github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-04-29 03:00:14 -04:00
Code Issues Packages Projects Releases Wiki Activity
12,781 Commits 248 Branches 14 Tags
0d6fc0f571dfb02ca5fc7480a916ad2a309411fe
Commit Graph

12781 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
nimlgen
0d6fc0f571 jit: graphing in uops (#15489) 
* jit: graphing as rewrite rule

* f

* +metal,cuda

* x

* cl

* x

* x

* simpler

* f

* m

* x

* revert?

* revert2

* back

* back

* t

* x

* m

* x

* c

* x

* l

* x

* comment

* smaller

* rv

* x

* x
 2026-03-27 19:09:02 +03:00
chenyu
30ebbe7f17 few more fold valid tests (#15509) 
from remove CORRECT_DIVMOD_FOLDING attempt
 2026-03-27 10:38:42 -04:00
Christopher Milan
9e0cc5c6ae create image buffers in late codegen (#15493) 2026-03-27 04:50:53 -04:00
chenyu
1198d6e908 move pow to mixin (#15507) 2026-03-27 03:16:40 -04:00
chenyu
323fcefd7d Revert "DEV is a ContextVar (#15505)" (#15506) 
This reverts commit fdb30cba96.
 2026-03-27 02:22:40 -04:00
Christopher Milan
fdb30cba96 DEV is a ContextVar (#15505) 2026-03-27 00:57:09 -04:00
wozeparrot
a65e958be9 llama: new apply_grad (#15503) 2026-03-26 19:39:25 -07:00
Christopher Milan
67a50fb738 move where on load with casts (#15492) 2026-03-26 22:11:27 -04:00
qazal
586c49642f viz/cli: test in CI (#15501) 
* viz cli work

* baseline test

* make cli test work without subprocess

* more checks

* check itrace

* s/return/return None

* change

* minimal

* colored
 2026-03-27 06:47:15 +09:00
qazal
3f9f0fa846 viz: yield sqtt alt events (#15500) 
* yield other

* less

* work

* less
 2026-03-27 04:43:41 +09:00
qazal
237c25031f sqtt: construct OTHER_SIMD op types with for loop (#15495) 
* other-lds from amd_copy_matmul

* more other

* other simd work
 2026-03-26 23:07:18 +09:00
nimlgen
7193f90746 test view input in jit (#15497) 
* will anything fail?

* add test
 2026-03-26 16:59:47 +03:00
nimlgen
de24b3fe37 jit: pass init params straight to base (#15496) 
* jit: pass init params straight to base

* linter
 2026-03-26 16:59:10 +03:00
qazal
ec5b7a249e viz: refactor sqtt timeline builder (#15494) 
* viz: refactor sqtt timeline builder

* barrier maps to waves

* clean up cli
 2026-03-26 21:16:15 +09:00
Christopher Milan
313937ad6d fix IMAGE TestEnd2End.test_linear_mnist (#15488) 2026-03-26 04:12:47 -04:00
Christopher Milan
bc180a963c deprecate <dev>=1 in favor of DEV=<dev> (#15467) 
* start work on target

* add test

* update actions to use DEV

* update docs

* update readmes

* tests need that too

* update example

* update tests (comments)

* fix that test

* ruff

* mypy

* oops

* remove getenvs

* don't add Target yet

* and the test

* lint

* and docs

* more stuff

* assert

* few more fixes

* test assert
 2026-03-26 03:48:03 -04:00
chenyu
8426f820a1 Tensor.sub to mixin (#15486) 
also _broadcasted skipped broadcasting shape if it does not have shape
 2026-03-25 23:20:56 -04:00
wozeparrot
1ca178f379 llama: stochastic rounding (#15456) 2026-03-25 18:16:31 -07:00
chenyu
7c8f992894 move EXPAND dtype cast back to gradient.py (#15481) 
only a concern for gradient, not mixin
 2026-03-25 19:25:26 -04:00
nimlgen
9d2d0774b4 remote: disk copies (#15482) 
* remote: disk copies

* lineter

* r

* nv

* x
 2026-03-25 22:14:25 +03:00
qazal
7c2c8d3905 viz: small ux improvements (#15483) 
* test

* better

* work
 2026-03-26 03:18:25 +09:00
qazal
737d5f67f9 viz: compute canvas dims for auto zoom (#15474) 2026-03-26 00:05:23 +09:00
qazal
60bd546593 sqtt: add cycle count to rdna3 enums (#15473) 
* update rdna3 sqtt enums to include cycle_count

