George HotzandGitHub f267a49639 UOp.param accepts single sint as shape (int,) (#17784) ...

* UOp.param accepts single sint as shape (int,)

Change UOp.param signature from shape:tuple[sint, ...]|None to shape:tuple[sint, ...]|sint|None.
A single sint means 1D shape (sint,). Update all callers from (n,) to n syntax.
No param_from_shape — multi-dim shapes stay as tuples.

* use single int syntax in hcq2 copy_with_kernel

2026-08-27 14:07:33 -07:00

..

mlperf_bert

deprecate Tensor.training [PR] (#16774)

2026-06-27 11:02:28 -04:00

mlperf_resnet

remove tfa dependency: use keras.optimizers.Lamb and tf.raw_ops for LARS (#13555)

2025-12-03 17:48:27 -05:00

mlperf_retinanet

RetinaNet losses (#9536)

2025-03-21 15:52:54 -04:00

mlperf_stable_diffusion

rename MULTI to UNSHARD (#17267)

2026-07-28 16:51:41 -07:00

mlperf_unet3d

[MLPerf] UNet3D dataloader (#4343)

2024-04-28 22:34:18 -04:00

openpilot

fix failed threefry (#10646)

2025-06-05 17:17:42 -07:00

process_replay

remove Ops.DEVICE from Ops.PROGRAM [PR] (#16797)

2026-06-29 17:41:31 -04:00

sglang_llama

refactor test/external/external_llama_eval.py (#10567)

2025-06-23 10:43:20 -07:00

external_benchmark_all2all.py

add external_benchmark_all2all.py (#17507)

2026-08-13 15:50:43 +09:00

external_benchmark_am.py

clean: becnhmark -> benchmark (#10620)

2025-06-03 19:28:18 -07:00

external_benchmark_bert_matmuls.py

DEFAULT_FLOAT/DEFAULT_INT ContextVar [pr] (#17265)

2026-07-28 19:08:18 -04:00

external_benchmark_bert_softmax.py

DEFAULT_FLOAT/DEFAULT_INT ContextVar [pr] (#17265)

2026-07-28 19:08:18 -04:00

external_benchmark_disk_raw.py

ci: add h machines (#11416)

2025-07-29 19:21:51 +03:00

external_benchmark_hip_compile.py

ops_gpu -> ops_cl (#12103)

2025-09-10 15:15:48 -04:00

external_benchmark_keccak.py

Tensor.keccak("sha3_256") (#7186)

2025-06-06 15:24:05 -07:00

external_benchmark_kernel_launch.py

cleanup tests, bump caches (#11746)

2025-08-19 21:21:07 -07:00

external_benchmark_llama_offload.py

llama: benchmark offload copy script (#16762)

2026-06-26 00:26:32 -07:00

external_benchmark_llama_schedule.py

more Tensor.training -> TRAINING [PR] (#16770)

2026-06-26 20:11:59 -04:00

external_benchmark_multitensor_allreduce.py

all2all (#13902)

2025-12-31 16:38:32 +03:00

external_benchmark_op_conv.py

UOp.param accepts single sint as shape (int,) (#17784)

2026-08-27 14:07:33 -07:00

external_benchmark_pyrender.py

rename BUFFERIZE to STAGE (#16125)

2026-05-10 09:26:46 -07:00

external_benchmark_resnet.py

more Tensor.training -> TRAINING [PR] (#16770)

2026-06-26 20:11:59 -04:00

external_benchmark_schedule.py

finish casted_consts migration [PR] (#17595)

2026-08-19 10:43:55 -04:00

external_cl_half_max.py

ops_gpu -> ops_cl (#12103)

2025-09-10 15:15:48 -04:00

external_fuzz_ampt.py

Fix/hcqfuzz harnesss bug (#10923)

2025-06-23 11:22:30 +03:00

external_fuzz_beam_timeout_recovery.py

am: use per pipe reset for gfx11+ (#16006)

2026-05-01 12:56:43 +03:00

external_fuzz_hcq_mp.py

system: fix flock on pcidevs (#13123)

2025-11-06 19:02:13 +08:00

external_fuzz_hcq_signals.py

hcqfuzz: init (#10049)

2025-04-25 23:19:21 +03:00

external_fuzz_qcom_cpu_cache.py

qcom: sync cpu cache when from_blob (#14518)

2026-02-03 21:51:03 +03:00

external_fuzz_sdma_warm_start.py

am: reset sdma to start from 0 (#15109)

2026-03-03 18:14:46 +03:00

external_fuzz_tlsf.py

Fix/hcqfuzz harnesss bug (#10923)

2025-06-23 11:22:30 +03:00

external_gpu_fail_osx.py

ops_gpu -> ops_cl (#12103)

2025-09-10 15:15:48 -04:00

external_jit_failure.py

fix jit realize issue (#3258)

2024-01-26 18:27:35 -08:00

external_llm_eval.py

llm is the only app (#15779)

2026-04-17 10:44:48 +08:00

external_metal_compile_fail.py

Use MTLCompiler directly (v2) (#7920)

2024-12-04 16:36:48 +08:00

external_model_benchmark.py

DEV is ContextVar, setting Device.DEFAULT is deprecated (#15508)

2026-03-30 17:10:49 -04:00

external_multi_gpu.py

deprecate <dev>=1 in favor of DEV=<dev> (#15467)

2026-03-26 03:48:03 -04:00

external_nhwc_conv.py

add fun with nhwc convs

2025-10-28 17:12:22 +08:00

external_openpilot_image_warp.py

openpilot warp (#13283)

