Sign in
apache
/
tvm
/
HEAD
4ab91d4
Use adapter.info when available instead of requestAdapterInfo (#17051)
by François Beaufort
· 8 hours ago
main
31f4721
[Runtime] Support PagedKVCache with tree attention (#17049)
by Ruihang Lai
· 8 hours ago
515c079
[DLight] Skip GEMV rules when more than one vector (#17052)
by Siyuan Feng
· 29 hours ago
nightly
1eac178
[Runtime] Fix PagedKVCache for PopN and enhance tests (#17045)
by Ruihang Lai
· 2 days ago
820f1b6
[Runtime] Compatibility with dmlc::Stream API changes (#16998)
by Eric Lunderberg
· 2 days ago
7c2c0d9
[Disco][QoL] Implement broadcast/scatter methods for Session (#17035)
by Eric Lunderberg
· 2 days ago
f6aab98
[Bugfix][Support] Fix copy constructor for support::OrderedSet (#17044)
by Eric Lunderberg
· 2 days ago
08b32a7
[Runtime][ROCm] Enable ROCm host memory support (#17037)
by Ruihang Lai
· 2 days ago
291c047
[TIR] Fix Bug in VectorizeLoop (#17039)
by Charlie Ruan
· 2 days ago
71f7af7
[Runtime] Use preferred host memory (pinned memory) in KV cache (#17036)
by Ruihang Lai
· 3 days ago
8bdd54b
[TOPI] Fix SME conv2d schedule import and intrin argument (#17040)
by Luke Hutton
· 3 days ago
d9240e4
[Relax][Bugfix] Apply FuseOps to nested DataflowBlock (#17033)
by Eric Lunderberg
· 3 days ago
7afac14
[BugFix][MSC] split name_string with index by colon from the right (#17000)
by Peng Sun
· 3 days ago
c9d87ef
[Relax][Bugfix] Annotate ComputePrimValue output as host function (#17032)
by Eric Lunderberg
· 4 days ago
b2c6116
[Relax][Bugfix] Bind symbolic variables in R.match_cast (#17034)
by Eric Lunderberg
· 4 days ago
d4b096f
[Web] Fix string to uint8 array for special characters (#17031)
by Charlie Ruan
· 4 days ago
cab54e0
[SME][TOPI] Add conv2d NHWC SME fp32 schedule (#17003)
by Andrei Hutu
· 4 days ago
430e02f
[SME] Add scalable fp16->fp32 dense schedule (#16981)
by Luke Hutton
· 4 days ago
20d8c53
[SVE] Add support for representing and creating buffer-level predicates (#16966)
by Luke Hutton
· 4 days ago
b598f28
[Contrib] Implement NDArray cache update (#17029)
by Wuwei Lin
· 5 days ago
7359313
[TIR] Fix Shuffle rewrite (#17030)
by Siyuan Feng
· 5 days ago
27a3b90
[Web] Add dtype and offset for CreateView in runtime (#17028)
by Charlie Ruan
· 6 days ago
4f1e2df
[picojson] Let objects be ordered when serializing (#17027)
by Yixin Dong
· 7 days ago
f498cef
[WebGPU] Update error messages to be more user-friendly (#17021)
by Nestor Qin
· 7 days ago
7f7762d
[DLight] Perf improvement for low_batch_gemv on Metal (#17026)
by Siyuan Feng
· 8 days ago
cf2753e
[Relax][UnitTest] Validate IRModule with multiple targets (#16960)
by Eric Lunderberg
· 8 days ago
604fbbd
Support multinomial_from_uniform dispatch (#17010)
by Siyuan Feng
· 8 days ago
7463b37
[Metal] Support metal device profiling (#17025)
by Siyuan Feng
· 8 days ago
b1951a7
[USMP] add missing const specifier for global_const_workspace (#16999)
by Philipp van Kempen
· 9 days ago
e978a44
[COMMUNITY] New committer: Balint Cristian (#17018)
by Leandro Nunes
· 10 days ago
ac9a943
[TOPI][Testing] Enable conv2d NHWC fp16 topi testing for `arm_cpu` (#17007)
by Andrei Hutu
· 10 days ago
a5862a5
[SVE] Use only powers of two as possible vscale values (#17001)
by Luke Hutton
· 11 days ago
2e56421
[DLight] Update Adreno GEMV Rules (#17016)
by Siyuan Feng
· 11 days ago
3b97658
