Commit Graph

Select branches

Hide Pull Requests

abstractions_4

add_local_buffer

after_qol

algebraic_upat

all_bv_no_idx

all_sink

amd_isel

amd_refac

amd_smu_reset_patch

amd_sqtt

amd_uop

amd_uop_cleanups

assign_copy

assign_try_3

before2

better_spec

better_syn

bitcast_spec

broadcast_shape_2

bt_ei

buf_heur

bump_nv_2404

cache_reshape

call_inline

call_is_kernels

callify

cast_vec_as_cast

cat_mop

changes_from_rangify

check_clsetarg

check_dims

cifar_30

cifar_30_clean

clean_load

clean_spec

clone_tg

codegen_try_2

contig_on_after

continue_mat

deepsek

default_late_allreduce

delete_kernel_py

delete_ones

delete_slow_rangeify

delta_attention

delta_attention_scan

demote_op

disable_rules

dsp_search

dsp_search_merged

dtype_shape

earlier_gater

emu_cdna

emulate_ctx

end_all_ranges

endrange

er_prioity

execution_unit

failing_assign

fancy_usb

fast_llm

fix_fa_fwd_clang_22

fix_fuse

fix_llvm_cast

fix_range_merging_sd

fix_rangeify_tests

fix_rc

fix_sched_copy

fix_slice_store

fix_src_spec

fixup_fix_kernel

full_spec

fun_w_egraphs

gate-on-load-store

gate_rewrite

gemma4_gpt

gen_pdf_fast

gh-pages

gl_dims

gldims

glm_flash

grad_outerworld

grok_x86_novec

group_div_rules

hcq2_vram_copies

heur_4096

heur_group

hk_gemm

http_threading

hwtest_fixes

id_store

if_range_cleanups

ignore_oob

image_idx_xy

image_in_mc

image_no_vec

image_refactors_stack

index_to_slice

invalid_try_3

jinja_template

kernel4

kernel_is_call

kernelless

kimi_slop

kitten_matmul

late_Locals

late_add_load

late_allreduce

late_codegen

late_decomp_is_late

late_lower_idx

late_view

late_weakint

lds_double_buffering

less_wmma_arg

lil_dg_change

lil_rangeify_td

llama_trainer

llm_262k_minimal

llm_flash_attn

llm_qwen

llm_serve_fixes

load_view_late

local_cache

locals_have_device

loop_end_op

loop_is_rng

lowerer_hacks

mac_pytest

manual_kernel_3

master

mega_lowerer

min_new_lin

min_outer_range

minor_views

mixins7

mlperf_training_v4.1

mlperf_training_v5.0

mlperf_training_v6.0

modern_scan

more_apps

more_early_comps

more_folding_rules

more_kernelcount

more_pcode

more_rangeify_pm_tests

more_ren_clean

more_uop_gc_test

more_uop_mm

more_vmap

move_devec

move_dtype_decomps

move_expander

move_gates_to_load_store

move_mop_cleanup

move_schedule

move_simp_views

moveleftright

mt_typing

multi_range

multioutput

new_ck

new_devec

new_invalid_and_image

new_linearizer

new_shape

new_test

no_assign_2

no_buf_0

no_decimals

no_late_simplify_marg

no_merge_ends

no_merge_views

no_mstack_in_allreduce

no_one_ranges

no_sip_usbgpu

no_tsink_base

no_tuplize

no_user_ck_contig

no_vec_x86_4

no_weakint_variable

noop_index

novecimg

num_batches_tracked_shape

olmo3_support

one_compiler

one_hcopt

only_assign_buffers

only_loop_backends

only_reg_emu

only_reg_emu2

outer_range

outerworld_work

pad_invalid_work

pad_mixin

param_call_is_function

parse_sqtt_more

pci_iommu

port_nir

postrange_hcopts

precompile_backward

prefill_qwen_fast

print_special_ops

ps_clean

pyremu_try

python_prg

python_speed

q4k_support

qcom-openpilot-13ms

qcom_830

qcom_mmapeak

qcom_speed_codex

qol_string

qwen36_27b_amd_900

qwen36_27b_amd_minimal

qwen36_amd_bugfixes

qwen_mergable

r_cleanups

r_new_opt

ranged_call

rangeify_2

rangeify_const_folding

rdna3_vibes

rdna4_asm

rdna4_gemm

real_pyrender

recurse_sub

reg_coalese

relu_pattern

remove_assign

remove_cast_hack

remove_ck_contig

remove_define_var

remove_forced_re

