https://tilelang-tutorial.huandu.me/https://tilelang-tutorial.huandu.me/01-why-tile-dsl/https://tilelang-tutorial.huandu.me/02-from-pytorch-to-kernels/https://tilelang-tutorial.huandu.me/03-design-philosophy/https://tilelang-tutorial.huandu.me/04-architecture-and-compilation-flow/https://tilelang-tutorial.huandu.me/05-comparisons/https://tilelang-tutorial.huandu.me/06-hardware-intro-nv-vs-ascend/https://tilelang-tutorial.huandu.me/07-quickstart/https://tilelang-tutorial.huandu.me/08-language-basics/https://tilelang-tutorial.huandu.me/09-core-primitives/https://tilelang-tutorial.huandu.me/10-layout-and-memory-intro/https://tilelang-tutorial.huandu.me/11-pipeline-and-autotuning/https://tilelang-tutorial.huandu.me/12-debugging-and-profiling/https://tilelang-tutorial.huandu.me/13-operators-elementwise-norm-rope/https://tilelang-tutorial.huandu.me/14-operators-gemm/https://tilelang-tutorial.huandu.me/15-operators-attention/https://tilelang-tutorial.huandu.me/16-operators-moe-mla/https://tilelang-tutorial.huandu.me/17-e2e-training-7b/https://tilelang-tutorial.huandu.me/18-e2e-inference-serving/https://tilelang-tutorial.huandu.me/19-ascend-hands-on/https://tilelang-tutorial.huandu.me/20-ir-and-pass-pipeline/https://tilelang-tutorial.huandu.me/21-layout-inference/https://tilelang-tutorial.huandu.me/22-pipeline-sync-multibuffer/https://tilelang-tutorial.huandu.me/23-autoschedule-smt/https://tilelang-tutorial.huandu.me/24-nvidia-backend-codegen/https://tilelang-tutorial.huandu.me/25-ascend-backend-codegen/https://tilelang-tutorial.huandu.me/26-runtime-jit-abi/https://tilelang-tutorial.huandu.me/27-low-precision-quantization/https://tilelang-tutorial.huandu.me/28-operator-design-methodology/https://tilelang-tutorial.huandu.me/29-compiler-engineering/https://tilelang-tutorial.huandu.me/30-frontier/https://tilelang-tutorial.huandu.me/31-contributing/https://tilelang-tutorial.huandu.me/32-deepseek-ascend-release/https://tilelang-tutorial.huandu.me/33-tilekernels-two-backends/https://tilelang-tutorial.huandu.me/34-tilekernels-engram-mhc/https://tilelang-tutorial.huandu.me/35-deepgemm-ascend/https://tilelang-tutorial.huandu.me/36-flashmla-ascend/https://tilelang-tutorial.huandu.me/37-deepselect-topk/https://tilelang-tutorial.huandu.me/38-deepep-ascend/https://tilelang-tutorial.huandu.me/39-deepseek-e2e-supernode/https://tilelang-tutorial.huandu.me/code/01-why-tile-dsl/https://tilelang-tutorial.huandu.me/code/01-why-tile-dsl/01-op-cost-model/https://tilelang-tutorial.huandu.me/code/01-why-tile-dsl/01-op-cost-model/op_cost_model.py/https://tilelang-tutorial.huandu.me/code/01-why-tile-dsl/01-op-cost-model/README.md/https://tilelang-tutorial.huandu.me/code/01-why-tile-dsl/02-first-tile-kernel/https://tilelang-tutorial.huandu.me/code/01-why-tile-dsl/02-first-tile-kernel/README.md/https://tilelang-tutorial.huandu.me/code/01-why-tile-dsl/02-first-tile-kernel/tile_gemm_sketch.py/https://tilelang-tutorial.huandu.me/code/01-why-tile-dsl/README.md/https://tilelang-tutorial.huandu.me/code/02-from-pytorch-to-kernels/https://tilelang-tutorial.huandu.me/code/02-from-pytorch-to-kernels/01-op-traffic-model/https://tilelang-tutorial.huandu.me/code/02-from-pytorch-to-kernels/01-op-traffic-model/op_traffic.py/https://tilelang-tutorial.huandu.me/code/02-from-pytorch-to-kernels/01-op-traffic-model/README.md/https://tilelang-tutorial.huandu.me/code/02-from-pytorch-to-kernels/02-step-budget/https://tilelang-tutorial.huandu.me/code/02-from-pytorch-to-kernels/02-step-budget/README.md/https://tilelang-tutorial.huandu.me/code/02-from-pytorch-to-kernels/02-step-budget/step_budget.py/https://tilelang-tutorial.huandu.me/code/02-from-pytorch-to-kernels/README.md/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/01-layer-inventory/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/01-layer-inventory/layer_inventory.py/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/01-layer-inventory/README.md/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/02-one-kernel-two-levels/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/02-one-kernel-two-levels/one_kernel_two_levels.py/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/02-one-kernel-two-levels/README.md/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/03-schedule-as-annotation/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/03-schedule-as-annotation/README.md/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/03-schedule-as-annotation/schedule_as_annotation.py/https://tilelang-tutorial.huandu.me/code/03-design-philosophy/README.md/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/compile-path-check/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/compile-path-check/check_compile_path.py/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/compile-path-check/README.md/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/pass-order/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/pass-order/dump_pass_order.py/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/pass-order/README.md/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/quickstart-gemm/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/quickstart-gemm/quickstart_gemm.py/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/quickstart-gemm/README.md/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/README.md/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/repo-map/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/repo-map/inventory.py/https://tilelang-tutorial.huandu.me/code/04-architecture-and-compilation-flow/repo-map/README.md/https://tilelang-tutorial.huandu.me/code/05-comparisons/https://tilelang-tutorial.huandu.me/code/05-comparisons/01-unified-gemm-relu/https://tilelang-tutorial.huandu.me/code/05-comparisons/01-unified-gemm-relu/ascendc_leakyrelu_excerpt.cpp/https://tilelang-tutorial.huandu.me/code/05-comparisons/01-unified-gemm-relu/cuda_gemm_relu_illustrative.cu/https://tilelang-tutorial.huandu.me/code/05-comparisons/01-unified-gemm-relu/cute_dsl_gemv_excerpt.py/https://tilelang-tutorial.huandu.me/code/05-comparisons/01-unified-gemm-relu/pto_dsl_matmul_excerpt.py/https://tilelang-tutorial.huandu.me/code/05-comparisons/01-unified-gemm-relu/README.md/https://tilelang-tutorial.huandu.me/code/05-comparisons/01-unified-gemm-relu/tilelang_gemm_relu.py/https://tilelang-tutorial.huandu.me/code/05-comparisons/01-unified-gemm-relu/triton_gemm_relu_sketch.py/https://tilelang-tutorial.huandu.me/code/05-comparisons/02-abstraction-layers/https://tilelang-tutorial.huandu.me/code/05-comparisons/02-abstraction-layers/abstraction_layers.py/https://tilelang-tutorial.huandu.me/code