nimlgen/tinygrad
1
0
Fork 0
forked from tinygrad/tinygrad
Code Pull Requests Activity
13,747 Commits 334 Branches 15 Tags
new_devec
Commit Graph
262 Commits
Author SHA1 Message Date
chenyuandGitHub 3a9b87e48f remove Ops.DEVICE from Ops.PROGRAM [PR] (#16797) 2026-06-29 17:41:31 -04:00
qazalandGitHub b8224e19a7 llama: remove unused fp8 transpose kernel (#16763) 2026-06-26 16:31:13 +09:00
wozeparrotandGitHub e0c69d7a12 llama: fused grad quantize (#16731) 2026-06-24 23:25:39 -04:00
wozeparrotandGitHub fe9b19b12d llama: more mp mem fixes (#16701) 
* llama: more mp mem fixes

* clean: unused

* fix: batch
 2026-06-22 10:54:35 -04:00
chenyuandGitHub 267af9c601 full_like to CreationMixin [PR] (#16702) 2026-06-22 09:33:23 -04:00
wozeparrotandGitHub 5164c21b44 gemm: keep shape thru mxfp8 quantize (#16692) 2026-06-20 22:28:53 -07:00
George HotzandGitHub 649971f02a remove DEFINE_LOCAL and DEFINE_REG (gpt) (#16673) 
* remove define_local and define_reg (gpt)

* fix precommit

* cleanups

* regalloc fix

* cleanups 2
 2026-06-19 10:07:50 -07:00
wozeparrotandGitHub 36f6d1b064 gemm: fix bf16 atb for mp sharding (#16637) 2026-06-16 15:58:47 -07:00
qazalandGitHub f998b9930a fp8 gemm inv_scale in epilogue (#16625) 
* fuse scale

* remove python inv_scale

* more inv_scale removal

* more cleanups

* cleaner

* diff polish

* work

* rename

* simpler

* simpler

* compute

* c

* Revert "c"

This reverts commit 8941fec7ca.

* Revert "compute"

This reverts commit 9db573a6d3.

* Revert "simpler"

This reverts commit 910ad33f87.

* Revert "simpler"

This reverts commit bf75d235a1.

* s_g

* update types

* less diff noise

* remove
 2026-06-15 18:44:41 +09:00
wozeparrotandGitHub 67a4f129c2 llama: fix bf16 gemm oob (#16603) 2026-06-12 19:43:05 -07:00
wozeparrotandGitHub e770805d21 llama: mxfp8 (#16574) 2026-06-11 22:15:24 -07:00
wozeparrotandGitHub c38d6a7e3a mxfp8 part 2 (#16561) 2026-06-10 23:36:11 -07:00
wozeparrotandGitHub 2bdc360606 gemm: mxfp8 hipkittens gemm (#16541) 
* gemm: mxfp8 hipkittens gemm

* feat: update hipkittens

* feat: kernel signature

* clean: just kernel

* feat: from tinygrad

* feat: test

* fix: add back utils

* clean: no diff

* clean: no diff
 2026-06-09 15:20:05 -07:00
wozeparrotandGitHub a1ec32cfd2 llama: current grad scaling (#16518) 2026-06-05 15:39:41 -07:00
qazalandGitHub 3b1a5f9770 llama: a_bT and aT_b bf16 gemms (#16487) 
* hk_bf16_gemm

* enable in 8b

* cleanups

* rename to USE_HK_BF16_GEMM

* work

* work

* work

* work

* change the gemms

* work

* work

* set as default

* work

* change
 2026-06-04 23:30:21 +09:00
wozeparrotandGitHub 7dcfd144b6 llama: columnwise fp8 scaling (#16480) 2026-06-02 18:55:45 -07:00
George HotzandGitHub ffadd7a315 remove intel and amx support (#16482) 2026-06-02 18:53:05 -07:00
George HotzandGitHub 58d58c1659 remove DEVECTORIZE (#16290) 
* remove DEVECTORIZE

