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

• df01360a7b x hcq2_vram_copies nimlgen 2026-08-17 13:20:15 +00:00
• b429f5e711 hcq2: speed nimlgen 2026-08-16 14:22:53 +00:00
• a521b1abe9 stack delta_attention geohot 2026-08-17 06:46:10 -07:00
• 02318048e9 Deployed 58edff6 with MkDocs version: 1.6.1 gh-pages github-actions[bot] 2026-08-17 13:13:01 +00:00
• 58edff61d9 hcq2: one submitter (#17556) master nimlgenandGitHub 2026-08-17 16:08:19 +03:00
• f381ad3b29 fix update_benchmark nimlgen 2026-08-17 05:27:40 -07:00
• 954d4f7797 add back beautiful_mnist_torch in CI (#17569) chenyuandGitHub 2026-08-17 08:23:34 -04:00
• 7fe8e350c5 delete bad torch backend function override (#17568) chenyuandGitHub 2026-08-17 07:56:05 -04:00
• 42714e1399 update a few is CONST check to check device None [pr] (#17563) chenyuandGitHub 2026-08-17 07:33:16 -04:00
• 821e80ff9a remove torch backend detach hack (#17565) chenyuandGitHub 2026-08-17 07:33:05 -04:00
• f96f3a32f6 fst nimlgen 2026-08-17 01:45:40 -07:00
• 5824a210bd x nimlgen 2026-08-17 00:06:44 -07:00
• 37a54dc7cf add some dels to jit for OOM fixes (#17566) George HotzandGitHub 2026-08-16 23:39:37 -07:00
• 36fb8803d0 Dx nimlgen 2026-08-16 23:06:43 -07:00
• 7fa239c909 bugfixes geohot 2026-08-16 23:00:42 -07:00
• 9213223c82 revrt nimlgen 2026-08-16 22:52:31 -07:00
• 38655e8123 Merge branch 'master' into hcq2_one_submitter nimlgenandGitHub 2026-08-17 08:42:51 +03:00
• 2d7fcf6df8 cleanups geohot 2026-08-16 19:15:11 -07:00
• 024d0ad4b2 refactor delta attention geohot 2026-08-16 19:03:51 -07:00
• e25f86721d more torch backend fixups (#17562) chenyuandGitHub 2026-08-16 21:31:53 -04:00
• 138fb4a783 delete dead DType.scalar [PR] (#17561) chenyuandGitHub 2026-08-16 21:12:17 -04:00
• bfd4048abf no dtype in vconst_like [PR] (#17560) chenyuandGitHub 2026-08-16 21:06:38 -04:00
• 057a18a07c fix emulated long cast to double (#17559) chenyuandGitHub 2026-08-16 20:52:42 -04:00
• c30bf116b7 few torch_backend fix (#17558) chenyuandGitHub 2026-08-16 20:07:25 -04:00
• d435280c05 Dx nimlgen 2026-08-16 13:16:25 -07:00
• 5ff2028a87 x nimlgen 2026-08-16 08:04:02 -07:00
• 2fcb40d400 x nimlgen 2026-08-16 14:02:23 +00:00
• 3c6f47b5f6 simpler nimlgen 2026-08-16 13:55:26 +00:00
• 8bdcc1b6d1 simpler nimlgen 2026-08-16 13:31:12 +00:00
• 52c07bf381 x nimlgen 2026-08-16 13:17:23 +00:00
• c9e15c7f99 Merge branch 'master' into delta_attention_scan delta_attention_scan George HotzandGitHub 2026-08-16 06:12:15 -07:00
• 90823835c8 x nimlgen 2026-08-16 13:04:44 +00:00
• 8063095c71 x nimlgen 2026-08-16 12:15:05 +00:00
• 0a2d1e7689 hcq2: c submitter nimlgen 2026-08-16 11:30:51 +00:00
• e7bf2a811d iface in device (#17554) nimlgenandGitHub 2026-08-16 12:09:24 +03:00
• f77ea3f8cc Merge branch 'master' into delta_attention_scan George HotzandGitHub 2026-08-15 20:08:08 -07:00
• e688e07758 add max_shape/max_numel to mixins + pad_to (#17553) George HotzandGitHub 2026-08-15 20:05:10 -07:00
• a231c73d06 Merge branch 'master' into delta_attention_scan George HotzandGitHub 2026-08-15 17:48:55 -07:00