/05-comparisons/02-abstraction-layers/README.md/https://tilelang-tutorial.huandu.me/code/05-comparisons/README.md/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/01-onchip-footprint/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/01-onchip-footprint/onchip_footprint.py/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/01-onchip-footprint/README.md/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/02-pipeline-timeline/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/02-pipeline-timeline/pipeline_timeline.py/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/02-pipeline-timeline/README.md/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/03-target-sketch/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/03-target-sketch/ascend_950_sketch.py/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/03-target-sketch/nv_hopper_sketch.py/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/03-target-sketch/README.md/https://tilelang-tutorial.huandu.me/code/06-hardware-intro-nv-vs-ascend/README.md/https://tilelang-tutorial.huandu.me/code/07-quickstart/https://tilelang-tutorial.huandu.me/code/07-quickstart/01-first-kernel/https://tilelang-tutorial.huandu.me/code/07-quickstart/01-first-kernel/README.md/https://tilelang-tutorial.huandu.me/code/07-quickstart/01-first-kernel/vecadd_tilelang.py/https://tilelang-tutorial.huandu.me/code/07-quickstart/02-inspect-source/https://tilelang-tutorial.huandu.me/code/07-quickstart/02-inspect-source/inspect_source.py/https://tilelang-tutorial.huandu.me/code/07-quickstart/02-inspect-source/README.md/https://tilelang-tutorial.huandu.me/code/07-quickstart/03-jit-and-cache/https://tilelang-tutorial.huandu.me/code/07-quickstart/03-jit-and-cache/jit_and_cache.py/https://tilelang-tutorial.huandu.me/code/07-quickstart/03-jit-and-cache/README.md/https://tilelang-tutorial.huandu.me/code/07-quickstart/04-api-surface-check/https://tilelang-tutorial.huandu.me/code/07-quickstart/04-api-surface-check/check_api_surface.py/https://tilelang-tutorial.huandu.me/code/07-quickstart/04-api-surface-check/README.md/https://tilelang-tutorial.huandu.me/code/07-quickstart/05-ascend-first-kernel/https://tilelang-tutorial.huandu.me/code/07-quickstart/05-ascend-first-kernel/README.md/https://tilelang-tutorial.huandu.me/code/07-quickstart/05-ascend-first-kernel/vecadd_ascend.py/https://tilelang-tutorial.huandu.me/code/07-quickstart/06-env-check/https://tilelang-tutorial.huandu.me/code/07-quickstart/06-env-check/check_env.py/https://tilelang-tutorial.huandu.me/code/07-quickstart/06-env-check/README.md/https://tilelang-tutorial.huandu.me/code/07-quickstart/README.md/https://tilelang-tutorial.huandu.me/code/08-language-basics/https://tilelang-tutorial.huandu.me/code/08-language-basics/01-kernel-grid/https://tilelang-tutorial.huandu.me/code/08-language-basics/01-kernel-grid/grid_index_model.py/https://tilelang-tutorial.huandu.me/code/08-language-basics/01-kernel-grid/kernel_grid.py/https://tilelang-tutorial.huandu.me/code/08-language-basics/01-kernel-grid/README.md/https://tilelang-tutorial.huandu.me/code/08-language-basics/02-dtypes/https://tilelang-tutorial.huandu.me/code/08-language-basics/02-dtypes/dtype_families.py/https://tilelang-tutorial.huandu.me/code/08-language-basics/02-dtypes/fp8_fp4_exprs.py/https://tilelang-tutorial.huandu.me/code/08-language-basics/02-dtypes/README.md/https://tilelang-tutorial.huandu.me/code/08-language-basics/03-memory-scopes/https://tilelang-tutorial.huandu.me/code/08-language-basics/03-memory-scopes/alloc_scopes.py/https://tilelang-tutorial.huandu.me/code/08-language-basics/03-memory-scopes/README.md/https://tilelang-tutorial.huandu.me/code/08-language-basics/03-memory-scopes/scope_footprint.py/https://tilelang-tutorial.huandu.me/code/08-language-basics/04-loops/https://tilelang-tutorial.huandu.me/code/08-language-basics/04-loops/loop_mapping.py/https://tilelang-tutorial.huandu.me/code/08-language-basics/04-loops/loops_demo.py/https://tilelang-tutorial.huandu.me/code/08-language-basics/04-loops/README.md/https://tilelang-tutorial.huandu.me/code/08-language-basics/05-control-flow/https://tilelang-tutorial.huandu.me/code/08-language-basics/05-control-flow/control_flow_demo.py/https://tilelang-tutorial.huandu.me/code/08-language-basics/05-control-flow/README.md/https://tilelang-tutorial.huandu.me/code/08-language-basics/06-python-boundary/https://tilelang-tutorial.huandu.me/code/08-language-basics/06-python-boundary/py_compat_scan.py/https://tilelang-tutorial.huandu.me/code/08-language-basics/06-python-boundary/README.md/https://tilelang-tutorial.huandu.me/code/08-language-basics/README.md/https://tilelang-tutorial.huandu.me/code/09-core-primitives/https://tilelang-tutorial.huandu.me/code/09-core-primitives/01-primitive-ir-map/https://tilelang-tutorial.huandu.me/code/09-core-primitives/01-primitive-ir-map/primitive_ir_map.py/https://tilelang-tutorial.huandu.me/code/09-core-primitives/01-primitive-ir-map/README.md/https://tilelang-tutorial.huandu.me/code/09-core-primitives/02-copy-predicate-model/https://tilelang-tutorial.huandu.me/code/09-core-primitives/02-copy-predicate-model/copy_predicate_model.py/https://tilelang-tutorial.huandu.me/code/09-core-primitives/02-copy-predicate-model/README.md/https://tilelang-tutorial.huandu.me/code/09-core-primitives/03-thread-reduce-model/https://tilelang-tutorial.huandu.me/code/09-core-primitives/03-thread-reduce-model/README.md/https://tilelang-tutorial.huandu.me/code/09-core-primitives/03-thread-reduce-model/thread_reduce_model.py/https://tilelang-tutorial.huandu.me/code/09-core-primitives/04-copy-gemm-sketches/https://tilelang-tutorial.huandu.me/code/09-core-primitives/04-copy-gemm-sketches/gemm_primitive_variants.py/https://tilelang-tutorial.huandu.me/code/09-core-primitives/04-copy-gemm-sketches/nv_gemm_copy_sketch.py/https://tilelang-tutorial.huandu.me/code/09-core-primitives/04-copy-gemm-sketches/README.md/https://tilelang-tutorial.huandu.me/code/09-core-primitives/05-ascend-primitives/https://tilelang-tutorial.huandu.me/code/09-core-primitives/05-ascend-primitives/ascend_dual_copy_simtvf_sketch.py/https://tilelang-tutorial.huandu.me/code/09-core-primitives/05-ascend-primitives/ascend_tile_library_sketch.py/https://tilelang-tutorial.huandu.me/code/09-core-primitives/05-ascend-primitives/README.md/https://tilelang-tutorial.huandu.me/code/09-core-primitives/README.md/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/01-layout-algebra/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/01-layout-algebra/layout_algebra.py/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/01-layout-algebra/README