Commit Graph

Select branches

Hide Pull Requests

geohot/updates

gh-pages

master

remove_anchors

#1

#2

#3

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

• e14cadb1fb gptoss: set ASM_GEMM (#17363) wozeparrotandGitHub 2026-08-02 21:52:36 +08:00
• 0258c7fefc minor argstr and alloc stack cleanup [PR] (#17361) chenyuandGitHub 2026-08-01 22:37:06 -04:00
• fb607fb990 faster devectorizer with one line (#17360) George HotzandGitHub 2026-08-01 13:06:35 -07:00
• 15c936db01 merge devectorize + indexing (#17354) George HotzandGitHub 2026-08-01 11:40:15 -07:00
• 98b700bad1 gptoss optim fixes (#17356) wozeparrotandGitHub 2026-08-02 00:10:06 +08:00
• 8e524ca467 CONST related cleanups [pr] (#17352) chenyuandGitHub 2026-08-01 09:54:28 -04:00
• 665822ab34 hcq2: faster replace (#17353) nimlgenandGitHub 2026-08-01 16:32:04 +03:00
• 6c0ec39279 UOp.is_invalid [PR] (#17351) chenyuandGitHub 2026-08-01 02:36:32 -04:00
• b502fc1367 more const arg -> val (#17350) chenyuandGitHub 2026-08-01 02:18:10 -04:00
• 161783d8f7 add _device_num back to ast.variables (kimi) (#17327) qazalandGitHub 2026-08-01 13:56:50 +08:00
• 5a1c641f79 more arg -> val (#17349) George HotzandGitHub 2026-07-31 22:44:24 -07:00
• 20b8ecff50 support new USB vendor ID (#17348) Adeeb ShihadehandGitHub 2026-07-31 21:45:06 -07:00
• 9082ecef5d use .val to access the value of Ops.CONST (#17347) George HotzandGitHub 2026-07-31 19:51:42 -07:00
• 099d69ff7d ci: split macos unit test into metal and mock runners (#17346) George HotzandGitHub 2026-07-31 19:41:18 -07:00
• a88f832f0c remove UOp.val (#17345) George HotzandGitHub 2026-07-31 18:58:38 -07:00
• 850989115d __int__ and __float__ work for weak (#17342) chenyuandGitHub 2026-07-31 19:56:37 -04:00
• 15d515299e heuristics: try multiple TC axes (#17341) sirhcmandGitHub 2026-07-31 19:25:18 -04:00
• 85ced44db6 tc: don't allow reduce over output dims (#17340) sirhcmandGitHub 2026-07-31 16:48:51 -04:00
• 277433259e fix sym_infer for CAST (#17338) chenyuandGitHub 2026-07-31 14:38:38 -04:00
• 8dc225e28e dtype_from_uop(INS) is None [PR] (#17337) chenyuandGitHub 2026-07-31 14:33:15 -04:00
• ad24750487 const(value, dtype) -> const(value).cast(dtype) in tests (#17335) chenyuandGitHub 2026-07-31 13:24:31 -04:00
• a11ee26bb8 viz: prep for faster cli DEBUG=3 (#17334) qazalandGitHub 2026-08-01 01:16:30 +08:00
• b95bd5b2a5 don't reset chestnut in benchmark (#17333) sirhcmandGitHub 2026-07-31 12:39:04 -04:00
• 1095bbe409 hcq2: fix ib reuse (#17330) nimlgenandGitHub 2026-07-31 18:43:23 +03:00
• 7f4dbb8090 remove shape= from UOp.const [PR] (#17331) chenyuandGitHub 2026-07-31 11:18:59 -04:00
• 8e2f175542 const(dtype, b) -> const(b, dtype) [PR] (#17328) chenyuandGitHub 2026-07-31 09:46:37 -04:00
• 155b84ee80 hcq2 faster schedule (#17324) nimlgenandGitHub 2026-07-31 16:08:18 +03:00
• 4f5cadd15d gptoss ci (#17325) wozeparrotandGitHub 2026-07-31 20:51:58 +08:00
• 0c4bfaeb48 coalesce ints (#17323) nimlgenandGitHub 2026-07-31 15:49:56 +03:00
• 6d2700f0b7 failing test for unbound _device_num err in BEAM (#17326) qazalandGitHub 2026-07-31 19:24:44 +08:00
• 93b74c75fc gptoss: grouped moe (#17322) wozeparrotandGitHub 2026-07-31 18:25:32 +08:00
• f7964acb64 llama with MXFP4 (#17321) qazalandGitHub 2026-07-31 17:25:25 +08:00
• 0a3325f9c2 add mxfp4 quantize and layout kernels (#17320) qazalandGitHub 2026-07-31 13:50:18 +08:00
• 13452b3775 benchmark comma big model (#17312) sirhcmandGitHub 2026-07-31 01:37:34 -04:00