• 97022960ae device: fix remap (#17549) nimlgenandGitHub 2026-08-16 01:01:36 +03:00
• 7b20625ffd Merge branch 'master' into delta_attention_scan George HotzandGitHub 2026-08-15 14:38:32 -07:00
• 70902a6aaf drop fxn alias geohot 2026-08-15 14:27:24 -07:00
• 7a3c11b25d locals sequentialize on renderers without workgroups in gpudims, drop CPU special case geohot 2026-08-15 14:24:59 -07:00
• 417563ca20 fix webgpu is_nan [pr] (#17551) chenyuandGitHub 2026-08-15 16:31:19 -04:00
• 72dc8b2b3f Tensor.max_shape on mixin, pad_to value like pad geohot 2026-08-15 13:15:44 -07:00
• 369ccfbd7f cleaner: single custom_kernel call path via scalar initial flag geohot 2026-08-15 12:48:17 -07:00
• dfe08dfcf7 variables should not be weakints no_weakint_variable geohot 2026-08-15 12:19:26 -07:00
• 7a83fec3ad Merge branch 'master' into rewrite_rangeify2 rewrite_rangeify2 George HotzandGitHub 2026-08-15 12:06:41 -07:00
• ff27f2ea74 adapt to Ops.BIND removal geohot 2026-08-15 11:08:39 -07:00
• b08c99704b cpu support geohot 2026-08-13 07:01:09 -07:00
• ceca344627 write generic delta attention scan geohot 2026-08-12 22:47:17 -07:00
• 5ca87f1bac fix cast to weak twice [pr] (#17548) chenyuandGitHub 2026-08-15 12:39:54 -04:00
• 26cbadd69a no pm_fold_cast_const in full_rewrite_to_sink [PR] (#17547) chenyuandGitHub 2026-08-15 10:18:59 -04:00
• fae893753b no pm_fold_cast_const in get_kernel_graph [pr] (#17546) chenyuandGitHub 2026-08-15 09:57:05 -04:00
• c17849a1f8 disk: neable polling (#17538) nimlgenandGitHub 2026-08-15 15:03:28 +03:00
• 539a03343a no casted const from sub and div [pr] (#17543) chenyuandGitHub 2026-08-15 07:47:40 -04:00
• a57569349c renumber invalids before callify (#17542) qazalandGitHub 2026-08-15 17:00:54 +08:00
• 5c43a89fb1 precompile_backward tests for sched_cache (#17544) qazalandGitHub 2026-08-15 14:25:31 +08:00
• 2015a4f4a7 llm kernels: adapt to Ops.BIND removal qwen_mergable geohot 2026-08-14 22:36:20 -07:00
• 72646094df Merge origin/master into qwen_mergable geohot 2026-08-14 21:58:34 -07:00
• e6f5bb9c09 simple test for Invalid clone cache miss regression (#17541) qazalandGitHub 2026-08-15 10:00:21 +08:00
• 4b0525e594 no pm_fold_cast_const in UOp.simplify and hcq2 [pr] (#17540) chenyuandGitHub 2026-08-14 21:39:47 -04:00
• 64ccbde3bb clean up STACK with a const [PR] (#17539) chenyuandGitHub 2026-08-14 21:07:01 -04:00
• 6ea665ed66 remove pm_fold_cast_const from dtype decomp [pr] (#17536) chenyuandGitHub 2026-08-14 16:24:51 -04:00
• 0725acc392 reenable hcq2 ci (#17532) nimlgenandGitHub 2026-08-14 23:13:18 +03:00
• 4a1f32977c gptoss: default GROUPED_MOE=1 (#17537) wozeparrotandGitHub 2026-08-15 03:24:37 +08:00
• 13c381b0c0 remove pm_fold_cast_const from initial symbolic [pr] (#17535) chenyuandGitHub 2026-08-14 14:14:40 -04:00
• ac7067ac60 fix deconstruct_function for python 3.11 (#17534) chenyuandGitHub 2026-08-14 13:38:34 -04:00
• 80169c6758 remove where push cast to branches from sym [pr] (#17533) chenyuandGitHub 2026-08-14 12:57:56 -04:00