[DLIGHT][GPU] Enhance opencl thread limit for schedules (#16972)
by krishnaraj36
· 11 days ago
209971a
[DLIGHT][GPU] Improved gemv outer fallback schedule (#16973)
by krishnaraj36
· 11 days ago
18a2a25
[KVCache] Support KVCache decode from forked sequence and pop more tokens (#16995)
by Yaxing Cai
· 12 days ago
3cd6673
[Runtime] Allow query of available device memory through DeviceAPI (#16994)
by Eric Lunderberg
· 13 days ago
afb6416
[WebGPU] Handle device OOM in createBuffer (#17005)
by Charlie Ruan
· 2 weeks ago
f044eef
[Runtime][Disco] Restore checks for hangup of disco pipe (#16997)
by Eric Lunderberg
· 2 weeks ago
b49468d
[SME] Introduce scalable fp32 dense schedule (#16921)
by Luke Hutton
· 2 weeks ago
cfe1711
chore: remove repetitive words (#16957)
by tianzedavid
· 3 weeks ago
b7467aa
[Relay] fixed to make TupleGetItem inherits the previous span (#16996)
by mawnja
· 3 weeks ago
d9dbbc9
[Bugfix][Disco] Handle NDArray larger than OS buffer for pipe (#16992)
by Eric Lunderberg
· 3 weeks ago
54c68d6
[Disco] Implement `num_workers` property for `disco.Session` (#16978)
by Eric Lunderberg
· 3 weeks ago
93233a9
[Disco] Treat hangup of disco worker process as kShutdown (#16989)
by Eric Lunderberg
· 3 weeks ago
c6a8a80
[Disco] Allow allocation that only exists on worker0 (#16993)
by Eric Lunderberg
· 3 weeks ago
c2d14ae
[Relax][Transform] Handle identical PrimFunc with distinct VDevice (#16959)
by Eric Lunderberg
· 3 weeks ago
5b5f8d0
[QoL][IR] Provide std::hash and std::equal_to for IR Variable types (#16909)
by Eric Lunderberg
· 3 weeks ago
0dfc5f9
[Unity] Check for transpose and dynamic shape in AdjustMatmulOrder (#16589)
by Eric Lunderberg
· 3 weeks ago
eb242ec
[DLight] Check for target in function attributes (#16958)
by Eric Lunderberg
· 3 weeks ago
2933744
[Cuda] Skip FreeDataSpace when CUDA driver is in inconsistent state (#16980)
by Eric Lunderberg
· 3 weeks ago
fd820ad
[Disco] Expose disco.Session.shutdown through the python API (#16979)
by Eric Lunderberg
· 3 weeks ago
1d4b9ea
[UnitTest] Use pytest's scope='session' for tvm.testing.parameter (#16930)
by Eric Lunderberg
· 3 weeks ago
d1ac1c0
[KVCache] Fix the aux data syncing order of paged KV cache (#16988)
by Rick Zhou
· 3 weeks ago
4403379
[JVM] Automatic Compatibility of JVM AttachCurrentThread (#16987)
by Tianqi Chen
· 3 weeks ago
825dc1f
[TOPI] Remove `blockIdx.z` in topi sort (#16977)
by Siyuan Feng
· 3 weeks ago
2565aa3
[BugFix][Relax] change FuseOpsByPattern strategy to pattern-match maximal subgraph (#16922)
by lazypanda
· 3 weeks ago
fffd168
[Unity][BYOC] Use arith.Analyzer to check batch equality of matmul in cublas (#16982)
by Rick Zhou
· 3 weeks ago
4c1ebcf
[Relax] Implement relax.op.view (#16955)
by Eric Lunderberg
· 3 weeks ago
c0a47ed
[CUBLAS][FP8] Enable R.matmul + R.multiply offloading (#16974)
by Ivan Sidorenko
· 3 weeks ago
02c4c55
[SVE] Add codegen support for `vscale_range()` function attribute (#16962)
by Andrei Hutu
· 3 weeks ago
819b002
[Relax] Support nested ModuleList in nn.Module (#16971)
by Wuwei Lin
· 4 weeks ago
28d32b5
[TIR] Support narrow dtype for let binding (#16947)
by Siyuan Feng
· 4 weeks ago
876f528
[LLVM] Stringref API deprecation fixes (#16968)
by Anirudh Sundar Subramaniam
· 4 weeks ago
9cfebca
[TVMScript] Fix error reporting inside Macro func (#16967)
by Siyuan Feng
· 4 weeks ago
59ef0ee
