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

autogen-comgr-2

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

ci-deps

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

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

explicit-dev

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

gitea-test-ci

gl_dims

gldims

glm_flash

grad_outerworld

grok_x86_novec

group_div_rules

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-chat-template

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

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

onnx-cpu-count

outer_range

outerworld_work

pad_invalid

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

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-lp

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

view_after_const

view_in_codegen

viz_cbroadcast

warp_fun

wgpu-f16-packed

working_postopt

x86_moves

x86_numel

#1

#2

#3

#4

#5

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

• 6ad9a688ed add failing test after "pend substitutes for speed" (#12457) qazalandGitHub 2025-10-05 16:10:04 +03:00
• 74b04f7dca test beautiful_mnist_multigpu (#12455) chenyuandGitHub 2025-10-05 20:45:01 +08:00
• 69857d0ab0 Stable Diffusion mlperf training (#11304) 3 people 2025-10-05 07:56:05 -04:00
• a976ace404 minor improvements to rewrite (#12454) George HotzandGitHub 2025-10-05 18:09:32 +08:00
• 4b60121498 fix bmnist torch with RANGEIFY=1 (#12442) qazalandGitHub 2025-10-05 12:34:27 +03:00
• b5f31d7505 earlier seen children (#12451) George HotzandGitHub 2025-10-05 15:55:13 +08:00
• dde61c3852 Revert "write tests for algebraic UPat" reprocess_node_experiments geohot 2025-10-05 15:11:07 +08:00
• 136aeaacd3 works geohot 2025-10-05 15:11:05 +08:00
• 18552a3040 canon geohot 2025-10-05 14:58:52 +08:00
• c600446299 experiments with reprocessing node geohot 2025-10-05 14:12:40 +08:00
• 6538935441 write tests for algebraic UPat geohot 2025-10-05 08:20:13 +08:00
• 865d5796f8 add a test for untested Tensor.assign behavior (#12448) qazalandGitHub 2025-10-04 12:44:56 +03:00
• e74be4a140 UOp.factor and add chain sorting (#12413) Sieds LyklesandGitHub 2025-10-04 06:05:38 +02:00
• 394dc24110 post index symbolic (#12446) Sieds LyklesandGitHub 2025-10-03 23:23:03 +02:00
• 9f2b69b870 enable few tests for PTX test_dtype (#12445) chenyuandGitHub 2025-10-03 20:56:30 +08:00
• 0b534f71c2 recursive substitute should be O(n) (#12444) George HotzandGitHub 2025-10-03 18:29:59 +08:00
• 40f0acbc23 make that a single rewrite recurse_sub geohot 2025-10-03 18:18:43 +08:00
• f3ac529438 even faster geohot 2025-10-03 18:14:41 +08:00
• a3991948e9 recursive substitute geohot 2025-10-03 18:03:55 +08:00
• b087663c35 RANGEIFY test_bert uses more ran somehow (#12443) chenyuandGitHub 2025-10-03 16:38:53 +08:00
• 940a8d5ba9 default IGNORE_OOB=1 (#12441) chenyuandGitHub 2025-10-03 16:16:19 +08:00
• d290e77a5b pend substitutes for speed (#12440) George HotzandGitHub 2025-10-03 15:49:19 +08:00
• 23d310bcc1 ptx: handle i8/u8 casts correctly (#12439) nimlgenandGitHub 2025-10-03 10:34:15 +03:00
• 1e8945a28c Training loop for Stable Diffusion mlperf (#12315) hoovedandGitHub 2025-10-03 02:45:38 -04:00
• c7849ac593 fix test lil model (#12437) George HotzandGitHub 2025-10-03 14:28:37 +08:00