• adacaa3e17 hcq2 fix hangs (#17529) nimlgenandGitHub 2026-08-14 16:23:59 +03:00
• 89ab344c42 fix assign into bitcast with no explicit realize (#17531) chenyuandGitHub 2026-08-14 09:23:01 -04:00
• 25c3bd027b remove pm_fold_cast_const in simplify_merge_adjacent [pr] (#17530) chenyuandGitHub 2026-08-14 09:10:05 -04:00
• 6b35220622 cpu hcq2 (#17503) nimlgenandGitHub 2026-08-14 15:06:03 +03:00
• b1859805b1 remove Ops.BIND (#17511) George HotzandGitHub 2026-08-13 23:52:16 -07:00
• faba071b1d don't enter CALL body in assign fixups (#17527) qazalandGitHub 2026-08-14 14:05:04 +08:00
• 81dc8ec232 Revert "amd: fix ALL2ALL speed on amdgpu (gpt) (#17518)" (#17526) qazalandGitHub 2026-08-14 09:39:14 +08:00
• 95ca5081fe hotfix: update extra/runbook_digitalocean_mi350x geohot 2026-08-13 17:23:47 -07:00
• 303d1677b3 qcomcl: use qemu for compilation (#17524) sirhcmandGitHub 2026-08-13 17:16:02 -07:00
• 673c6463f9 disable HCQ2 for AMD CI (#17523) George HotzandGitHub 2026-08-13 13:14:40 -07:00
• 849074f0db gptoss: use fa swa (#17522) wozeparrotandGitHub 2026-08-14 01:39:22 +08:00
• 0252cb8fa7 remove anchors from CI flow (#17521) George HotzandGitHub 2026-08-13 10:17:22 -07:00
• cc6d33bde7 amd: fix ALL2ALL speed on amdgpu (gpt) (#17518) qazalandGitHub 2026-08-13 15:24:45 +08:00
• 16c5ff2490 add external_benchmark_all2all.py (#17507) qazalandGitHub 2026-08-13 14:50:43 +08:00
• 1b7f040984 fa: paas through window (#17517) wozeparrotandGitHub 2026-08-13 14:45:57 +08:00
• 39d144546e fix mxfp4 mem estimate (#17515) qazalandGitHub 2026-08-13 13:48:15 +08:00
• e103fb2a10 more lil llm improvements (#17514) George HotzandGitHub 2026-08-12 20:13:20 -07:00
• 922506676a Merge branch 'master' into qwen_mergable George HotzandGitHub 2026-08-12 19:30:25 -07:00
• 2297118541 lil llm improvements (#17513) George HotzandGitHub 2026-08-12 19:29:32 -07:00
• cd6d0d6ee3 allow running QCOMCL compiler in docker (#17499) sirhcmandGitHub 2026-08-12 19:27:55 -07:00
• 26f2049d43 Merge branch 'master' into bitcast_spec bitcast_spec George HotzandGitHub 2026-08-12 19:24:33 -07:00
• 89a63a3550 cleanup cast geohot 2026-08-12 19:17:09 -07:00
• e556609538 quant 256 multiple geohot 2026-08-12 19:13:01 -07:00
• 23744a5e55 AMD geohot 2026-08-12 19:09:30 -07:00
• 672747b16b Merge branch 'master' into qwen_mergable George HotzandGitHub 2026-08-12 16:04:58 -07:00
• 138afe15bc mergable fast RDNA3 Qwen 3.6 geohot 2026-08-12 13:53:37 -07:00
• ff0cb28c21 skip slow whisper tests geohot 2026-08-12 13:12:37 -07:00
• ed8297a102 kerenl opts test from nan in llama 8b (#17510) qazalandGitHub 2026-08-13 03:07:30 +08:00
• de04781b36 simplify equivalent const max (#17505) RaineandGitHub 2026-08-12 12:39:29 -03:00
• 4f106ebe87 hcq2: enqueue speed (#17504) nimlgenandGitHub 2026-08-12 13:08:53 +03:00
• 04c271ac41 simplify digitalocean_mi350x (#17502) qazalandGitHub 2026-08-12 15:00:04 +08:00
• 2e5a9a4121 no hardcoded device names in test_sliced_buffer_function (#17501) qazalandGitHub 2026-08-12 14:19:25 +08:00