sirhcm/tinygrad
1
0
Fork 0
forked from tinygrad/tinygrad
Code Pull Requests 2 Actions 8 Activity
14,032 Commits 336 Branches 15 Tags
fcb1d3fc9f47ecb0720dcf63aee3595dd976e51c
Commit Graph
14032 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
RaineandGitHub fcb1d3fc9f add x86 loops (#17179) 
* add x86 loops

* lint
 2026-07-24 07:26:09 -07:00
wozeparrotandGitHub c9b60caf8c gptoss: moe gemm kernels (#17178) 2026-07-24 07:25:15 -07:00
chenyuandGitHub 40bfd84956 remove unneeded expand [PR] (#17176) 2026-07-24 10:03:46 -04:00
nimlgenandGitHub 017d1ab56e cpu prep for hcq2 (#17153) 
* cpu prep for hcq2

* x
 2026-07-24 15:57:21 +03:00
nimlgenandGitHub 0ecef210bb hcq2 cleanup 2 (#17177) 
* hcq uops

* x
 2026-07-24 13:04:21 +03:00
nimlgenandGitHub 709d7b2343 hcq2 cleanups (#17162) 2026-07-24 10:14:09 +03:00
chenyuandGitHub 2dd2e3d786 don't support Tensor(list(np.array)) [pr] (#17175) 2026-07-24 03:13:28 -04:00
George HotzandGitHub 08eceb0ad9 llm: fix generated tokens in usage accounting (#17174) 2026-07-23 21:38:07 -07:00
chenyuandGitHub a292dab8ab corner cases from weak const branch (#17173) 2026-07-24 00:09:56 -04:00
chenyuandGitHub f5d9c31d15 arange upcast to int64 with big N (#17167) 
simplifies _one_hot_along_dim
 2026-07-23 20:11:39 -04:00
George HotzandGitHub bf5989ea48 support loops in nir (kimi) (#17166) 
* support loops in nir

* tpying
 2026-07-23 17:05:15 -07:00
chenyuandGitHub 0f9edb02f6 don't use promo_lattice in fast_idiv [PR] (#17165) 
int64 will link to weakfloat
 2026-07-23 19:20:53 -04:00
chenyuandGitHub 3e031d6c4d AND OR XOR SHL SHR cannot have float operands [PR] (#17164) 
* AND OR XOR SHL SHR cannot have float operands [PR]

* dont
 2026-07-23 18:51:56 -04:00
chenyuandGitHub 1b616669d2 few more symbolic rules [pr] (#17161) 
bitcast twice can skip middle one; bool != False is self; two more where combine
 2026-07-23 18:05:32 -04:00
sirhcmandGitHub 6ee16eb1fa split out LLVM compilers (#17160) 2026-07-23 17:45:23 -04:00
chenyuandGitHub 00cf9c0c25 tweak add CONTIGUOUS to tagged UOps [PR] (#17159) 
dont call replace when the tag is alreayd None
 2026-07-23 17:06:56 -04:00
chenyuandGitHub 528aa4fc8c more stack const in gguf (#17157) 
better IQ4_XS and IQ3_XXS
 2026-07-23 14:04:04 -04:00
chenyuandGitHub 250de4b142 use stack for same shape cat [pr] (#17156) 
same device multi cat benefits too
 2026-07-23 13:50:28 -04:00
chenyuandGitHub 9267fca91a use STACK in wino (#17155) 
STACK is great
 2026-07-23 13:02:42 -04:00
chenyuandGitHub 2983987321 clean up some tests (#17154) 
duplicated or typo or slow
 2026-07-23 11:22:18 -04:00
chenyuandGitHub 33279b5d4c guard move add/mul consts to end [PR] (#17152) 
make sure the rules won't cycle
 2026-07-23 10:53:05 -04:00
wozeparrotandGitHub 2e7db75668 gptoss: faster swa (#17150) 2026-07-23 06:56:27 -07:00
nimlgenandGitHub 62273d50fc simpler hcq buffer (#17148) 
* simpleer hcq buffer

* x
 2026-07-23 13:28:51 +03:00
chenyuandGitHub c5b2b9242d python speed tweak for compile3 (#17147) 2026-07-23 02:02:44 -04:00
chenyuandGitHub e9a86c99ed more symmetrical x//d<c -> x<c*d [pr] (#17146) 
* more symmetrical x//d<c -> x<c*d [pr]