.md/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/02-plot-layout/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/02-plot-layout/plot_layout_demo.py/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/02-plot-layout/README.md/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/03-annotate-layout/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/03-annotate-layout/annotate_swizzle_gemm.py/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/03-annotate-layout/README.md/https://tilelang-tutorial.huandu.me/code/10-layout-and-memory-intro/README.md/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/01-pipeline-timeline/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/01-pipeline-timeline/pipeline_timeline.py/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/01-pipeline-timeline/README.md/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/02-shared-memory-budget/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/02-shared-memory-budget/README.md/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/02-shared-memory-budget/smem_budget.py/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/03-autotune-config-space/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/03-autotune-config-space/config_space.py/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/03-autotune-config-space/README.md/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/04-tilelang-pipelined-gemm/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/04-tilelang-pipelined-gemm/pipelined_gemm.py/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/04-tilelang-pipelined-gemm/README.md/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/05-tilelang-autotune-gemm/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/05-tilelang-autotune-gemm/autotune_gemm.py/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/05-tilelang-autotune-gemm/README.md/https://tilelang-tutorial.huandu.me/code/11-pipeline-and-autotuning/README.md/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/01-tool-inventory/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/01-tool-inventory/README.md/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/01-tool-inventory/tool_inventory.py/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/02-tolerance-model/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/02-tolerance-model/README.md/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/02-tolerance-model/tolerance_model.py/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/03-compile-only/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/03-compile-only/add_kernel_compile_only.py/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/03-compile-only/README.md/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/04-lower-trace/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/04-lower-trace/gemm_lower_trace.py/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/04-lower-trace/README.md/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/05-print-and-assert/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/05-print-and-assert/debug_print_demo.py/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/05-print-and-assert/README.md/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/06-ascend-dump/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/06-ascend-dump/ascend_dump_demo.py/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/06-ascend-dump/README.md/https://tilelang-tutorial.huandu.me/code/12-debugging-and-profiling/README.md/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/01-ledger/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/01-ledger/ledger.py/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/01-ledger/README.md/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/02-elementwise/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/02-elementwise/elementwise_variants.py/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/02-elementwise/README.md/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/03-norm/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/03-norm/norm_kernels.py/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/03-norm/README.md/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/04-rope/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/04-rope/README.md/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/04-rope/rope_kernels.py/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/05-fusion/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/05-fusion/README.md/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/05-fusion/rmsnorm_rope_quant.py/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/06-ascend/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/06-ascend/ascend_adapter_sketch.py/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/06-ascend/ascend_mainline_kernels.py/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/06-ascend/README.md/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/07-api-audit/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/07-api-audit/api_audit.py/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/07-api-audit/README.md/https://tilelang-tutorial.huandu.me/code/13-operators-elementwise-norm-rope/README.md/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/01-gemm-ledger/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/01-gemm-ledger/gemm_ledger.py/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/01-gemm-ledger/README.md/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/02-tilelang-ladder/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/02-tilelang-ladder/gemm_intrinsic.py/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/02-tilelang-ladder/gemm_ladder.py/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/02-tilelang-ladder/README.md/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/03-gemm-variants/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/03-gemm-variants/gemm_variants.py/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/03-gemm-variants/README.md/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/04-low-precision/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/04-low-precision/low_precision_gemm.py/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/04-low-precision/README.md/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/05-ascend-gemm/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/05-ascend-gemm/ascend_gemm.py/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/05-ascend-gemm/README.md/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/06-bench-methodology/