branch: master
Commits on master
- b2ea610 fix tqdm unit_scale and support hours in time (#5227) 1 year ago
- f374fb7 assert bool dtype for valid [run_process_replay] (#5214) 1 year ago
- 3f4eeb8 late UOps.IF generation [run_process_replay] [no_assert] (#5027) 1 year ago
- 42d1f92 simpler tqdm (#5221) 1 year ago
- dd7eef7 libc defs to autogen (#5217) 1 year ago
- 6b08cb5 ptx runs on nv in benchmarks (#5224) 1 year ago
- b4c49ae remove cudacpu in favour of mockgpu (#5225) 1 year ago
- ee02dcb nv supports PTX=1 (#5222) 1 year ago
- 7bcb74a feat: tag 0.9.1 (#5220) 1 year ago
- 7f46bfa hotfix: docs touchup 1 year ago
- c941a58 amd refactor queue creation (#5216) 1 year ago
- 7ba4938 simplify View.permute arg check [run_process_replay] (#5218) 1 year ago
- 80ac212 hotfix: linearizer test fixup 1 year ago
- c9714df rename graph to children [run_process_replay] (#5215) 1 year ago
- 6c456b6 remove uopgraph dedup + slight speedup (#5199) 1 year ago
- 9b08a93 amd inline bf16 funcs (#5212) 1 year ago
- 7090eac validate sdxl output and put it in benchmark (#5211) 1 year ago
- 63fa4e2 fix seed = 0 in sdxl (#5209) 1 year ago
- 4688f97 Add SDXL Inference to Examples (#5206) 1 year ago
- 3e56c84 remu err handling (#5208) 1 year ago
- 7f7fa26 allow hugepage failure in memadvise (#5207) 1 year ago
- 73395b9 better error msg for TinyJit inside TinyJit (#5202) 1 year ago
- ac748cc nv apply relocs (#5165) 1 year ago
- 540ebdf missing init files (#5196) 1 year ago
- d8dc43a remove JIT_BATCH_SIZE=4 from gpt2 NV benchmark (#5198) 1 year ago
- 345bcc2 move graph_dedup out of class [run_process_replay] (#5197) 1 year ago
- d094a68 single pass rewrite (#5159) 1 year ago
- 1ff9bba ruff: close file handle (#5180) 1 year ago
- 83da8b3 use NV instead of CUDA in benchmark (#5192) 1 year ago
- 0c6c7c5 CACHELEVEL=0 -> IGNORE_BEAM_CACHE=1 in benchmark (#5191) 1 year ago