remove_on_stack

remove_programspec

remove_pylint

remove_slow_tests

remove_vec_2

remove_vec_3

renderers_use_shapes_2

replace_if_with_range

reprocess_node_experiments

reshape_trait

resnet-ci

revert-11547-fix_mismatch_red

revert-11715-revert-11711-amd_llvm_tc_cast

revert-11809-no_ki_in_gpudims

revert-11811-better_div_nesting

revert-11876-memplan_vbig

revert-12104-simplify_valid_in_symbolic

revert-12308-bufferize_early

revert-12731-jitlink_init_check

revert-12813-after_cleanups

revert-13046-better_cleanup_arange_buffers

revert-13311-prof_link_sqtt

revert-13583-rocr_desc_2

revert-14734-tload

revert-17141-full_rewrite_to_sink

revert-17408-_broadcasted-weak

revert-9438-simple_matmul_np_rand

revert_wrong_ck_fix

rewrite_rangeify2

rtoposort

scan_assign

scan_op_8

sched_cache

sched_lin

sd_load_simple

self_type

set_testpath

shrink_in_render

simp_spec

simple_gidxs

simple_pool

simple_priority

simple_rangeify

simpler-sd-compile

simpler_end

simpler_fusion

simpler_image

simpler_postrange

simpler_reduce

slice_like_shrink

spec_3

spec_cleanup

spec_ren

speed_one_linear

split_buf_idx

split_decomp_pass

split_sub

stack_0

store_is_void

string_ranges

support_opts_in_contig

swizzle_in_kernel

sym_llm

sym_work

symbolic-chunk-split

szdiff_bot_test

tc_is_heur

tensor_graph_single_sink

test_amd_crash

test_fa

test_from_postopt

test_hgemm

test_unit

tiny_scan

tmr

tool_call_support

topoprop

ttomsa-new_linearizer

tuplegtuple

two_stage_remove

typed_checks

typing_work

unused_permute_arg

uop_prg

upd_test4

update_benchmark

update_benchmark_search

update_mlperf

user_failure

vecless_x86_glm

verify_sink_dims_clean

vf_ish

view_after_const

view_in_codegen

viz_cbroadcast

warp_fun

wgpu-f16-packed

working_postopt

x86_moves

x86_numel

v0.10.0

v0.10.1

v0.10.2

v0.10.3

v0.11.0

v0.12.0

v0.13.0

v0.4.0

v0.5.0

v0.6.0

v0.7.0

v0.8.0

v0.9.0

v0.9.1

v0.9.2

Mono Color

• 011ef8fa9d Fix incorrect jit current batch devs reset (#11505) uuuvnandGitHub 2025-08-05 05:16:16 +00:00
• f02720ca2d fix fuse gate_contiguous unique (#11504) chenyuandGitHub 2025-08-04 20:43:31 -07:00
• 7ae02dea19 Merge branch 'master' into moveleftright moveleftright George HotzandGitHub 2025-08-04 19:10:27 -07:00
• 7f6acfb0d5 give define global and friends a shape (#11502) George HotzandGitHub 2025-08-04 19:09:39 -07:00
• b349e55c66 fix fuse unique fix_fuse geohot 2025-08-04 19:07:49 -07:00
• 0e91b6fd30 bugfixes geohot 2025-08-04 18:46:32 -07:00
• 823dfbde70 move view to swizzle geohot 2025-08-04 18:27:27 -07:00
• e6945b90b0 test both kernelless geohot 2025-08-04 17:01:45 -07:00
• dbd95a71dd unroll geohot 2025-08-04 16:11:44 -07:00
• 83385e7abc update gradient src in ramp.py (#11499) chenyuandGitHub 2025-08-04 15:58:03 -07:00
• 969680a246 rangeify to reduce geohot 2025-08-04 14:20:14 -07:00
• 846a2826ab viz: remove TracingKey.fmt (#11482) qazalandGitHub 2025-08-05 05:00:03 +08:00
• 01d44e8f16 tiny reduce_gradient cleanup [pr] (#11498) chenyuandGitHub 2025-08-04 13:12:53 -07:00
• dce989ed94 ctx.regs geohot 2025-08-04 11:51:58 -07:00
• 8a11af01ed remove broken paperswithcode links in doc (#11497) chenyuandGitHub 2025-08-04 10:12:33 -07:00
• 4f0ee4e982 BPE tokenizer (#11415) 3 people 2025-08-04 18:52:38 +02:00
• 11abcbd182 Merge branch 'master' into kernelless George HotzandGitHub 2025-08-04 08:48:23 -07:00
• 06af9f9236 fix double exception + add name,loc in error msg (#11487) 3 people 2025-08-04 18:41:23 +08:00