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/06-bench-methodology/bench_methodology.py/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/06-bench-methodology/README.md/https://tilelang-tutorial.huandu.me/code/14-operators-gemm/README.md/https://tilelang-tutorial.huandu.me/code/15-operators-attention/https://tilelang-tutorial.huandu.me/code/15-operators-attention/01-traffic-model/https://tilelang-tutorial.huandu.me/code/15-operators-attention/01-traffic-model/README.md/https://tilelang-tutorial.huandu.me/code/15-operators-attention/01-traffic-model/traffic_model.py/https://tilelang-tutorial.huandu.me/code/15-operators-attention/02-online-softmax/https://tilelang-tutorial.huandu.me/code/15-operators-attention/02-online-softmax/online_softmax.py/https://tilelang-tutorial.huandu.me/code/15-operators-attention/02-online-softmax/README.md/https://tilelang-tutorial.huandu.me/code/15-operators-attention/03-split-kv-combine/https://tilelang-tutorial.huandu.me/code/15-operators-attention/03-split-kv-combine/README.md/https://tilelang-tutorial.huandu.me/code/15-operators-attention/03-split-kv-combine/split_kv_combine.py/https://tilelang-tutorial.huandu.me/code/15-operators-attention/04-block-flash-attention/https://tilelang-tutorial.huandu.me/code/15-operators-attention/04-block-flash-attention/block_flash_attention.py/https://tilelang-tutorial.huandu.me/code/15-operators-attention/04-block-flash-attention/README.md/https://tilelang-tutorial.huandu.me/code/15-operators-attention/05-official-example-map/https://tilelang-tutorial.huandu.me/code/15-operators-attention/05-official-example-map/map_official_examples.py/https://tilelang-tutorial.huandu.me/code/15-operators-attention/05-official-example-map/README.md/https://tilelang-tutorial.huandu.me/code/15-operators-attention/06-ascend-map/https://tilelang-tutorial.huandu.me/code/15-operators-attention/06-ascend-map/ascend_attention_map.py/https://tilelang-tutorial.huandu.me/code/15-operators-attention/06-ascend-map/README.md/https://tilelang-tutorial.huandu.me/code/15-operators-attention/README.md/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/01-cost-ledger/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/01-cost-ledger/decode_step_ledger.py/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/01-cost-ledger/README.md/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/02-moe-routing/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/02-moe-routing/README.md/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/02-moe-routing/routing_sim.py/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/03-mla-kv/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/03-mla-kv/kv_budget.py/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/03-mla-kv/README.md/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/04-moe-kernel-map/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/04-moe-kernel-map/kernel_map.py/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/04-moe-kernel-map/README.md/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/05-sparse-index-map/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/05-sparse-index-map/README.md/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/05-sparse-index-map/sparse_index_map.py/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/06-ascend-map/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/06-ascend-map/ascend_map.py/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/06-ascend-map/README.md/https://tilelang-tutorial.huandu.me/code/16-operators-moe-mla/README.md/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/01-plan-ledger/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/01-plan-ledger/README.md/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/01-plan-ledger/train_plan.py/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/02-data-pipeline/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/02-data-pipeline/pack_tokens.py/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/02-data-pipeline/README.md/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/03-tilelang-norm-torch/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/03-tilelang-norm-torch/README.md/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/03-tilelang-norm-torch/rmsnorm_tl.py/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/04-verify-numerics/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/04-verify-numerics/compare_ref.py/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/04-verify-numerics/README.md/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/05-ascend-integration/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/05-ascend-integration/invoke_ascend_op.py/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/05-ascend-integration/README.md/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/06-train-loop/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/06-train-loop/README.md/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/06-train-loop/train_step.py/https://tilelang-tutorial.huandu.me/code/17-e2e-training-7b/README.md/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/01-prefill-decode-roofline/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/01-prefill-decode-roofline/prefill_decode_roofline.py/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/01-prefill-decode-roofline/README.md/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/02-kv-ledger/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/02-kv-ledger/kv_ledger.py/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/02-kv-ledger/README.md/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/03-decode-step-share/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/03-decode-step-share/decode_step_share.py/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/03-decode-step-share/README.md/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/04-serving-schedule-sim/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/04-serving-schedule-sim/README.md/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/04-serving-schedule-sim/serving_schedule_sim.py/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/05-bench-methodology/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/05-bench-methodology/bench_methodology.py/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/05-bench-methodology/README.md/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/06-tilelang-decode-attention/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/06-tilelang-decode-attention/decode_attention_splitkv.py/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/06-tilelang-decode-attention/README.md/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/07-tilelang-moe-and-quant/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/07-tilelang-moe-and-quant/moe_dispatch_contract.py/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/07-tilelang-moe-and-quant/README.md/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/07-tilelang-moe-and-quant/skeletons.py/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/08-ascend-aclgraph/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/08-ascend-aclgraph/aclgraph_decode_block.py/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/08-ascend-aclgraph/README.md/https://tilelang-tutorial.huandu.me/code/18-e2e-inference-serving/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/01-env-check/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/01-env-check/env_check.