* parse_valid too
 2026-07-23 00:19:53 -04:00
chenyuandGitHub 6a9c23b1c1 more symmetrical div_and_mod_symbolic [pr] (#17145) 2026-07-22 23:34:03 -04:00
chenyuandGitHub 723309b5c0 update hand_coded_optimizations [pr] (#17144) 
only upcast when they share valids
 2026-07-22 22:26:27 -04:00
chenyuandGitHub 2aafca411d Revert "reorder sym in full_rewrite_to_sink [pr] (#17141)" (#17142) 
This reverts commit 7755480f46.
 2026-07-22 19:47:47 -04:00
chenyuandGitHub 7755480f46 reorder sym in full_rewrite_to_sink [pr] (#17141) 
* reorder sym in full_rewrite_to_sink [pr]

* fix
 2026-07-22 19:41:00 -04:00
chenyuandGitHub 855175123b fix z3 AND in validate (#17139) 2026-07-22 16:09:28 -04:00
chenyuandGitHub a7c693d2fd rework pm_lower_weakint [pr] (#17136) 
deleted pm_no_index and many "remove hanging casts"
 2026-07-22 15:53:11 -04:00
chenyuandGitHub 6b82c0cb95 minor test fixes (#17138) 
TestCustomKernel square().max() masks NaN, and fixed test_tensor_cores_multi_reduce
 2026-07-22 15:08:52 -04:00
sirhcmandGitHub c7027db715 lvp: skip test_masked_select (#17137) 2026-07-22 14:53:09 -04:00
chenyuandGitHub b9fa7e519c remove dead Metadata.caller (#17135) 2026-07-22 12:28:56 -04:00
nimlgenandGitHub d1f215d377 cpu: uop worker (#17133) 
* cpu: uop worker

* x

* x

* Dx

* amd only

* Dx

* x

* x
 2026-07-22 19:01:08 +03:00
chenyuandGitHub 557e674861 delete pm_index_invalid [pr] (#17130) 2026-07-22 11:50:08 -04:00
chenyuandGitHub 17557d7fdf bring back fold_where_closure [pr] (#17134) 
* bring back fold_where_closure [pr]

generic enough to delete delete pm_index_invalid

* cannot cache self
 2026-07-22 11:08:14 -04:00
wozeparrotandGitHub 5f2eaeee40 gptoss: rope freqs in fp32 (#17132) 2026-07-22 06:11:06 -07:00
nimlgenandGitHub a6fda6b102 add volatile to paramarg (#17127) 
* volatile

* x

* x

* fo
 2026-07-22 15:47:37 +03:00
nimlgenandGitHub 6e979b879b call inside c (#17131) 2026-07-22 13:49:06 +03:00
George HotzandGitHub 39924387b1 LOOP is srcless RANGE (kimi) (#17129) 
* LOOP is srcless RANGE (kimi)

* upd

* cleanups

* cleanups

* we don't float anymore
 2026-07-21 18:41:31 -07:00
chenyuandGitHub 9433790adb move _broadcasted to ElementwiseMixin [PR] (#17128) 2026-07-21 19:42:07 -04:00
chenyuandGitHub 92f9c850b4 fix pow(int, float) (#17126) 
* fix pow(int, float)

* onnx
 2026-07-21 18:48:24 -04:00
chenyuandGitHub b1060ca708 don't promote dtype in _pad_constant [pr] (#17125) 2026-07-21 18:13:50 -04:00
chenyuandGitHub b1a72299ab more _broadcast_to cleanup [PR] (#17124) 2026-07-21 17:46:04 -04:00
sirhcmandGitHub 8fa5993923 replace pytest-timeout with SIGABRT hook (#17122) 2026-07-21 17:42:58 -04:00
nimlgenandGitHub f41e4a758f drop Ops.WAIT (#17121) 2026-07-22 00:42:45 +03:00
chenyuandGitHub f19a2ad771 single where mixin [pr] (#17118) 
* single where mixin [pr]

no shape broadcasting in ufix and _broadcasted anymore

* QCOM vectorized bool is broken
 2026-07-21 17:38:36 -04:00
nimlgenandGitHub 787b2f2db2 hcq2: use ins for hcq ir (#17120) 2026-07-21 23:42:34 +03:00
Armand du Parc LocmariaandGitHub ef37830d13 allow freeing buffers when pickling/unpickling (#16799) 
* allow pickling out of band buffers

* also need to release when loading

* test peak ram

* lint

* sync before yielding next buffer for backends with async copy in

* skip on mock devices

* reason

* or always bytearray, always free?

* Revert "or always bytearray, always free?"

This reverts commit a017bb68742985a5b7431e0b4e973c2997c92b6a.

* one less copy
 2026-07-21 16:11:23 -04:00