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

• 90c777d815 remove apply_auto_opt (#2063) George HotzandGitHub 2023-10-13 07:44:14 -07:00
• bd42fa0b73 kernel cache (#2035) nimlgenandGitHub 2023-10-13 16:32:01 +03:00
• 6f1810af2d with unroll, the action space goes from 161 -> 127 (#2060) George HotzandGitHub 2023-10-12 20:52:23 -07:00
• 6b7ac5c431 ModNode __mod__ rule (#2039) Umut ZenginandGitHub 2023-10-12 21:30:10 +03:00
• 3187962476 CIFAR HALF mode (#2041) Yixiang GaoandGitHub 2023-10-12 12:19:51 -05:00
• c5edb3c374 train value net, improve API, add BCE (#2047) George HotzandGitHub 2023-10-12 07:56:38 -07:00
• 0ba629c7b9 add world dataset (#2045) George HotzandGitHub 2023-10-11 15:54:30 -07:00
• 0c3b6f13a8 Latest opt (#2044) George HotzandGitHub 2023-10-11 15:46:14 -07:00
• 8d6cecb25c Torch eq fix (#1562) geohotstanandGitHub 2023-10-12 03:57:11 +08:00
• 41bfeb2c1e start work on auto opt (#2034) George HotzandGitHub 2023-10-11 12:54:53 -07:00
• 1c980517c5 s/var_vals_from_ast/vars_from_ast (#2038) chenyuandGitHub 2023-10-10 23:21:55 -04:00
• 81c7d750db test: fix test_linearizer.test_tensor_core test (#2036) Francis LamandGitHub 2023-10-10 14:48:28 -07:00
• e2b83f1b42 Variable.bind newer (#2017) chenyuandGitHub 2023-10-10 13:03:01 -04:00
• 71d93ffd79 Refactor GPU and Metal langauges in their own separate renderers (#2033) qazalandGitHub 2023-10-10 17:46:41 +03:00
• f139060103 Rewrite hand coded opt with action space (#2030) George HotzandGitHub 2023-10-10 07:38:38 -07:00
• e27fedfc7b Fix stable diffusion output error on WebGPU (#2032) Ahmed HarmoucheandGitHub 2023-10-10 15:40:51 +02:00
• e40f141203 Refactor and add more unit tests for disktensors (#2022) qazalandGitHub 2023-10-10 04:46:29 +03:00
• 45f0891a8f use "<" instead of "<=" in codegen for loop (#2027) chenyuandGitHub 2023-10-09 20:26:36 -04:00
• 25555c836f llama default to JIT only if device supports JIT (#2028) chenyuandGitHub 2023-10-09 20:26:02 -04:00
• 16ca8410f8 op logger + replay (#2021) George HotzandGitHub 2023-10-08 15:10:18 -07:00
• 46f354b49f Fix comment to describe code (#2023) calleditandGitHub 2023-10-08 23:28:14 +02:00
• 0e2e041faf CI for using tinygrad as an external pkg (#2019) qazalandGitHub 2023-10-08 20:50:48 +03:00
• 8db92bd060 fix tvm gemm example geohot 2023-10-08 05:57:41 -07:00
• af6e2f31ca whisper: cast model output token to int32 (#2013) 3 people 2023-10-08 05:56:22 -07:00
• e93e240a6c adapting test/external/external_osx_profiling.py to the new code base (#2002) Luca SciarpaandGitHub 2023-10-08 14:55:00 +02:00
• c4e8ea73bd feat: add tinygrad.features to setup.py (#2016) wozeparrotandGitHub 2023-10-08 00:55:50 -04:00
• dece9958f8 wmma: clean up to make WMMA arg order consistent (#2014) Francis LamandGitHub 2023-10-07 17:45:40 -07:00
• cea4cbfc7a move image+kopt to features (#2015) George HotzandGitHub 2023-10-07 15:41:08 -07:00
• 44ed94ef5c use the device abstraction in handcode_resnet50_opt geohot 2023-10-07 13:22:20 -07:00
• 6ee9cae44f don't extract CIFAR every time / use the cache geohot 2023-10-07 12:33:24 -07:00
• d07ac379f9 add var_vals to kopt with symbolic (#2008) nimlgenandGitHub 2023-10-07 19:34:21 +03:00
• 121f7aa8c5 Schedule item (#2012) George HotzandGitHub 2023-10-07 08:59:25 -07:00