• 0f82d92b9d use float for softmax in llm.py (#12438) chenyuandGitHub 2025-10-03 14:27:56 +08:00
• b9f7a7e218 Merge branch 'master' into add_local_buffer George HotzandGitHub 2025-10-03 13:06:14 +08:00
• 4c63f7e786 skip copies of reshaped buffers (#12430) George HotzandGitHub 2025-10-03 13:05:47 +08:00
• 0047bcc535 undo loaded comparison swap (#12436) Sieds LyklesandGitHub 2025-10-03 06:57:29 +02:00
• f203d8b221 update RANGEIFY kernel count and test_masked_select (#12435) chenyuandGitHub 2025-10-03 12:41:34 +08:00
• a6dd5a224b skip webgpu tests (#12433) wozeparrotandGitHub 2025-10-02 21:31:07 -07:00
• bf99de7b1e update a few more tests for RANGEIFY (#12434) chenyuandGitHub 2025-10-03 12:16:58 +08:00
• 9273d7d404 adding a local buffer is very simple now geohot 2025-10-03 11:17:57 +08:00
• a734437da8 skip copies of reshaped buffers geohot 2025-10-03 10:55:58 +08:00
• a8dca47fbc fix support_opts_in_contig geohot 2025-10-03 10:52:18 +08:00
• 769db23df6 that geohot 2025-10-03 10:45:47 +08:00
• 1cd40941c8 delete junk geohot 2025-10-03 10:35:12 +08:00
• 9a607e69e1 Merge branch 'master' into support_opts_in_contig George HotzandGitHub 2025-10-03 10:32:08 +08:00
• 9cd365c12e little changes from double gemm (#12429) George HotzandGitHub 2025-10-03 10:31:51 +08:00
• 16a65b4fd0 fix test_symbolic_gcd_div hang (#12427) Sieds LyklesandGitHub 2025-10-03 04:21:16 +02:00
• 797e74ba1f normal lil_dg_change geohot 2025-10-03 10:12:12 +08:00
• 0f0b0cf17f small changes from double gemm work geohot 2025-10-03 10:07:46 +08:00
• 5b05bf4ab4 Merge branch 'master' into support_opts_in_contig George HotzandGitHub 2025-10-03 10:03:14 +08:00
• 2d24af888b REWRITE_STACK_LIMIT (#12426) chenyuandGitHub 2025-10-03 09:51:04 +08:00
• 1b58ef0d60 Increase stack size limit in unified_rewrite (#12424) hoovedandGitHub 2025-10-02 21:06:47 -04:00
• 17d36d0952 don't tag MSTACK/MSELECT on global buffers (#12423) qazalandGitHub 2025-10-02 13:32:15 +03:00
• e59e0aadc1 opts geohot 2025-10-02 18:25:38 +08:00
• 3688afa513 fix swap geohot 2025-10-02 18:17:03 +08:00
• dae164ffb1 warp geohot 2025-10-02 18:04:45 +08:00
• 3fb3dd4c06 flash attention with two gemms geohot 2025-10-02 17:48:48 +08:00
• e5028d58e9 flash attention sort of works geohot 2025-10-02 17:29:23 +08:00
• 7b3912d8e4 relax atol for some tests (#12422) chenyuandGitHub 2025-10-02 17:04:44 +08:00
• 98163832e4 update RANGEIFY test_cast_padded (#12421) chenyuandGitHub 2025-10-02 16:37:35 +08:00
• 5a602e6c36 double wmma works geohot 2025-10-02 16:06:46 +08:00
• 37beef6de3 add null bert training test in ci (#12420) chenyuandGitHub 2025-10-02 16:05:19 +08:00
• 6640514555 demote works on both matmuls geohot 2025-10-02 15:42:22 +08:00
• f21851b099 ops: n^2 .device property fix (#12419) 3 people 2025-10-02 10:28:12 +03:00
• ec177c80c2 rangeify: fix test_where_fold (llvm) (#12416) b1tgandGitHub 2025-10-02 14:57:49 +08:00
• 13a25b2e67 rangeify: don't shape INDEX on kernelize (#12417) qazalandGitHub 2025-10-02 09:45:37 +03:00
• 5d9035f5a6 Eval for Stable Diffusion mlperf (#12316) hoovedandGitHub 2025-10-02 02:35:38 -04:00
• 0f804c9a83 Stable Diffusion model init for mlperf (#12314) 3 people 2025-10-02 02:28:41 -04:00
• 2fbd7d21f9 tc passes geohot 2025-10-02 14:27:52 +08:00
• 0eee93f0c0 hotfix: disable split ranges for non rangeify geohot 2025-10-02 13:15:24 +08:00