* dispatch_to_exec
 2026-03-25 23:19:54 +09:00
chenyu
142bf11926 logical_not to mixin [pr] (#15472) 
also UPat.cast skips same dtype
 2026-03-25 09:16:45 -04:00
George Hotz
25ff7146f2 add a status line to REMOTE with DEBUG=1 (#15471) 
* python speedups of hot paths

* add a status line to REMOTE with DEBUG=1

* pc

* t
 2026-03-25 20:54:56 +08:00
qazal
c973b508b8 viz/cli: pass ctrlc (#15470) 2026-03-25 21:13:28 +09:00
George Hotz
c1a7d90ccc python speedups of hot paths (#15469) 2026-03-25 20:02:42 +08:00
George Hotz
ae7090b13b print function timing with DEBUG=2 (#15468) 
* add DEBUG=2 function timing

* remove those functions, they aren't useful

* fix spec
 2026-03-25 19:07:32 +08:00
Christopher Milan
e7f389efda fix height=1 images on macos (#15460) 2026-03-25 05:59:56 -04:00
George Hotz
789628df2e hotfix: add USE_BOT flag to ASM24 USB 2026-03-25 15:00:08 +08:00
George Hotz
cd1a276f47 llm: support gguf path or url (#15464) 
* llm: support gguf path or url

* one line
 2026-03-25 14:43:19 +08:00
chenyu
713b322e70 add weakint to promo_lattice (#15463) 
sits between bool and smallest int
 2026-03-25 00:27:34 -04:00
chenyu
02878c5a2f move _broadcasted to OpMixin (#15461) 
it needs both ElementwiseMixin and MovementMixin
 2026-03-24 23:56:01 -04:00
chenyu
519ba22470 more Tensor._broadcasted cleanup (#15459) 
prep moving to mixin
 2026-03-24 22:55:45 -04:00
George Hotz
fe2690399b llm: support assistant prefill + refactor to TransformerConfig (#15457) 
* llm: support assistant prefill

* refactor to ModelConfig

* TransformerConfig

* more
 2026-03-25 10:50:48 +08:00
Christopher Milan
fd92aec094 cleanup unused image pitch code (#15458) 2026-03-24 22:47:16 -04:00
chenyu
f6ed4da268 Tensor.ufix (#15452) 
* Tensor.ufix

prep moving _broadcasted to mixin

* remove backward_cast
 2026-03-24 22:34:43 -04:00
qazal
1b3d00d6ac viz/cli: remove --offset and --limit flags (#15439) 
* work

* also no more no-color

* reorder

* update llama

* sqtt readme

* itertools

* rm that

* signals back
 2026-03-25 09:52:27 +09:00
wozeparrot
da2031266a llama: correct 8b init (#15397) 2026-03-24 13:41:41 -07:00
qazal
652bab8aad viz: support nested track_rewrites (#15454) 
* simple test

* stack active groups
 2026-03-25 05:01:30 +09:00
qazal
41eb2cc41b viz: preserve zoom between re renders (#15451) 2026-03-25 03:11:10 +09:00
Salman Chishti
84049fdc07 Upgrade GitHub Actions to latest versions (#15446) 
Signed-off-by: Salman Muin Kayser Chishti <13schishti@gmail.com>
Co-authored-by: chenyu <chenyu@fastmail.com>
 2026-03-24 10:28:49 -04:00
Salman Chishti
9567075e20 Upgrade GitHub Actions for Node 24 compatibility (#15445) 
Signed-off-by: Salman Muin Kayser Chishti <13schishti@gmail.com>
Co-authored-by: chenyu <chenyu@fastmail.com>
 2026-03-24 10:28:19 -04:00
chenyu
b7960841af support shape broadcast in UOp.alu (#15442) 
i think it can integrate tighter, but now Tensor also does ufix from UOp and implicit dtype upcast
 2026-03-24 10:14:57 -04:00
George Hotz
a33ac869aa llm server: temperature + test client (#15444) 
* improvements to the llm server

* eval script

* eval llm

* better eval gets 58.71

* cleanups

* add temperature, but multinomial is absurdly slow

* claude is so smart

* lint

* remove slop

* no more stop
 2026-03-24 21:07:15 +08:00
nimlgen
9db5d677c7 jit in viz (#15447) 2026-03-24 18:23:53 +08:00
Christopher Milan
2e4fbbcc9c ir3: fix texture mapping and benchmark (#15443) 2026-03-24 04:52:54 -04:00
Christopher Milan
d5320a9ddf QCOM cleanups (#15435) 2026-03-23 22:18:38 -04:00
George Hotz
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
chenyu
018a9e2d3c remove match_dtype arg in Tensor._broadcasted (#15440) 
reworked Tensor.where to not need it, also updated dtypes.from_py to use isinstance because ConstFloat issues
 2026-03-23 22:10:39 -04:00