• f1f64bc88d remove val_vars from the linearizer (#2009) George HotzandGitHub 2023-10-07 07:47:28 -07:00
• dea8bb0938 triton isn't tested, and allows this refactor (#2007) George HotzandGitHub 2023-10-07 07:29:59 -07:00
• 23de1db727 strip whitespace geohot 2023-10-07 06:06:27 -07:00
• 26fcc8dff6 fix: remove runtime imports (#1982) Roelof van DijkandGitHub 2023-10-07 14:23:08 +02:00
• f54959e5cd move print tree into graph (#2003) George HotzandGitHub 2023-10-07 04:39:21 -07:00
• 2114dc13d1 Allow multi-input model export (#1995) Ahmed HarmoucheandGitHub 2023-10-07 13:13:34 +02:00
• ffa33d743a good changes from openpilot_compile2 (#2000) George HotzandGitHub 2023-10-06 13:33:24 -07:00
• 05be57f57f Fix llama with empty prompt (#1997) chenyuandGitHub 2023-10-06 09:48:07 -04:00
• 7a68060422 Revert "allow local + grouped reduce in hand_coded (#1996)" (#1998) George HotzandGitHub 2023-10-06 06:43:28 -07:00
• 219a1f7063 allow local + grouped reduce in hand_coded (#1996) nimlgenandGitHub 2023-10-06 16:11:28 +03:00
• fa9945dac0 remove stale tests geohot 2023-10-06 02:14:56 -07:00
• 94b21c41a7 ci: use mypy.ini (#1993) Vidhan BhattandGitHub 2023-10-06 04:45:28 -04:00
• e43d8977f8 Revert "chore: add py.typed marker. (#1991)" (#1994) George HotzandGitHub 2023-10-06 01:44:34 -07:00
• 6d581e8911 chore: add py.typed marker. (#1991) Vidhan BhattandGitHub 2023-10-05 19:27:33 -04:00
• da2b3e55f4 simpler llama - don't shrink twice (#1981) chenyuandGitHub 2023-10-05 17:31:46 -04:00
• 972d9ea215 fix: PRUNEGRAPH is unused (#1985) Roelof van DijkandGitHub 2023-10-05 23:28:43 +02:00
• 21a2c5df73 fix up contiguous (#1978) George HotzandGitHub 2023-10-05 07:22:05 -07:00
• c99fa58dd2 simplify gpt2 example (#1973) chenyuandGitHub 2023-10-05 10:09:29 -04:00
• 2d0c1037b1 Fix up latest openpilot model (#1976) George HotzandGitHub 2023-10-05 05:24:28 -07:00
• 1862e14a4f fix gemv triggering for gemm (#1975) George HotzandGitHub 2023-10-05 05:23:00 -07:00
• 0ba75c4370 optimizer: add matvec optimizations (#1972) Francis LamandGitHub 2023-10-04 14:16:27 -07:00
• 3d5127038c don't create linearizer if we are in the method cache (#1969) George HotzandGitHub 2023-10-04 12:42:58 -07:00
• de5d603ec1 corealize + remove realize from lazybuffer (#1968) George HotzandGitHub 2023-10-04 10:59:31 -07:00
• 88b6ed6945 disable broken optim_conv2d geohot 2023-10-04 07:33:50 -07:00
• d449b3bef1 think about removing realize from lazybuffer (#1965) George HotzandGitHub 2023-10-04 07:18:58 -07:00
• 2ea1dd3e87 no process() in Linearizer (#1966) nimlgenandGitHub 2023-10-04 17:18:42 +03:00
• 0945848b5f schedule the loadops like everything else (#1964) George HotzandGitHub 2023-10-04 02:36:04 -07:00
• fb4d830a2a Fix cast error in render_load in wgsl (#1956) Ahmed HarmoucheandGitHub 2023-10-04 11:29:14 +02:00
• 6a79d4044a unrealized consts everywhere (#1963) George HotzandGitHub 2023-10-04 01:48:10 -07:00
• f04c1a63ae Rand works in jit (#1960) 3 people 2023-10-03 22:55:25 +03:00
• f64d5b3ba8 move to realize.py (#1961) George HotzandGitHub 2023-10-03 07:25:40 -07:00
• 717451a244 Revert "optimizer: add matvec optimizations (#1753)" (#1959) George HotzandGitHub 2023-10-03 00:28:42 -07:00
• f520323054 optimizer: add matvec optimizations (#1753) 3 people 2023-10-03 00:01:59 -07:00