• a8c1e89500 fp4 asm gemm 6+ pflops (#17315) qazalandGitHub 2026-07-31 13:25:23 +08:00
• d65ea465ed cleanup gemm fragment + add store unshard (#17313) George HotzandGitHub 2026-07-30 20:53:28 -07:00
• 6608b9d8e2 adjust lower weak order [pr] (#17314) chenyuandGitHub 2026-07-30 23:36:12 -04:00
• f2c2f4456b Add softmin (#17292) Noah SchiroandGitHub 2026-07-30 20:05:11 -04:00
• fe8ece7efa fix weak for image gate fusion [pr] (#17310) chenyuandGitHub 2026-07-30 19:47:47 -04:00
• ce500c1946 broadcast doesn't cast CONST [pr] (#17308) chenyuandGitHub 2026-07-30 18:51:59 -04:00
• b290372121 ftdi reset chestnut before running comma benchmark (#17306) sirhcmandGitHub 2026-07-30 14:28:45 -04:00
• da15c43e51 update openpilot benchmarks (#17304) sirhcmandGitHub 2026-07-30 14:22:20 -04:00
• b488cc7df2 update dtype_from_uop for SHR/SHL (#17307) chenyuandGitHub 2026-07-30 13:57:58 -04:00
• 341c4ed4f5 rewrite writes new dtype with dtype_from_uop try 2 [pr] (#17305) chenyuandGitHub 2026-07-30 13:47:21 -04:00
• d05a3e6c0b Revert "rewrite writes new dtype with dtype_from_uop [pr] (#17302)" (#17303) chenyuandGitHub 2026-07-30 13:08:26 -04:00
• 417245ab26 rewrite writes new dtype with dtype_from_uop [pr] (#17302) chenyuandGitHub 2026-07-30 12:46:24 -04:00
• e25bf77ce9 kimi delta attention (#17281) b1tgandGitHub 2026-07-30 23:23:56 +08:00
• b5a2a5666a gptoss moe routing (#17284) wozeparrotandGitHub 2026-07-30 22:49:14 +08:00
• 3f6f0a19ab hcq2: less graph_rewrites (#17300) nimlgenandGitHub 2026-07-30 16:44:45 +03:00
• 7b7b9c1ae0 spec tests for movement ops before custom_kernel (#17299) qazalandGitHub 2026-07-30 18:20:24 +08:00
• aba5ba4b40 update rangeify comment for mop after AFTER (#17298) qazalandGitHub 2026-07-30 16:09:32 +08:00
• d52ef3077c fix do_devectorize dtype (#17295) chenyuandGitHub 2026-07-29 23:04:25 -04:00
• bfc9fc6e0a nicer TinyJit decorator (#17293) sirhcmandGitHub 2026-07-29 21:48:00 -04:00
• 060f447db6 qcom: match cl for SP_CS_INSTR_SIZE (#17289) sirhcmandGitHub 2026-07-29 18:47:02 -04:00
• fd912b348c generic c0*x<c1 [pr] (#17291) chenyuandGitHub 2026-07-29 18:45:23 -04:00
• 138676ab81 improve fragment example + index unshard (kimi) (#17288) George HotzandGitHub 2026-07-29 15:38:38 -07:00
• 027907a544 fix c0*x<c1 symbolic [pr] (#17290) chenyuandGitHub 2026-07-29 18:32:31 -04:00
• d4ba8b6e0f hcq2: use stack (#17286) nimlgenandGitHub 2026-07-29 22:09:34 +03:00
• b30c7e00d4 support 2d on UNSHARD (kimi) (#17285) George HotzandGitHub 2026-07-29 12:01:59 -07:00
• aab51fb7b6 64-bit UOp.variable support, try 2 (#17256) sirhcmandGitHub 2026-07-29 14:39:36 -04:00
• dd86a30798 test case for weakfloat cast to weakint INDEX (#17287) chenyuandGitHub 2026-07-29 14:15:52 -04:00
• 52c9e5a99e rename LOOP -> WEAK and STRONGLOOP -> LOOP (#17283) George HotzandGitHub 2026-07-29 10:38:36 -07:00
• 6c2b9fac08 weak const cleanups [PR] (#17282) chenyuandGitHub 2026-07-29 13:14:20 -04:00
• bd296a7359 enable alloc_fragment support with UNSHARD (kimi) (#17272) George HotzandGitHub 2026-07-29 09:46:45 -07:00
• 3df1b07c86 Patch half precision ops for older gpu archs (#17274) Noah SchiroandGitHub 2026-07-29 12:42:24 -04:00
• 3803f1583b fix add_raw_barrier [pr] (#17278) chenyuandGitHub 2026-07-29 12:30:00 -04:00
• 6ea7d366fa test permuted input in custom Ops.PROGRAM test (#17279) qazalandGitHub 2026-07-29 17:29:05 +08:00