* fully remove DEVECTORIZE
 2026-05-20 13:25:49 -07:00
qazalandGitHub bfb2d1f89a Revert "fp8 gemm speedup (#16236)" (#16245) 
This reverts commit d95bf394e1.
 2026-05-19 02:01:44 +09:00
qazalandGitHub d95bf394e1 fp8 gemm speedup (#16236) 
* add asm_gemm option

* milestone

* work

* edit

* only the fast kernel

* diff
 2026-05-17 04:58:28 +09:00
wozeparrotandGitHub e97f2c1114 llama: only gemm + fa custom kernel (#16180) 
* llama: tie store to grad directly

* llama: set mp flags

* llama: non fused grad fp8 quantize path
 2026-05-12 21:03:49 -07:00
wozeparrotandGitHub 730fa66bf3 llama speed 6 (#16071) 2026-05-06 20:51:03 -07:00
wozeparrotandGitHub 528d35e306 llama speed 4 (#15993) 2026-04-30 17:14:41 -07:00
nimlgenandGitHub dfd2d07005 remove CompiledRunner (#15970) 
* rm usage of CompiledRunner

* more tests

* last

* linter

* sink

* remove

* linter
 2026-04-29 22:45:48 +03:00
qazalandGitHub a37b605523 remove arch from asm kernel class (#15977) 
* rm arch from kernel

* update other tests

* update abstractions4.py
 2026-04-30 03:39:52 +09:00
wozeparrotandGitHub ef09071073 llama: speed 2 (#15960) 2026-04-28 20:44:37 -07:00
nimlgenandGitHub 77965a22e5 local optimize as rewrite (#15953) 
* local optimize as rewrite

* better

* x

* slighly rename

* fix

* ugh

* remove

* x

* remove

* not weak
 2026-04-28 22:51:04 +03:00
nimlgenandGitHub 4164666c72 programinfo (#15942) 
* programinfo

* fix

* m

* x

* x

* changes

* x

* fix

* rm
 2026-04-27 23:12:03 +03:00
nimlgenandGitHub bb652352c7 remove execitem (#15932) 
* remove execitem

* f

* x
 2026-04-25 19:33:04 +03:00
nimlgenandGitHub 768106a542 remove schedule from extra/docs/examples (#15929) 
* remove schedule from extra/docs/examples

* f
 2026-04-25 14:09:12 +03:00
nimlgenandGitHub f2751955cb remove linear_to_schedule from tests (#15912) 
* remove linear_to_schedule from tests

* x
 2026-04-24 20:02:10 +03:00
chenyuandGitHub 9192c93b7e Tensor.invalid -> Tesnor.invalids (#15849) 
matches ones and zeros, and to not share name with UOp.invalid
 2026-04-21 11:19:51 -04:00
nimlgenandGitHub bfe28ee2ad rm run_schedule (#15847) 2026-04-21 18:14:30 +03:00
wozeparrotandGitHub 9e60e4a7e7 llama: native fp8 (#15733) 2026-04-16 22:16:05 -07:00
qazalandGitHub 12c653a743 remove opts arg in get_program, everything uses opts_to_apply [pr] (#15767) 
* check Ops.BEAM in process replay

* remove opts from the get_program api

* lint

* simplify

* cleanup
 2026-04-16 22:42:43 +03:00
chenyuandGitHub 3394d18066 size*itemsize -> nbytes (#15729) 
and some UOp.size removal to prep for size to mixin change
 2026-04-14 16:27:54 -04:00
wozeparrotandGitHub 55bcd7cc9e llama amax outside (#15670) 2026-04-09 23:08:03 -07:00
George HotzandGitHub 48a7627b04 add RDNA4 support to copy WMMA (#15663) 
* add RDNA4 supportt to copy WMMA

* simpler

* simpler

* comment

* assert
 2026-04-09 22:48:20 +08:00
George HotzGitHubqazalqazal
 		1ebeb52e59 RDNA4 asm gemm (#15427) 
* sqtt: rdna4 decoder work

