MANIFEST.in
README.md
pyproject.toml
b12x/__init__.py
b12x.egg-info/PKG-INFO
b12x.egg-info/SOURCES.txt
b12x.egg-info/dependency_links.txt
b12x.egg-info/requires.txt
b12x.egg-info/top_level.txt
b12x/attention/__init__.py
b12x/attention/workspace.py
b12x/attention/_cute/__init__.py
b12x/attention/_cute/copy.py
b12x/attention/_cute/ops.py
b12x/attention/_cute/pipeline.py
b12x/attention/contiguous/README.md
b12x/attention/contiguous/__init__.py
b12x/attention/contiguous/api.py
b12x/attention/contiguous/block_info.py
b12x/attention/contiguous/cute_dsl_utils.py
b12x/attention/contiguous/forward.py
b12x/attention/contiguous/layout_utils.py
b12x/attention/contiguous/mask.py
b12x/attention/contiguous/named_barrier.py
b12x/attention/contiguous/pack_gqa.py
b12x/attention/contiguous/seqlen_info.py
b12x/attention/contiguous/softmax.py
b12x/attention/contiguous/tile_scheduler.py
b12x/attention/indexer/MSA.md
b12x/attention/indexer/__init__.py
b12x/attention/indexer/api.py
b12x/attention/indexer/contiguous_kernel.py
b12x/attention/indexer/fused_indexer.py
b12x/attention/indexer/kernel.py
b12x/attention/indexer/msa_reference.py
b12x/attention/indexer/paged.py
b12x/attention/indexer/persistent_topk.py
b12x/attention/indexer/reference.py
b12x/attention/indexer/schedule_metadata.py
b12x/attention/indexer/scratch.py
b12x/attention/indexer/tiled_topk.py
b12x/attention/mla/__init__.py
b12x/attention/mla/api.py
b12x/attention/mla/compressed_api.py
b12x/attention/mla/compressed_config.py
b12x/attention/mla/compressed_reference.py
b12x/attention/mla/decode_math.py
b12x/attention/mla/io.py
b12x/attention/mla/io_mg.py
b12x/attention/mla/kernel.py
b12x/attention/mla/merge.py
b12x/attention/mla/packed.py
b12x/attention/mla/prefill.py
b12x/attention/mla/prefill_mg.py
b12x/attention/mla/reference.py
b12x/attention/mla/smem.py
b12x/attention/mla/smem_mg.py
b12x/attention/mla/traits.py
b12x/attention/mla/legacy/__init__.py
b12x/attention/mla/legacy/kernel.py
b12x/attention/mla/legacy/kernel_onepass.py
b12x/attention/mla/legacy/split.py
b12x/attention/mla/legacy/traits.py
b12x/attention/paged/__init__.py
b12x/attention/paged/api.py
b12x/attention/paged/forward_extend_generic.py
b12x/attention/paged/forward_paged.py
b12x/attention/paged/graph_replay.py
b12x/attention/paged/merge.py
b12x/attention/paged/planner.py
b12x/attention/paged/reference.py
b12x/attention/paged/traits.py
b12x/attention/paged/workspace.py
b12x/attention/paged/tuning/__init__.py
b12x/attention/paged/tuning/registry.py
b12x/cute/__init__.py
b12x/cute/compiler.py
b12x/cute/fp4.py
b12x/cute/runtime_control.py
b12x/cute/runtime_patches.py
b12x/cute/scratch.py
b12x/cute/utils.py
b12x/distributed/__init__.py
b12x/distributed/_cuda_ipc.py
b12x/distributed/pcie_oneshot.cu
b12x/distributed/pcie_oneshot.py
b12x/gemm/__init__.py
b12x/gemm/block_fp8_linear.py
b12x/gemm/dense.py
b12x/gemm/mxfp8_linear.py
b12x/gemm/wo_projection.py
b12x/integration/__init__.py
b12x/integration/attention.py
b12x/integration/compressed_scratch.py
b12x/integration/mla.py
b12x/integration/paged_attention_scratch.py
b12x/integration/residual.py
b12x/integration/residual_kernels.py
b12x/integration/scratch.py
b12x/integration/scratch_layout.py
b12x/integration/sparse_mla_scratch.py
b12x/integration/tp_moe.py
b12x/integration/triton_route.py
b12x/moe/__init__.py
b12x/moe/fused/__init__.py
b12x/moe/fused/activations.py
b12x/moe/fused/dynamic.py
b12x/moe/fused/micro.py
b12x/moe/fused/reference.py
b12x/moe/fused/relu2.py
b12x/moe/fused/silu.py
b12x/moe/fused/w4a16/__init__.py
b12x/moe/fused/w4a16/host.py
b12x/moe/fused/w4a16/kernel.py
b12x/moe/fused/w4a16/prepare.py
b12x/moe/fused/w4a16/route_pack.py
b12x/moe/fused/w4a8/__init__.py
b12x/moe/fused/w4a8/act.py
b12x/moe/fused/w4a8/gemm.py
b12x/moe/fused/w4a8/pipeline.py
b12x/moe/fused/w4a8/quant.py
b12x/moe/fused/w4a8/route.py
b12x/moe/tuning/__init__.py
b12x/moe/tuning/decode.max_active_clusters.py
b12x/moe/tuning/registry.py
b12x/quantization/__init__.py
b12x/quantization/bf16_to_fp4_tma.py