[Bugfix][ONNX] Improve broadcast and batch_matmul conversion (#16961)
by XinhuaHamiMelon
· 4 weeks ago
944d180
[SVE] Add get_active_lane_mask builtin (#16965)
by Luke Hutton
· 4 weeks ago
effa5d7
[CUBLAS] Enable offloading of R.matmul + R.dequantize (#16896)
by Ivan Sidorenko
· 4 weeks ago
20d7696
[Relax] Express dynamic arguments of strided_slice as arguments (#16826)
by Eric Lunderberg
· 4 weeks ago
a320b63
[Unity][Cutlass] Fix C source generation of dense operation (#16476)
by Jinbae Park
· 5 weeks ago
6252fa5
[TIR] Enhance CLZ intrinsic support (#16952)
by Siyuan Feng
· 5 weeks ago
bc8742b
[Misc] Add script for testing release package (#16956)
by ysh329
· 5 weeks ago
c8deb7f
Overriding the StructuralEqual() for easy usage (#16908)
by sdalvi-quic
· 5 weeks ago
114ad70
[TOPI] Revert unification of conv2d NHWC hybrid scheduling for `arm_cpu` targets (#16951)
by Andrei Hutu
· 5 weeks ago
b4a69de
Enable gemv schedule for adreno (#16932)
by krishnaraj36
· 5 weeks ago
c0385c7
[Runtime] Allow offset to be specified in NDArray::CreateView (#16938)
by Eric Lunderberg
· 5 weeks ago
dd09c85
[CI] Update image tag to 20240428-060115-0b09ed018 (#16948)
by Yong Wu
· 5 weeks ago
2d7663c
[CI] Use LLVM17 for tests on `ci_cpu` (#16931)
by Luke Hutton
· 5 weeks ago
e10cdc5
[tir][Compute-at] Make compute-ated block simple when the predicate could be merged (#16945)
by wrongtest
· 5 weeks ago
b00fc55
[CI] Enable Conda setup v3 (#16942)
by Tianqi Chen
· 5 weeks ago
081c23b
[Relax] Allow PrimValue as index in relax.op.take (#16940)
by Eric Lunderberg
· 5 weeks ago
b54f57a
[TFLite] Add support for GELU conversion (#16936)
by Luke Hutton
· 5 weeks ago
0b09ed0
[3rdparty] Bump FlashInfer for sampling functions (#16935)
by Ruihang Lai
· 5 weeks ago
63e0a0f
[Thrust] Increase static workspace size (#16937)
by Ruihang Lai
· 5 weeks ago
3ff3daa
[CI] Upgrade CUDA to 12.4 (#16939)
by Yong Wu
· 5 weeks ago
1453893
[CLML] Fix in clml pattern check condition (#16933)
by krishnaraj36
· 5 weeks ago
97ff7cc
[VM][OPENCL] Take advantage of OpenCL host ptr for improved copy (#16929)
by Siva
· 5 weeks ago
278a6af
[Relax][TIR] Introduce new `cumsum` op for gpu (#16934)
by Siyuan Feng
· 5 weeks ago
5bd1047
[SCRIPT][ADRENO] Fix in build config for adreno (#16927)
by krishnaraj36
· 5 weeks ago
51cfb70
[Fix][Dlight] Fix GeneralReduction for log-sum-exp (#16923)
by Ruihang Lai
· 5 weeks ago
39f2482
[Fix] Fix SSA conversion for SizeVar retention (#16924)
by Ruihang Lai
· 5 weeks ago
4f8c03f
[TVMScript] Support `T.launch_thread` with i64 dtype (#16916)
by Siyuan Feng
· 5 weeks ago
5cf4ca6
[Marvell BYOC]: Marvell AI Accelerator Integration - Phase 2 (#16915)
by Krishna Bindumadhavan
· 5 weeks ago
2f395f1
[SVE][TOPI] Add conv2d NHWC hybrid SVE schedule for `arm_cpu` (#16899)
by Andrei Hutu
· 5 weeks ago
11f2253
Restore "pytest.mark.gpu" for RELAX tests (#16741)
by apeskov
· 6 weeks ago
342f472
[Disco] Improve error message for CallPacked (#16919)
by Wuwei Lin
· 6 weeks ago
b0143d1
[CMAKE] Make LOG_BEFORE_THROW explicit (#16914)
by Tianqi Chen
· 6 weeks ago
29534b7
[SVE] Check for SVE target in VectorizeLoop (#16893)
by Elen Kalda
· 6 weeks ago
57316da
[Web] Support string[] in setPackedFunc() and exceptionally long arrays (#16910)
by Charlie Ruan
· 6 weeks ago
6b77cba
[Misc] Enhance Release Note Script and Remove Useless File (#16913)
by ysh329
· 6 weeks ago
Next »