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/01-env-check/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/02-elementwise-ub/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/02-elementwise-ub/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/02-elementwise-ub/vec_add_adapter.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/03-elementwise-pipeline/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/03-elementwise-pipeline/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/03-elementwise-pipeline/vec_add_pipeline.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/04-gemm-l1-l0c/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/04-gemm-l1-l0c/gemm_adapter.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/04-gemm-l1-l0c/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/05-gemm-dialect/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/05-gemm-dialect/gemm_relu_dialect.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/05-gemm-dialect/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/06-ascendc-baseline/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/06-ascendc-baseline/add_ascendc.cpp/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/06-ascendc-baseline/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/07-vector-modes/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/07-vector-modes/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/07-vector-modes/vecadd_modes.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/08-msprof-bench/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/08-msprof-bench/bench_msprof.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/08-msprof-bench/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/09-aclgraph/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/09-aclgraph/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/09-aclgraph/rms_rope_aclgraph.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/10-debug-and-dump/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/10-debug-and-dump/print_and_dump.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/10-debug-and-dump/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/11-exception-dump/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/11-exception-dump/exception_dump.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/11-exception-dump/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/12-tuning-sweep/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/12-tuning-sweep/README.md/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/12-tuning-sweep/tune_sweep.py/https://tilelang-tutorial.huandu.me/code/19-ascend-hands-on/README.md/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/01-pass-order/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/01-pass-order/dump_pass_order.py/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/01-pass-order/README.md/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/02-ir-snippet-extract/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/02-ir-snippet-extract/extract_ir_snippets.py/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/02-ir-snippet-extract/README.md/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/03-lower-trace-walkthrough/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/03-lower-trace-walkthrough/lower_trace_walkthrough.py/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/03-lower-trace-walkthrough/README.md/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/04-version-diff/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/04-version-diff/compare_ascend_pipelines.py/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/04-version-diff/README.md/https://tilelang-tutorial.huandu.me/code/20-ir-and-pass-pipeline/README.md/https://tilelang-tutorial.huandu.me/code/21-layout-inference/https://tilelang-tutorial.huandu.me/code/21-layout-inference/01-layout-algebra/https://tilelang-tutorial.huandu.me/code/21-layout-inference/01-layout-algebra/layout_algebra.py/https://tilelang-tutorial.huandu.me/code/21-layout-inference/01-layout-algebra/README.md/https://tilelang-tutorial.huandu.me/code/21-layout-inference/02-plot-layout/https://tilelang-tutorial.huandu.me/code/21-layout-inference/02-plot-layout/plot_layout_demo.py/https://tilelang-tutorial.huandu.me/code/21-layout-inference/02-plot-layout/README.md/https://tilelang-tutorial.huandu.me/code/21-layout-inference/03-infer-visualization/https://tilelang-tutorial.huandu.me/code/21-layout-inference/03-infer-visualization/gemm_layout_visualization.py/https://tilelang-tutorial.huandu.me/code/21-layout-inference/03-infer-visualization/README.md/https://tilelang-tutorial.huandu.me/code/21-layout-inference/04-layout-conflict/https://tilelang-tutorial.huandu.me/code/21-layout-inference/04-layout-conflict/annotate_layout_conflict.py/https://tilelang-tutorial.huandu.me/code/21-layout-inference/04-layout-conflict/README.md/https://tilelang-tutorial.huandu.me/code/21-layout-inference/05-ascend-nz-layout/https://tilelang-tutorial.huandu.me/code/21-layout-inference/05-ascend-nz-layout/ascend_nz_layout.py/https://tilelang-tutorial.huandu.me/code/21-layout-inference/05-ascend-nz-layout/README.md/https://tilelang-tutorial.huandu.me/code/21-layout-inference/README.md/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/01-num-stages-ir-diff/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/01-num-stages-ir-diff/lower_num_stages.py/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/01-num-stages-ir-diff/README.md/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/02-buffer-version-model/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/02-buffer-version-model/buffer_versions.py/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/02-buffer-version-model/README.md/