2025-11-14 13:55:32 -08:00

external_osx_profiling.py

ops_gpu -> ops_cl (#12103)

2025-09-10 15:15:48 -04:00

external_test_am_fault_recovery.py

standardize Program class (#17189)

2026-07-25 00:17:38 -04:00

external_test_am.py

am: large allocs aligned to 2mb to use 2mb pages (#15609)

2026-04-05 18:01:31 +03:00

external_test_amd.py

move schedule into schedule (#15736)

2026-04-15 11:03:25 +08:00

external_test_datasets.py

skip test TestKiTS19Dataset::test_training_set (#10936)

2025-06-23 14:27:24 -04:00

external_test_dev_var.py

deprecate <dev>=1 in favor of DEV=<dev> (#15467)

2026-03-26 03:48:03 -04:00

external_test_example.py

remove getenv(CI) (#16365)

2026-05-25 20:23:33 -04:00

external_test_gpu_crash.py

UOp.param accepts single sint as shape (int,) (#17784)

2026-08-27 14:07:33 -07:00

external_test_hcq_open.py

test hcq open with pytest (#13124)

2025-11-06 20:09:51 +08:00

external_test_hcq.py

remove copyin/copyout from Buffer (#17020)

2026-07-14 19:47:22 +03:00

external_test_hip_compile.py

move schedule into schedule (#15736)

2026-04-15 11:03:25 +08:00

external_test_hive_reset.py

mi350 hive_reset am repro (#15014)

2026-02-25 21:30:18 +03:00

external_test_image.py

ops_gpu -> ops_cl (#12103)

2025-09-10 15:15:48 -04:00

external_test_jit_on_models.py

DEFAULT_FLOAT/DEFAULT_INT ContextVar [pr] (#17265)

2026-07-28 19:08:18 -04:00

external_test_keccak.py

Tensor.keccak("sha3_256") (#7186)

2025-06-06 15:24:05 -07:00

external_test_llama3_ff.py

work to make GEMV fast (#5824)

2024-07-30 17:41:40 -07:00

external_test_llama3_layer.py

remove requires_grad use in extra/examples (#16238)

2026-05-16 18:40:26 -04:00

external_test_lm_head.py

isolate the 134ms kernel in train_gpt2.py (#4773)

2024-05-29 17:26:24 -04:00

external_test_losses.py

RetinaNet losses (#9536)

2025-03-21 15:52:54 -04:00

external_test_lr_schedule.py

ci: remove tensorflow from testoptim (#16717)

2026-06-23 18:11:48 -04:00

external_test_mamba.py

mark slow tests as slow instead of as CI (#13736)

2025-12-17 10:29:57 -04:00

external_test_metrics.py

log_perplexity metrics (#10912)

2025-06-21 10:44:47 -04:00

external_test_mnist_data_select.py

add quick external data select test

2024-03-02 05:38:32 -08:00

external_test_onnx_backend.py

move is_dtype_supported to renderer (#16226)

2026-05-20 21:19:37 -04:00

external_test_onnx_ops.py

no device= into arange and eye (#16503)

2026-06-04 09:21:50 -04:00

external_test_onnx_runner.py

move is_dtype_supported to renderer (#16226)

2026-05-20 21:19:37 -04:00

external_test_opt.py

more Tensor.training -> TRAINING [PR] (#16770)

2026-06-26 20:11:59 -04:00

external_test_optim.py

more Tensor.training -> TRAINING [PR] (#16770)

2026-06-26 20:11:59 -04:00

external_test_process_count.py

add external_test_process_count

2025-12-02 17:26:30 -08:00

external_test_schedule_scaling.py

don't enter CALL body in assign fixups (#17527)

2026-08-14 15:05:04 +09:00

external_test_simple_tokenizer.py

llm is the only app (#15779)

2026-04-17 10:44:48 +08:00

external_test_speed_llama.py

hcq2 to extra (#16126)

2026-05-11 17:17:30 +03:00

external_test_speed_theoretical.py

test flops (and allow wide ALU in UOps) [run_process_replay] (#5749)

2024-07-26 21:07:28 -07:00

external_test_tlsf.py

factor out generic MemoryManager (#10910)

2025-06-21 16:18:33 +03:00

external_test_usb_asm24.py

fix TestDevCopySpeeds SIZE (#17602)

2026-08-19 14:53:17 -04:00

external_test_whisper_librispeech.py

names shadowing builtins (#5179)

2024-06-27 08:15:01 -04:00

external_test_yolov8.py

padded maxpool with -inf as done in pytorch (#16858)

2026-07-07 16:10:44 -07:00

external_uop_gc.py

update a few is CONST check to check device None [pr] (#17563)

2026-08-17 07:33:16 -04:00

fuzz_fast_idiv.py

remove dtype.vec from transcendental (glm) [pr] (#16773)

2026-06-26 21:41:31 -07:00

fuzz_graph.py

remove copyin/copyout from Buffer (#17020)

2026-07-14 19:47:22 +03:00

fuzz_kfd.py

hcq: hcqbuf in copy (#15595)

2026-04-03 22:47:36 +03:00

fuzz_shape_ops.py

mark slow tests as slow instead of as CI (#13736)

2025-12-17 10:29:57 -04:00

fuzz_symbolic_div.py

ci: parallelize fuzz tests (#16805)

2026-06-30 14:14:32 -04:00

fuzz_symbolic.py

const(dtype, b) -> const(b, dtype) [PR] (#17328)

2026-07-31 09:46:37 -04:00

speed_v_theoretical.py

lower nv test_gemm_4096 target (#13107)

2025-11-05 11:05:16 -05:00