b12x 0.4.0

Unapologetically SM120-only CuTe DSL kernels for NVFP4 GEMM and MoE.

b12x

b12x is an SM120-only CuTe DSL kernel library for Blackwell NVFP4 dense GEMM and routed Mixture-of-Experts inference.

It is intentionally narrow. This is not a generic CUDA kernel collection or a full model-serving stack. It does not intend to target any other GPU architectures, including SM100. It is a focused package for a small number of hand-tuned, high-performance SM120 kernels plus the runtime glue needed to launch them cleanly from PyTorch and sglang.

Installation

Runtime install

python -m pip install b12x

Development install from source

git clone <repo-url>
cd b12x
python -m pip install -e '.[dev]'

Requirements

• Blackwell SM120 GPU
• CUDA 13 toolchain
• Python >=3.10,<4.0
• CUDA 13 PyTorch, torch>=2.10.0
• nvidia-cutlass-dsl[cu13]==4.4.1
• FlashInfer available if you want reference and benchmark comparisons, but it's not a runtime dependency
• Qwen3.5-397B A17B NVFP4 checkpoint available through B12X_MODEL_PATH for the end-to-end MoE benchmark

Package layout

• b12x.cute
  • Low-level CuTe and FP4 helpers
• b12x.gemm
  • Standalone dense NVFP4 GEMM
• b12x.integration
  • Public runtime entrypoints such as b12x_moe_fp4
• b12x.moe.fused
  • Static and dynamic fused MoE kernels and reference paths
• b12x.quant
  • Torch-side NVFP4 packing and quantization helpers
• b12x.sglang
  • Thin sglang integration shims

MoE runtime contract

• b12x.integration.tp_moe.b12x_moe_fp4 requires a caller-owned workspace.
• b12x selects its fused MoE backend from shape alone:
  • compact routed workloads use the static backend
  • all larger routed workloads use dynamic
• Use allocate_tp_moe_workspace(...) for one exact unchunked launch shape.
• Use allocate_tp_moe_workspace_pool() for variable-size or chunked workloads.
• Keep one workspace pool per process/device, and let the pool partition scratch by CUDA stream internally.
• During CUDA graph capture, output= must also be caller-owned and stable across replays.

Benchmarks and tests

Benchmarks

• benchmarks/benchmark_moe.py
  • End-to-end Qwen3.5-397B TP=4 MoE benchmark
  • micro batch profile: [1, 2, 4, 8]
  • sglang-single-request batch profile: [1, 23, 80]
  • chunked-prefill batch profile: [8192, 16384, 24576, 32768]
• benchmarks/benchmark_dense_gemm.py
  • Dense FP4 GEMM vs FlashInfer/cuDNN/CUTLASS

Tests

• tests/test_tp_moe_reference.py
  • Independent oracle-backed MoE correctness test
• tests/test_moe_equivalence.py
  • Real-weight smoke and CUDA-graph replay routing-safety checks
• tests/test_gemm_stack.py
  • Dense GEMM exactness vs FlashInfer/cuDNN

Common commands

# Graph-first benchmark defaults with auto-dispatch
B12X_MODEL_PATH=/path/to/Qwen3.5-397B-A17B-NVFP4 python benchmarks/benchmark_moe.py

# Measure eager launches instead of CUDA graph replay
B12X_MODEL_PATH=/path/to/Qwen3.5-397B-A17B-NVFP4 python benchmarks/benchmark_moe.py --no-cuda-graph

# Include routing in the timed region
B12X_MODEL_PATH=/path/to/Qwen3.5-397B-A17B-NVFP4 python benchmarks/benchmark_moe.py --include-routing

# Use the recorded single-request sglang profile
B12X_MODEL_PATH=/path/to/Qwen3.5-397B-A17B-NVFP4 python benchmarks/benchmark_moe.py --batch-size-profile sglang-single-request

# Graph-first prefill-scale sweep aligned with chunked-prefill serving
B12X_MODEL_PATH=/path/to/Qwen3.5-397B-A17B-NVFP4 python benchmarks/benchmark_moe.py --batch-size-profile chunked-prefill

# Multi-layer CUDA-graph replay validation with real consecutive MoE layers
B12X_MODEL_PATH=/path/to/Qwen3.5-397B-A17B-NVFP4 python benchmarks/benchmark_moe.py --graph-mode multi-layer --reference none --validate none

# Dense GEMM microbenchmark
python benchmarks/benchmark_dense_gemm.py

# Oracle-backed MoE correctness
python tests/test_tp_moe_reference.py --impls b12x --scale-contract per-expert

# Real-weight CUDA-graph smoke
pytest tests/test_moe_equivalence.py