• e1f2c2cc19 fix jitted dist (#1955) nimlgenandGitHub 2023-10-02 18:45:13 +03:00
• 35ac60775b simplify line (#1950) 3 people 2023-10-02 12:19:15 +02:00
• 08e884217c metal batch executor (#1920) 3 people 2023-10-02 13:18:31 +03:00
• d48a90859c use the opts from the default device (#1954) George HotzandGitHub 2023-10-02 03:13:46 -07:00
• c27971d51f fix llvm nan/inf const (#1951) 3 people 2023-10-02 13:08:57 +03:00
• 6a4ec4776e fix CI (#1953) George HotzandGitHub 2023-10-02 02:58:58 -07:00
• 579cabf668 Fix examples/train_efficientnet (#1947) Daniel RiegeandGitHub 2023-10-02 11:23:38 +02:00
• d4671cd8e3 use schedule in more places in linearizer tests (#1946) David HouandGitHub 2023-10-02 02:22:56 -07:00
• e7a49e84c8 perf: assert behind if is not optimized (#1847) 4 people 2023-09-29 20:07:24 +02:00
• 8e9db88474 expand after expr_idxs in Linearizer.global_load (#1818) 3 people 2023-09-29 10:33:34 -07:00
• 692bec7b6f simplify CacheCollector (#1944) nimlgenandGitHub 2023-09-29 20:13:04 +03:00
• 90326dbdc3 resnet50 hand coded optimization (#1945) George HotzandGitHub 2023-09-29 09:34:51 -07:00
• a677a1e2cd winograd test prints op count geohot 2023-09-29 05:41:29 -07:00
• 4ff35e2b97 better resnet eval (#1943) George HotzandGitHub 2023-09-29 05:40:25 -07:00
• 48c8d130ae simpler GPT2 (#1941) George HotzandGitHub 2023-09-29 04:41:09 -07:00
• 81cb120b0f winograd speed test (#1942) George HotzandGitHub 2023-09-29 04:40:35 -07:00
• d52df788d3 remove RawConst and add test (#1939) George HotzandGitHub 2023-09-29 01:21:51 -07:00
• 22b8576887 more lazy cleanup (#1938) George HotzandGitHub 2023-09-29 00:53:29 -07:00
• 2a49f7e456 fix transfer to mapped buffers (#1923) nimlgenandGitHub 2023-09-29 10:50:24 +03:00
• f445e056ed wmma: add test and tensor core shape (#1925) Francis LamandGitHub 2023-09-28 18:04:28 -07:00
• 094d3d71be with Tensor.train() (#1935) Yixiang GaoandGitHub 2023-09-28 20:02:31 -05:00
• 10f0dc0c85 keep only one comment from git action bot (#1936) Yixiang GaoandGitHub 2023-09-28 19:24:53 -05:00
• 70671d9625 fix test_collectives (#1934) wozeparrotandGitHub 2023-09-28 14:02:22 -04:00
• c36d0e3bd8 tvm import hook geohot 2023-09-28 09:24:32 -07:00
• adab724caa schedule2, keep the tests working with small changes (#1932) George HotzandGitHub 2023-09-28 09:14:43 -07:00
• c6d5e471d0 Do not import typing_extensions at runtime (#1927) Antoine AdamandGitHub 2023-09-28 10:57:28 +02:00
• 164f8a1923 fix hipgraph exec (#1929) nimlgenandGitHub 2023-09-27 19:54:21 +03:00
• 9c6bb7ff13 fix: add sentencepiece to testing dependencies (#1919) Sean D'SouzaandGitHub 2023-09-25 11:22:01 -04:00
• c907efbf4a reorder a few things (#1915) George HotzandGitHub 2023-09-25 10:17:21 +08:00
• 25a767cd5d Remove LtNode.__mul__ and AndNode.__mul__ (#1913) chenyuandGitHub 2023-09-24 19:03:59 -04:00
• eaa8d343d8 Remove str type from map_buffers (#1912) chenyuandGitHub 2023-09-24 19:03:22 -04:00
• ae9529e678 chore: remove redundant noise in stable diffusion example (#1910) Dat D. NguyenandGitHub 2023-09-24 20:33:45 +07:00
• 6d9065ed1c Minor cleanups (#1911) George HotzandGitHub 2023-09-24 21:32:50 +08:00
• 20059dc55b Make ShapeTracker Immutable (#1909) George HotzandGitHub 2023-09-24 21:09:03 +08:00
• 45f02393f0 HipGraph support (#1880) nimlgenandGitHub 2023-09-24 15:14:36 +03:00