• 4877aa965a ast seems to probe nv as well (#11494) nimlgenandGitHub 2025-08-04 11:47:07 +03:00
• e0106b6b25 1/(x*c) -> (1/c)*(1/x) (#11491) chenyuandGitHub 2025-08-03 20:35:46 -07:00
• 5870352fe1 viz: factorize llvm-mca call (#11490) qazalandGitHub 2025-08-04 05:31:23 +08:00
• edf6caae81 Merge branch 'master' into kernelless George HotzandGitHub 2025-08-03 13:05:36 -07:00
• dbc7807c61 enable WEBGPU tests with buffer limit (#11489) chenyuandGitHub 2025-08-03 13:02:44 -07:00
• 8f374ee1f7 nv: print devfmr in gsp logs (#11484) nimlgenandGitHub 2025-08-03 15:12:53 +03:00
• 823f1a01db move cast around expand backward to tensor.py (#11483) chenyuandGitHub 2025-08-02 20:03:54 -07:00
• 96869abaf0 test upcast geohot 2025-08-02 18:48:02 -07:00
• af6481eaba upcast geohot 2025-08-02 18:29:49 -07:00
• b5ad66cdf2 this geohot 2025-08-02 16:56:45 -07:00
• 0ce0f51010 generic double cast folding (#11481) chenyuandGitHub 2025-08-02 16:26:37 -07:00
• 72e0d1d0dc viz: profile the compiler in TINY device (#11457) qazalandGitHub 2025-08-03 07:03:20 +08:00
• 7976a01d59 flash attention geohot 2025-08-02 14:20:11 -07:00
• 66be747908 few more dtype cast convinience methods (#11480) chenyuandGitHub 2025-08-02 12:47:09 -07:00
• e22e5da9a5 move some test_dtype tests to unit (#11479) chenyuandGitHub 2025-08-02 12:25:00 -07:00
• da0b955be4 hcq: cpu can be graphed (#11474) nimlgenandGitHub 2025-08-02 21:01:19 +03:00
• f7965f85aa Revert "feat: faster index building (#11462)" (#11478) chenyuandGitHub 2025-08-02 09:50:48 -07:00
• ef7e01cadf Fix SVD shape bug + Fix batched SVD bug (#11477) kevvzandGitHub 2025-08-02 09:47:41 -07:00
• 6ecaf8e7b2 refactor: use less index and simplify reduce axes check [pr] (#11476) 3 people 2025-08-03 00:44:51 +08:00
• 3a4deb08d2 feat: faster index building (#11462) wozeparrotandGitHub 2025-08-02 08:50:18 -07:00
• 8cc2d64edb amd: reuse create_queues for usb iface (#11473) nimlgenandGitHub 2025-08-02 14:40:46 +03:00
• 6f516e46c6 not used geohot 2025-08-01 21:42:07 -07:00
• 439ada1ada children support geohot 2025-08-01 21:39:51 -07:00
• c342809b3e test_float4_basic sort of works geohot 2025-08-01 17:10:18 -07:00
• b4ab6de416 opt work geohot 2025-08-01 16:56:47 -07:00
• 3d31b0b5f6 axistype in range geohot 2025-08-01 16:22:54 -07:00
• 5bab842337 beautiful mnist geohot 2025-08-01 15:12:17 -07:00
• 9e8e6b45ab grad acc train llama (#11467) chenyuandGitHub 2025-08-01 12:54:50 -07:00
• 427f773bc2 all test ops pass geohot 2025-08-01 11:53:56 -07:00
• 45a09207f9 test ops work (except empty) geohot 2025-08-01 11:46:15 -07:00
• 494e951e90 nicer geohot 2025-08-01 10:23:11 -07:00
• c4410e91fd rendering more geohot 2025-08-01 09:18:14 -07:00
• 7ad7329257 data parallel train llama (#11466) chenyuandGitHub 2025-08-01 09:13:51 -07:00
• 905019a4ec test gemm passes geohot 2025-08-01 09:07:44 -07:00
• 9f2182f92f cpu: start threading (#11324) nimlgenandGitHub 2025-08-01 15:35:07 +03:00
• c7ae1bd474 viz: more consistent border styling (#11464) qazalandGitHub 2025-08-01 14:31:06 +08:00
• 149c3f8fe9 test plus passes geohot 2025-07-31 22:39:42 -07:00
• 5706e2d845 something geohot 2025-07-31 22:37:26 -07:00
• c66d2082c6 kerneless will replace kernel and lowerer geohot 2025-07-31 21:46:20 -07:00
• 954f30a1c1 work mega_lowerer geohot 2025-07-31 21:11:28 -07:00
• 58e79f3fc8 bugfixes geohot 2025-07-31 19:57:15 -07:00
• aee4ebe52f linearize works geohot 2025-07-31 19:37:38 -07:00
• cb9bcc60de Merge branch 'master' into mega_lowerer George HotzandGitHub 2025-07-31 18:12:47 -07:00
• 8ff03806e8 add llama layers (#11460) George HotzandGitHub 2025-07-31 16:28:04 -07:00