* diff cleanup

* more diff

* test

* 125

* r4

---------

Co-authored-by: qazal <[email protected]>
Co-authored-by: qazal <[email protected]>
 2026-04-08 21:26:44 +08:00
wozeparrotandGitHub 70dbd35023 llama: move custom_kernel into flat_llama (#15643) 2026-04-08 00:19:14 -07:00
wozeparrotGitHubqazal
 		7e54992bf6 fp8 llama (#15588) 
Co-authored-by: qazal <[email protected]>
 2026-04-04 18:24:57 -07:00
sirhcmandGitHub 0ed8d9271d Renderers accept Target or nothing (#15590) 2026-04-03 01:09:41 -04:00
qazalandGitHub fefb0ebc2a gemm/asm: fp8 cleanups (#15580) 
* normal gemm here

* s/dtypes.fp8e4m3/FP8_DTYPE

* gemm_bw

* device UOp stays NULL
 2026-04-02 19:02:38 +09:00
chenyuandGitHub 1aa04eab08 simple CreationMixin (#15567) 
start with full_like, zeros_like, ones_like
 2026-04-01 23:00:56 -04:00
qazalandGitHub 8feb8edc68 gemm/asm: add fp8 support to cdna asm_gemm (#15542) 
* work

* hmm, mixins

* rhs_transposed

* also fix the dtype

* check for hipcc

* Exception

* select dev

* default
 2026-03-31 19:32:54 +09:00
George HotzandGitHub 85dee83f5d amd flash attention cleanups + emulator fixes (#15431) 
* amd flash attention cleanups

* simpler

* params

* fix emulator bugs

* fix idiv bug

* remove that test

* more emu fixes
 2026-03-24 10:10:46 +08:00
George HotzGitHubAmpqazalchenyugg
 		c62dea6881 ai slop flash attention (it works) (#15401) 
* ai slop flash attention (it works)

* speed up, 2 TFLOPS + 7 GB/s

* simpler

* simpler

* optimize

* faster

* warp shuffle

* sqtt: link dispatch to exec (#15396)

* sqtt packet linking infra

python

* javascript

* ~doubly linked list

* ui works

* work

* exec can also highlight the pc, coloring work

* more work

* rm sqtt/model.py, doesn't need to be upstreamed

* viz: no context enters in cli, update llama profile (#15404)

* removed unused named arg in rules [pr] (#15414)

* viz: sqtt printer in viz/cli.py (#15411)

* work

* sqtt timeline in CLI

* format all printers nicely

* s/Showed/Printed

* ansistrip

* sys.exit

* keep colors in list

* work from amd_copy_matmul

* has_more always gets returned

* linter

* don't print colors

* more colors

* wow this is so deep

* work

* minor details

* selected

* improve progress bar

* remove it

* 22, global_load_vaddr is so long

* remove *0 hack in sign, gradient materializes zeros for unconnected nodes (#15416)

Amp-Thread-ID: https://ampcode.com/threads/T-019d1612-6322-706b-a94d-a812400a55cb

Co-authored-by: Amp <[email protected]>

* works

* cnt=20

* revert that

* uop slice tests

* simpler

---------

Co-authored-by: qazal <[email protected]>
Co-authored-by: chenyu <[email protected]>
Co-authored-by: gg <[email protected]>
Co-authored-by: Amp <[email protected]>
 2026-03-23 16:15:10 +08:00
George HotzandGitHub c13d9d29ff add SHAPED_WMMA (#15400) 
* add SHAPED_WMMA

* shaped wmma

* less bad
 2026-03-21 16:16:03 +08:00
George HotzandGitHub 41a9b09683 minimal vec in amd_copy_matmul (#15398) 
* minimal vec in amd_copy_matmul

* unified

* unify

* reshape/permute

* cleanups

* simpler

* move index

* cleanups

* more shared
 2026-03-21 14:57:21 +08:00
George HotzandGitHub 1a2a203f48 add wmma support to amd_copy_matmul (#15384) 
* add wmma support to amd_copy_matmul

* 15 TFLOPS and merged

* unify

* simpler

* simpler

* simpler

* cleanups

* TM/TN is the full regs

* comments

* WAVES_PER_SH + SQTT_EVENT

* Add WAVERDY support

* no split warp

* 3 range
 2026-03-20 19:02:19 +08:00