. When the build cannot detect a local GPU, set
+`TORCH_CUDA_ARCH_LIST=12.0` explicitly.
+
+## Installation
+
+The current alpha is distributed as source and builds the CUDA extension on the
+target machine:
+
+```bash
+TORCH_CUDA_ARCH_LIST=12.0 \
+  python -m pip install --pre FlashRWKV2
+```
+
+For development from a checkout:
+
+```bash
+git clone https://github.com/rwkv-rs/FlashRWKV2.git
+cd FlashRWKV2
+uv sync
+TORCH_CUDA_ARCH_LIST=12.0 \
+  ./.venv/bin/python -m pip install -v --no-build-isolation -e .
+```
+
+## Kernel API
+
+See the [Kernel API reference](https://github.com/rwkv-rs/FlashRWKV2/blob/main/docs/kernel_api.md)
+for the complete public operator surface and tensor contracts.
+
+## Tests
+
+```bash
+./.venv/bin/python -m pytest -q
+```
+
+CUDA tests require a successfully built `flashrwkv2._C` extension and a
+supported GPU.
+
+## Benchmarks
+
+Operator benchmarks live in [`benchmarks/`](benchmarks/). For example, run the
+WKV7 recurrent correctness benchmark with:
+
+```bash
+./.venv/bin/python -m benchmarks.tmix.wkv7.bench \
+  --shapes h32d64 \
+  --dtype bfloat16 \
+  --correctness-only \
+  --output /tmp/flashrwkv2-wkv7-correctness.json
+```
+
+These benchmarks measure individual operators. They do not report or infer
+model-level latency.
+
+For Albatross-compatible TMix low-rank inference, `varlen` means that the
+operator consumes packed token rows; it does not mean that one fused kernel is
+used for every row count.  The public composite callers automatically use the
+canonical fused rank-in window at `M<=7`, the fused rank-out/value-residual
+window at `M<=4`, and the canonical large-row linear dispatcher otherwise.
+Callers may provide both the original checkpoint layout and a runtime layout.
+Both layouts must be prepared outside the timed forward region and retained for
+the lifetime of the inference weights; FlashRWKV2 never transposes or copies a
+missing layout during dispatch.
+
+The existing low-rank operator benchmark reports correctness and latency as
+JSON without adding a model-level benchmark:
+
+```bash
+./.venv/bin/python benchmarks/tmix/linear/bench.py \
+  --operator projection-group --layout both --channels 4096
+```
+
+## License
+
+FlashRWKV2 is distributed under the
+[MIT License](https://github.com/rwkv-rs/FlashRWKV2/blob/main/LICENSE).