• f32a497f08 bug geohot 2025-10-02 13:10:07 +08:00
• 3fd25a425b fix gfr geohot 2025-10-02 12:58:31 +08:00
• 3da569c20b Merge branch 'master' into support_opts_in_contig George HotzandGitHub 2025-10-02 12:57:43 +08:00
• 583553f467 split ranges (#12411) George HotzandGitHub 2025-10-02 12:57:22 +08:00
• 6fc6b51b59 fix limit_bufs with kernelize (#12415) qazalandGitHub 2025-10-02 07:49:11 +03:00
• 9d5d4b248c Merge branch 'master' into support_opts_in_contig George HotzandGitHub 2025-10-02 12:39:50 +08:00
• d1c868f990 fix limit_bufs with multi (#12414) qazalandGitHub 2025-10-02 05:51:56 +03:00
• 2fcd55583f allow less kernels in external_test_opt (#12412) qazalandGitHub 2025-10-02 05:05:42 +03:00
• 8b48e19ce2 skip more multi remote tests (#12410) qazalandGitHub 2025-10-02 04:50:46 +03:00
• 3770dd9d80 annotate bufferize in viz geohot 2025-10-02 09:19:41 +08:00
• 5b649616ff rangeify: detect and assert cycles (#12405) qazalandGitHub 2025-10-02 03:39:43 +03:00
• 9a64fc0d28 Load alt value with cast try 2 (#12407) Sieds LyklesandGitHub 2025-10-02 00:55:29 +02:00
• 3e0e0290ce increase timeout in test_module_runs (#12408) nimlgenandGitHub 2025-10-01 22:01:44 +03:00
• 2f8ac77c25 add allow_any_len to gated load patterns in renderers (#12406) Sieds LyklesandGitHub 2025-10-01 20:35:32 +02:00
• 89bed28716 split reduceop (#12404) George HotzandGitHub 2025-10-01 18:45:16 +08:00
• 74ee305948 some rangeify tests fixed (#12403) George HotzandGitHub 2025-10-01 18:23:37 +08:00
• f198a9e1ba skip test_multihost_aware_schedule, assign devices mismatch (#12396) qazalandGitHub 2025-10-01 13:09:15 +03:00
• ac3d457d5e rangeify: TestReduceOpsConstFolding (#12397) 3 people 2025-10-01 17:58:19 +08:00
• c449e8eb17 don't change that geohot 2025-10-01 17:47:43 +08:00
• 3dc1b2e98e broken geohot 2025-10-01 17:27:17 +08:00
• 8e6126160f Merge branch 'master' into support_opts_in_contig George HotzandGitHub 2025-10-01 17:20:51 +08:00
• 60e52fbe36 support opts in contig, simpler (#12400) George HotzandGitHub 2025-10-01 17:20:04 +08:00
• 6c95b1f39d explicitly set device for CI unit test (#12399) chenyuandGitHub 2025-10-01 17:16:54 +08:00
• 115810d13e dual tc geohot 2025-10-01 16:51:04 +08:00
• 215c3e19d9 always print that geohot 2025-10-01 16:47:59 +08:00
• eedf570e7b late locals geohot 2025-10-01 16:36:49 +08:00
• 09ec0f7464 tag TC geohot 2025-10-01 16:18:14 +08:00
• 6ba8bf282f skip test_masked_select for RANGEIFY PYTHON (#12395) chenyuandGitHub 2025-10-01 16:13:31 +08:00
• 689ab9151b more RANGEIFY tests (#12393) chenyuandGitHub 2025-10-01 15:43:58 +08:00
• 481ffa9aed support opts in rangeify contig geohot 2025-10-01 15:39:23 +08:00
• adc8c3b28f Revert "load alt value with cast (#12384)" (#12392) chenyuandGitHub 2025-10-01 15:20:04 +08:00
• 154d114364 rangeify: fix abstractions2.py (#12386) b1tgandGitHub 2025-10-01 14:58:56 +08:00
• fe96c8d345 add HALF flag to tinygrad.apps.llm geohot 2025-10-01 14:43:50 +08:00
• f205352cd7 remove ranges with 1s (#12388) George HotzandGitHub 2025-10-01 14:43:29 +08:00
• 90b1c0dd96 rangeify: test_where_fold kernel count (#12379) qazalandGitHub 2025-10-01 09:35:12 +03:00
• 42748ccb92 rangeify: fix test_prequant_conv2d_1x1 (#12391) b1tgandGitHub 2025-10-01 14:33:47 +08:00
• 310ef10542 Merge branch 'master' into no_one_ranges no_one_ranges George HotzandGitHub 2025-10-01 14:31:04 +08:00