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/03-sync-count-vs-perf/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/03-sync-count-vs-perf/README.md/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/03-sync-count-vs-perf/sync_count_model.py/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/04-ascend-flag-budget/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/04-ascend-flag-budget/flag_budget.py/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/04-ascend-flag-budget/README.md/https://tilelang-tutorial.huandu.me/code/22-pipeline-sync-multibuffer/README.md/https://tilelang-tutorial.huandu.me/code/23-autoschedule-smt/https://tilelang-tutorial.huandu.me/code/23-autoschedule-smt/pass_order/https://tilelang-tutorial.huandu.me/code/23-autoschedule-smt/pass_order/ascend_pass_order.py/https://tilelang-tutorial.huandu.me/code/23-autoschedule-smt/pass_order/README.md/https://tilelang-tutorial.huandu.me/code/23-autoschedule-smt/README.md/https://tilelang-tutorial.huandu.me/code/23-autoschedule-smt/toy_scheduler/https://tilelang-tutorial.huandu.me/code/23-autoschedule-smt/toy_scheduler/README.md/https://tilelang-tutorial.huandu.me/code/23-autoschedule-smt/toy_scheduler/toy_schedule.py/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/01-copy-inst-selection/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/01-copy-inst-selection/copy_inst_selection.py/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/01-copy-inst-selection/README.md/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/02-gemm-inst-selection/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/02-gemm-inst-selection/gemm_inst_selection.py/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/02-gemm-inst-selection/README.md/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/03-tma-descriptor/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/03-tma-descriptor/README.md/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/03-tma-descriptor/tma_descriptor.py/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/04-warp-specialize-gemm/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/04-warp-specialize-gemm/README.md/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/04-warp-specialize-gemm/ws_gemm.py/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/05-tcgen05-gemm/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/05-tcgen05-gemm/README.md/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/05-tcgen05-gemm/tcgen05_gemm.py/https://tilelang-tutorial.huandu.me/code/24-nvidia-backend-codegen/README.md/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/01-route-map/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/01-route-map/README.md/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/01-route-map/route_map.py/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/02-scope-dma-pipes/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/02-scope-dma-pipes/README.md/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/02-scope-dma-pipes/scope_dma_pipes.py/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/03-api-presence-matrix/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/03-api-presence-matrix/api_matrix.py/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/03-api-presence-matrix/README.md/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/04-cite-check/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/04-cite-check/check_citations.py/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/04-cite-check/README.md/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/05-emit-ascend-c/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/05-emit-ascend-c/print_ascend_c.py/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/05-emit-ascend-c/README.md/https://tilelang-tutorial.huandu.me/code/25-ascend-backend-codegen/README.md/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/01-runtime-anchor-check/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/01-runtime-anchor-check/README.md/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/01-runtime-anchor-check/verify_runtime_anchors.py/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/02-cache-dir-inspect/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/02-cache-dir-inspect/inspect_cache_dir.py/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/02-cache-dir-inspect/README.md/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/03-packed-abi-sim/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/03-packed-abi-sim/packed_abi_sim.py/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/03-packed-abi-sim/README.md/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/04-torch-stream-call/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/04-torch-stream-call/README.md/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/04-torch-stream-call/torch_stream_call.py/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/05-compile-only-inspect/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/05-compile-only-inspect/inspect_via_compile_only.py/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/05-compile-only-inspect/README.md/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/06-ascend-aclgraph/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/06-ascend-aclgraph/aclgraph_capture.py/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/06-ascend-aclgraph/README.md/https://tilelang-tutorial.huandu.me/code/26-runtime-jit-abi/README.md/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/01-number-formats/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/01-number-formats/number_formats.py/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/01-number-formats/README.md/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/02-quant-dequant/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/02-quant-dequant/quant_dequant.py/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/02-quant-dequant/README.md/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/03-error-model/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/03-error-model/error_model.py/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/03-error-model/README.md/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/04-vs-reference/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/04-vs-reference/README.md/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/04-vs-reference/reference_matching.py/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/05-mx-scale-layout/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/05-mx-scale-layout/mx_scale_layout.py/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