• 719827b95d viz: add flops / mem bw to device programs (#11459) qazalandGitHub 2025-08-01 07:12:30 +08:00
• 3f742a5a7c comma space lab models benchmark (#11461) chenyuandGitHub 2025-07-31 16:06:18 -07:00
• 474ee9daa5 hotfix: add contiguous_backward to llama geohot 2025-07-31 15:06:19 -07:00
• fa66d9772d viz: show const node when it's root (#11456) qazalandGitHub 2025-08-01 06:01:58 +08:00
• 056dabda5a viz: refactor to color scheme (#11455) qazalandGitHub 2025-08-01 05:17:50 +08:00
• 92160f1cf1 stuff geohot 2025-07-31 13:50:04 -07:00
• e5b6149dfb more typing in drivers (#11454) nimlgenandGitHub 2025-07-31 23:26:33 +03:00
• bad3cf5731 viz: add LLVM machine code analysis (#11421) qazalandGitHub 2025-08-01 01:59:26 +08:00
• e847677e8a use AxisType in search instead of colors (#11452) chenyuandGitHub 2025-07-31 10:07:33 -07:00
• 75c2c42def suppress exceptions only during finalization (#11451) nimlgenandGitHub 2025-07-31 13:57:12 +03:00
• 24dd0d52ed feat: test remove to cpu (#11444) wozeparrotandGitHub 2025-07-30 20:18:56 -07:00
• c3cfcb50cb Add linalg_det and test for torch backend (#11405) 3 people 2025-07-30 19:04:44 -07:00
• cba3655de5 Add Test for Setitem (#10559) 3 people 2025-07-31 05:03:41 +03:00
• 6252f7770e feat: fake data (#11447) wozeparrotandGitHub 2025-07-30 17:18:20 -07:00
• 2584fe8907 naive shrink pushes everything left geohot 2025-07-30 17:12:38 -07:00
• 8dff2c1375 triple gemm geohot 2025-07-30 17:03:02 -07:00
• e300451f3a update llama3 (#11446) chenyuandGitHub 2025-07-30 16:34:21 -07:00
• 3cff1a6b13 run the lowerer on the big graph geohot 2025-07-30 15:23:43 -07:00
• 5fb975351a feat: flag for training on val (#11441) wozeparrotandGitHub 2025-07-30 14:29:45 -07:00
• 4ca430e5bf fix search dedup (#11439) chenyuandGitHub 2025-07-30 14:24:16 -07:00
• d3da20eca6 feat: bump mlperf workflow timeout to 6 hours (#11440) wozeparrotandGitHub 2025-07-30 14:12:12 -07:00
• 6c370cd524 stuff local_cache geohot 2025-07-30 14:12:06 -07:00
• 825b6a2505 feat: llama3 dataloader (#11340) wozeparrotandGitHub 2025-07-30 13:27:55 -07:00
• af357b5dc8 disable TRACK_MATCH_STATS in BEAM workers [pr] (#11437) qazalandGitHub 2025-07-31 04:22:08 +08:00
• 9185e962a3 add local caching to opts geohot 2025-07-30 13:20:11 -07:00
• 7c2d2eff86 check tensor core dims (#11436) George HotzandGitHub 2025-07-30 13:06:59 -07:00
• 5fc5bb5237 ci: clear processes (#11434) nimlgenandGitHub 2025-07-30 22:15:18 +03:00
• e051a3536f Merge branch 'master' into check_dims check_dims George HotzandGitHub 2025-07-30 12:05:58 -07:00
• c5573c989d check tc dims geohot 2025-07-30 12:05:29 -07:00
• 4f26a9ad32 check elements_per_thread in tensorcore [pr] (#11435) George HotzandGitHub 2025-07-30 11:55:48 -07:00
• 91dc19c713 check elements_per_thread in tensorcore [pr] geohot 2025-07-30 11:35:26 -07:00
• 4b4ba5454c ci: move driver start higher (#11431) nimlgenandGitHub 2025-07-30 10:48:38 +03:00
• 1bef2d80c1 unrolls are all in the same scope (#11429) George HotzandGitHub 2025-07-29 16:55:37 -07:00
• 204da24cfc increase driverbenchmark timeout-minutes to 15 (#11428) chenyuandGitHub 2025-07-29 16:45:05 -07:00
• d5fc6af4a2 remove unused ShapeTracker.consecutive [pr] (#11426) chenyuandGitHub 2025-07-29 15:36:19 -07:00
• 49a2583584 real new lowerer (#11419) George HotzandGitHub 2025-07-29 15:35:51 -07:00
• 0e5d8d5c3c remove tests that used .to_uop() (#11425) chenyuandGitHub 2025-07-29 12:52:16 -07:00
• c88e401d0e ci: fix typos in h machine benchmarks (#11423) nimlgenandGitHub 2025-07-29 22:11:47 +03:00