• 527e57300c fix smu reset for kernel >= 7 (#17277) George HotzandGitHub 2026-07-28 23:11:21 -07:00
• e684fcc689 pm_reduce_collapse fix for re enabling stack for cat of same shape (codex) (#17251) qazalandGitHub 2026-07-29 13:47:28 +08:00
• 451120c6e1 make .barrier implicit (kimi) (#17275) George HotzandGitHub 2026-07-28 22:34:57 -07:00
• 2f8f2d2d37 remove no-op explicit dtype= or cast [PR] (#17276) chenyuandGitHub 2026-07-29 00:47:35 -04:00
• dd16d5aead apply shrink bugfix for 3.11 (#17271) George HotzandGitHub 2026-07-28 19:03:06 -07:00
• 291ee435bb qwen3.6 for 27b and 35b-a3b (#17268) chenyuandGitHub 2026-07-28 21:11:49 -04:00
• a17387d6a0 add UNSHARD to spec (#17269) George HotzandGitHub 2026-07-28 17:00:21 -07:00
• 57ae1bc7a7 rename MULTI to UNSHARD (#17267) George HotzandGitHub 2026-07-28 16:51:41 -07:00
• 1757067fa6 add device range as src[1] to multi (kimi) (#17264) George HotzandGitHub 2026-07-28 16:39:34 -07:00
• f11f884f49 update dtypes.md (#17266) chenyuandGitHub 2026-07-28 19:30:13 -04:00
• 23e9e76e8c DEFAULT_FLOAT/DEFAULT_INT ContextVar [pr] (#17265) chenyuandGitHub 2026-07-28 19:08:18 -04:00
• 755dfb243b rename CPU_COUNT to NUM_CPU_THREADS with cgroup awareness (#17263) 3 people 2026-07-28 15:33:57 -07:00
• 0cdddf3495 remove unneeded default args in renderers (#17262) George HotzandGitHub 2026-07-28 14:53:04 -07:00
• fde3a8faff CAPTURE_PROCESS_REPLAY=0 default chaging test [PR] (#17261) chenyuandGitHub 2026-07-28 17:32:35 -04:00
• 9ce65b71c0 delete dsp_pm_late (#17260) George HotzandGitHub 2026-07-28 14:05:41 -07:00
• 7d48926293 switch _device_num to AxisType.DEVICE range (kimi) (#17252) George HotzandGitHub 2026-07-28 13:52:47 -07:00
• d1c3ae0f91 preserve typed ranges through reshape (#17259) drkandGitHub 2026-07-28 13:51:06 -07:00
• 749e002286 remove some dtype= when construct UOp [pr] (#17257) chenyuandGitHub 2026-07-28 16:05:35 -04:00
• c9e11544df delete explicit casts [pr] (#17255) chenyuandGitHub 2026-07-28 13:57:13 -04:00
• a9ad08064e make the github actions runners generic for gitea (#17254) George HotzandGitHub 2026-07-28 10:48:49 -07:00
• 37cf15942b UOp.const(dtype=None) infers from from_py [pr] (#17253) chenyuandGitHub 2026-07-28 12:06:55 -04:00
• 05e4727feb arange stack regression test (#17250) qazalandGitHub 2026-07-28 19:00:22 +08:00
• e3b3eea1e2 test for extra copy in allreduce_cast (#16740) qazalandGitHub 2026-07-28 18:02:54 +08:00
• 1380d6cc5a pm_long_decomp doesn't depend on operand dtype [pr] (#17249) chenyuandGitHub 2026-07-28 01:15:04 -04:00
• 4b7022e8f4 Revert "64-bit UOp.variable support (#17246)" (#17248) chenyuandGitHub 2026-07-28 00:00:26 -04:00
• ab8fb191b2 64-bit UOp.variable support (#17246) sirhcmandGitHub 2026-07-27 23:56:09 -04:00
• f837ca3587 don't match const dtype in UPat [pr] (#17247) chenyuandGitHub 2026-07-27 22:50:08 -04:00
• 654d475d37 update pm_long_decomp [pr] (#17245) chenyuandGitHub 2026-07-27 22:28:50 -04:00
• 550225f603 clean up pm_lower_weak [pr] (#17243) chenyuandGitHub 2026-07-27 19:11:39 -04:00
• 0bb36c9989 make Tensor(None) weakfloat (#17241) chenyuandGitHub 2026-07-27 16:45:28 -04:00
• 896afad9bf don't match strong typed const in UPat [pr] (#17240) chenyuandGitHub 2026-07-27 16:19:01 -04:00
• 8b9ef157d1 run_linear in external_test_gpu_crash (#17239) nimlgenandGitHub 2026-07-27 22:00:11 +03:00
• bdbb1d702f fix shard axis through symbolic reshape (#17238) b1tgandGitHub 2026-07-27 23:32:15 +08:00