/05-mx-scale-layout/README.md/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/06-tilelang-audit/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/06-tilelang-audit/README.md/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/06-tilelang-audit/tilelang_audit.py/https://tilelang-tutorial.huandu.me/code/27-low-precision-quantization/README.md/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/01-op-ledger/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/01-op-ledger/op_ledger.py/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/01-op-ledger/README.md/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/02-roofline-misuse/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/02-roofline-misuse/README.md/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/02-roofline-misuse/roofline_misuse.py/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/03-carver-cost-model/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/03-carver-cost-model/carver_cost_model.py/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/03-carver-cost-model/README.md/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/04-tile-size-decision/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/04-tile-size-decision/README.md/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/04-tile-size-decision/tile_decision.py/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/05-fusion-split/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/05-fusion-split/fusion_split.py/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/05-fusion-split/README.md/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/06-case-norm-rope-quant/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/06-case-norm-rope-quant/case_norm_rope_quant.py/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/06-case-norm-rope-quant/README.md/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/07-ascend-latency-model/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/07-ascend-latency-model/ascend_pipe_model.py/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/07-ascend-latency-model/README.md/https://tilelang-tutorial.huandu.me/code/28-operator-design-methodology/README.md/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/01-python-pass/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/01-python-pass/hoist_constant_broadcast.py/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/01-python-pass/README.md/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/01-python-pass/test_hoist_constant_broadcast.py/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/02-cpp-pass-template/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/02-cpp-pass-template/count_parallel_loops.cc/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/02-cpp-pass-template/python_wrapper_snippet.py/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/02-cpp-pass-template/README.md/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/02-cpp-pass-template/test_count_parallel_loops.py/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/03-verify-pass-template/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/03-verify-pass-template/extract_verify_skeleton.py/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/03-verify-pass-template/README.md/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/03-verify-pass-template/test_verify_local_alloc_limit.py/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/03-verify-pass-template/verify_local_alloc_limit.cc/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/04-regression-judge/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/04-regression-judge/judge_regression.py/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/04-regression-judge/README.md/https://tilelang-tutorial.huandu.me/code/29-compiler-engineering/README.md/https://tilelang-tutorial.huandu.me/code/30-frontier/https://tilelang-tutorial.huandu.me/code/30-frontier/01-state-audit/https://tilelang-tutorial.huandu.me/code/30-frontier/01-state-audit/README.md/https://tilelang-tutorial.huandu.me/code/30-frontier/01-state-audit/state_audit.py/https://tilelang-tutorial.huandu.me/code/30-frontier/02-tree-diff/https://tilelang-tutorial.huandu.me/code/30-frontier/02-tree-diff/README.md/https://tilelang-tutorial.huandu.me/code/30-frontier/02-tree-diff/tree_diff.py/https://tilelang-tutorial.huandu.me/code/30-frontier/03-upstream-watch/https://tilelang-tutorial.huandu.me/code/30-frontier/03-upstream-watch/README.md/https://tilelang-tutorial.huandu.me/code/30-frontier/03-upstream-watch/upstream_watch.py/https://tilelang-tutorial.huandu.me/code/30-frontier/README.md/https://tilelang-tutorial.huandu.me/code/31-contributing/https://tilelang-tutorial.huandu.me/code/31-contributing/01-pr-scope-linter/https://tilelang-tutorial.huandu.me/code/31-contributing/01-pr-scope-linter/pr_scope_linter.py/https://tilelang-tutorial.huandu.me/code/31-contributing/01-pr-scope-linter/README.md/https://tilelang-tutorial.huandu.me/code/31-contributing/02-rfc-lint/https://tilelang-tutorial.huandu.me/code/31-contributing/02-rfc-lint/README.md/https://tilelang-tutorial.huandu.me/code/31-contributing/02-rfc-lint/rfc_lint.py/https://tilelang-tutorial.huandu.me/code/31-contributing/03-checklist-gate/https://tilelang-tutorial.huandu.me/code/31-contributing/03-checklist-gate/checklist_gate.py/https://tilelang-tutorial.huandu.me/code/31-contributing/03-checklist-gate/README.md/https://tilelang-tutorial.huandu.me/code/31-contributing/README.md/https://tilelang-tutorial.huandu.me/code/32-deepseek-ascend-release/https://tilelang-tutorial.huandu.me/code/32-deepseek-ascend-release/01-verify-checklist/https://tilelang-tutorial.huandu.me/code/32-deepseek-ascend-release/01-verify-checklist/extract_sections.py/https://tilelang-tutorial.huandu.me/code/32-deepseek-ascend-release/01-verify-checklist/README.md/https://tilelang-tutorial.huandu.me/code/32-deepseek-ascend-release/README.md/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/01-topk-gate-minimal/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/01-topk-gate-minimal/README.md/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/01-topk-gate-minimal/topk_gate_asc_min.py/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/01-topk-gate-minimal/topk_gate_cuda_min.py/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/02-call-the-library/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/02-call-the-library/call_quant_and_rope.py/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/02-call-the-library/call_topk_gate.py/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/02-call-the-library/README