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
12,776 Commits 248 Branches 14 Tags
fdb30cba96d0f02e471a98bbf0efa4e419094559
Commit Graph

119 Commits

Author SHA1 Message Date
Christopher Milan
fdb30cba96 DEV is a ContextVar (#15505) 2026-03-27 00:57:09 -04:00
Christopher Milan
67a50fb738 move where on load with casts (#15492) 2026-03-26 22:11:27 -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
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
chenyu
713b322e70 add weakint to promo_lattice (#15463) 
sits between bool and smallest int
 2026-03-25 00:27:34 -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
qazal
652bab8aad viz: support nested track_rewrites (#15454) 
* simple test

* stack active groups
 2026-03-25 05:01:30 +09:00
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
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
nimlgen
fa4cdb422e memplan on linears (#15422) 
* memplan

* test

* x

* arenas

* correct

* set any size

* ugh

* make hevc happy

* x

* x

* held

* rm old

* del

* x

* fu

* f

* cl

* cl

* ok
 2026-03-23 19:50:16 +08:00
George Hotz
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 <amp@ampcode.com>

* works

* cnt=20

* revert that

* uop slice tests

* simpler

---------

Co-authored-by: qazal <77887910+Qazalin@users.noreply.github.com>
Co-authored-by: chenyu <chenyu@fastmail.com>
Co-authored-by: gg <ggordbegli@gmail.com>
Co-authored-by: Amp <amp@ampcode.com>
 2026-03-23 16:15:10 +08:00
nimlgen
9656d97d97 jit: captures linears, not execitems (#15399) 
* jit: captures linears, not execitems

* x

* um

* etsts

* mockcuda
 2026-03-21 16:32:12 +08:00
Christopher Milan
1560b534a5 remove IMAGE=2 (#15312) 2026-03-20 06:26:52 -04:00
Christopher Milan
0c89340a1e automatically emulate unsupported (tiny) floats [skip_process_replay] (#15366) 2026-03-20 02:31:44 -04:00
chenyu
da1700e16b dtypes.index -> dtypes.weakint (#15377) 2026-03-20 01:08:46 -04:00
nimlgen
86eec01f97 limit gl*lc (#15359) 2026-03-19 12:38:55 +08:00
nimlgen
d4836ddbb0 canonicalize device from tuple (#15348) 
* will it ifx ci?

* test

* um
 2026-03-18 20:35:52 +08:00
George Hotz
5524916e39 llama compute gradients explicitly + 243 GB of RAM on MP=8 (#15343) 
* llama compute gradients explicitly

* apply grads

* fix multi issue

* multi BUFFER_VIEW support

* simpler

* skip the flaky test
 2026-03-18 19:54:40 +08:00
nimlgen
f853371c83 fix compilers autoselect (#15346) 2026-03-18 18:19:53 +08:00
chenyu
94926d00d8 fix rand > uint32.max (#15330) 
need to keep low and high as 1D tensor.
`PYTHONPATH=. LLAMA3_SIZE=405B python3 examples/mlperf/models/flat_llama.py` works now
 2026-03-17 22:00:01 -04:00
qazal
00817cf65e viz: all tests can run on the NULL device (#15328) 
* remove that

* move to test_viz

* get_cfg

* do not use os.environ

* hm

* it's always on NULL

* import renderer

* no import *
 2026-03-18 04:14:20 +09:00
chenyu
14eb8170e4 skip TestRunAsModule if libclang is loaded (#15323) 
reverse rule of TestAutogen skip, otherwise `NULL=1 python -m pytest test/null/test_autogen.py test/null/test_device.py` crashes for me
 2026-03-17 06:02:53 -04:00
Christopher Milan
9047249a7c m.where(x.pad_to(m.shape), Invalid) ranges shrink (#15275) 2026-03-14 07:26:36 -04:00
Christopher Milan
dabdc986df shrink guarded ranges, try 2 (#15272) 2026-03-14 04:24:05 -04:00
Christopher Milan
7cf4b16c91 Revert "shrink guarded ranges" (#15271) 2026-03-14 03:44:38 -04:00
Christopher Milan
d9951e2f8e shrink guarded ranges (#15263) 2026-03-14 03:38:48 -04:00
chenyu
90b7f4341d failed two level divmod recombine case (#15233) 2026-03-12 04:04:36 -04:00
chenyu
842c978df3 remove staticmethod dtypes.max/min (#15227) 
always use x.dtype.max/min
 2026-03-11 23:11:24 -04:00
chenyu
fce87f19a8 better fold_add_divmod_recombine (#15214) 2026-03-10 23:24:22 -04:00
chenyu
df8deec949 test for nest_by_factor selection (#15213) 2026-03-10 22:41:31 -04:00
chenyu
be6b0bce1f variations of (x%c)+(x//c)*c (#15212) 
put those into one function
 2026-03-10 22:41:14 -04:00
chenyu
8389a8d7c5 remove_nested_mod can work with negative (#15205) 2026-03-10 03:10:08 -04:00
Christopher Milan
ffaafd391a Invalid in Tensor (#15154) 2026-03-10 02:49:54 -04:00
chenyu
68c7c3ca84 divmod test_gcd_with_remainder (#15204) 
test cases for gcd_with_remainder
 2026-03-09 23:51:47 -04:00
chenyu
60215deb60 tiebreak in fold_divmod_congruence (#15190) 
need to try both direction
 2026-03-09 03:40:39 -04:00
chenyu
a8d8351e5a match IDIV and MOD in nest_by_factor (#15188) 2026-03-09 00:50:38 -04:00
chenyu
83b80da8f3 even more divmod recombine (#15163) 2026-03-08 23:52:26 -04:00
qazal
25e82a9aca viz: exclude redundant traceback from SDMA (#15185) 
* viz: exclude redundant traceback from SDMA

* ctx

* cpu_profile
 2026-03-09 05:12:14 +09:00
nimlgen
6ac99fd4c9 memplanner opt copy bufs (#15110) 
* mtp

* x

* tests

* ss

* simp

* less slop

* x

* cleaner

* rm

* m

* c

* x

* f
 2026-03-08 22:28:01 +03:00
Roelof van Dijk
4ed8bb7445 tie break for divmod (#15169) 2026-03-06 08:05:38 -05:00
George Hotz
6fd18ef875 rename CAT to VCAT (#15167) 2026-03-06 18:46:28 +08:00
chenyu
da61088ca4 more divmod recombine (#15162) 2026-03-05 12:53:22 -05:00
chenyu
167a1d56a6 improve divmod folding (#15148) 
canonicalize to div than mod which enables more simplifcation
 2026-03-05 10:07:36 -05:00
qazal
5bf542469d viz: python traceback for USER device (#15160) 
* start

* ux

* unittests
 2026-03-05 20:22:09 +09:00
George Hotz
8a82b26522 llm: print the prefill cache size (#15146) 
* print the llm prefill cache size

* mock that too
 2026-03-05 12:13:28 +08:00
George Hotz
72a9ed6e23 fix render depth bug + add warmup to serve + no realize default (#15144) 
* fix render depth bug + add warmup to serve

* make realize not the default
 2026-03-05 11:21:16 +08:00
chenyu
04da527a7a minor div_and_mod_symbolic cleanups (#15138) 2026-03-04 19:05:44 -05:00
chenyu
4cce283790 relax test_tqdm_perf (#15134) 2026-03-04 12:58:47 -05:00
Christopher Milan
592f9bf6c6 set OPENPILOT_HACKS=1 to enable replace assign (#15123) 2026-03-04 05:26:04 -05:00