---

CubeK: high-performance multi-platform kernels in CubeCL

Algorithms

Algorithms	Variants
Random	bernoulli normal uniform
Quantization	symmetric per-block per-tensor q2 q4 q8 fp4
Reduction	mean sum prod max min arg[max|min] per-cube per-plane
Matmul	mma unit tma multi-stage specialization ordered multi-rows
Convolution	mma unit tma multi-stage im2col
Attention	mma unit multi-rows

Contributing

If you want to contribute new kernels, please read the GUIDE.md.

Running tests

Note: This applies to most kernels, but reduce works slightly differently for now, see its README.

Command

Three test suites are available:

• Smoke test suite: a tractable subset of representative tests that run on the CI.
• Extended test suite: usually auto-generated combinatorial tests covering many configurations. Good to run when developing kernels. Normally kept tractable.
• Full test suite: all generable test combinations; may be too large to compile or run practically.

Run tests with

# Replace <runtime> with cpu, cuda, rocm, wgpu, vulkan or metal

# Smoke test suite
cargo test-<runtime>

# Extended test suite
cargo test-<runtime>-extended

# Full test suite
cargo test-<runtime>-full

Cube test mode

You can control test behavior by setting the CUBE_TEST_MODE environment variable.
For more details, see Test Mode.

Modes

• CUBE_TEST_MODE=correct (default)
  Tests pass if results are numerically correct or if the kernel was launched with an invalid configuration.

  • Useful when tests are auto-generated from multiple parameter combinations, where some invalid configurations are expected.
  • Failing tests display only the first index with a discrepancy.

• CUBE_TEST_MODE=strict
  Tests pass only if they compile, run, and produce numerically accurate results.

  • Ideal for debugging to avoid false positives that can occur in correct mode.

• CUBE_TEST_MODE=printfail
  Similar to correct mode: tests pass if results are correct or if the kernel is invalid.

  • Failing tests show all tensor discrepancies.
  • Supports filtering, e.g.: CUBE_TEST_MODE=printfail:0,.,10-20 shows elements from the 0th first dimension, all of the second, and elements 10–20 in the third.

• CUBE_TEST_MODE=printall
  All tests fail, displaying all tensor discrepancies.

  • Filtering works the same as in printfail.

• CUBE_TEST_MODE=failifrun
  Only tests that compile and run will fail; others succeed.

  • Useful for tracking critical tests in large suites.