.md/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/03-map-and-diff/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/03-map-and-diff/map_and_diff.py/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/03-map-and-diff/README.md/https://tilelang-tutorial.huandu.me/code/33-tilekernels-two-backends/README.md/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/01-engram-gate-shapes/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/01-engram-gate-shapes/engram_gate_shapes.py/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/01-engram-gate-shapes/README.md/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/02-sinkhorn-iterations/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/02-sinkhorn-iterations/README.md/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/02-sinkhorn-iterations/sinkhorn_iterations.py/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/03-mhc-mix-flow/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/03-mhc-mix-flow/mhc_mix_flow.py/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/03-mhc-mix-flow/README.md/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/04-engram-hash-lookup/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/04-engram-hash-lookup/engram_hash_lookup.py/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/04-engram-hash-lookup/README.md/https://tilelang-tutorial.huandu.me/code/34-tilekernels-engram-mhc/README.md/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/01-python-api/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/01-python-api/deepgemm_ascend_call.py/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/01-python-api/README.md/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/02-perf-ledger/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/02-perf-ledger/perf_ledger.py/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/02-perf-ledger/README.md/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/03-config-search/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/03-config-search/config_search.py/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/03-config-search/README.md/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/04-hc-prenorm-tilelang/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/04-hc-prenorm-tilelang/hc_prenorm_sketch.py/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/04-hc-prenorm-tilelang/README.md/https://tilelang-tutorial.huandu.me/code/35-deepgemm-ascend/README.md/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/01-head-roofline/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/01-head-roofline/head_roofline.py/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/01-head-roofline/README.md/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/02-fixpipe-bound/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/02-fixpipe-bound/fixpipe_bound.py/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/02-fixpipe-bound/README.md/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/03-online-softmax-skip-scale/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/03-online-softmax-skip-scale/online_softmax_skip_scale.py/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/03-online-softmax-skip-scale/README.md/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/04-tilelang-expressibility/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/04-tilelang-expressibility/expressibility_map.py/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/04-tilelang-expressibility/README.md/https://tilelang-tutorial.huandu.me/code/36-flashmla-ascend/README.md/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/01-deepselect-python/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/01-deepselect-python/deepselect_python.py/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/01-deepselect-python/README.md/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/02-block-order-experiment/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/02-block-order-experiment/block_order_experiment.py/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/02-block-order-experiment/README.md/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/03-complexity-tables/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/03-complexity-tables/complexity_tables.py/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/03-complexity-tables/README.md/https://tilelang-tutorial.huandu.me/code/37-deepselect-topk/README.md/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/01-ep-minimal/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/01-ep-minimal/ep_minimal.py/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/01-ep-minimal/README.md/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/02-comm-ledger/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/02-comm-ledger/comm_ledger.py/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/02-comm-ledger/README.md/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/03-buffer-size/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/03-buffer-size/buffer_size.py/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/03-buffer-size/README.md/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/04-bandwidth-normalize/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/04-bandwidth-normalize/bandwidth_normalize.py/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/04-bandwidth-normalize/README.md/https://tilelang-tutorial.huandu.me/code/38-deepep-ascend/README.md/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/01-train-step-budget/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/01-train-step-budget/README.md/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/01-train-step-budget/train_step_budget.py/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/02-inference-latency-model/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/02-inference-latency-model/inference_latency_model.py/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/02-inference-latency-model/README.md/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/03-bottleneck-map/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/03-bottleneck-map/bottleneck_map.py/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/03-bottleneck-map/README.md/https://tilelang-tutorial.huandu.me/code/39-deepseek-e2e-supernode/README.md/https://tilelang-tutorial.huandu.me/notes/https://tilelang-tutorial.huandu.me/notes/ascend-route/https://tilelang-tutorial.huandu.me/notes/compiler-internals-map/https://tilelang-tutorial.huandu.me/notes/knowledge-system/https://tilelang-tutorial.huandu.me/references/https://tilelang-tutorial.huandu.me/versions/