branch: master
Commits on master
- 5c6ed5d lower test_conv_3x3_256_32_32_256_256 expectation (#8060) 1 year ago
- c6f5bb0 YoloV8 WebGPU fixes (#8057) 1 year ago
- 78c01a5 amd general _gpu_alloc (#8056) 1 year ago
- 8071600 nv one _gpu_alloc (#8055) 1 year ago
- ff9a89f Proper dtypes for input/output of exported WebGPU model (#8053) 1 year ago
- 435a51e reduce folding simple tests [pr] (#8040) 1 year ago
- 20878be lower test_gemv_4096_16384 expectations 1 year ago
- 83aecbd do gpuocelot copy manually [pr] (#8050) 1 year ago
- 4a208bf bump download cache version 1 year ago
- df18e7c accept filename decorator [pr] (#8049) 1 year ago
- c318708 QwQ-32B-Preview support (#7962) 1 year ago
- b3220ca test cases of always True/False lt (#8048) 1 year ago
- 8bb8068 hook_overflow -> safe_exp2 [pr] (#8047) 1 year ago
- 99abdc6 minor push_swizzle_down_through_elementwise cleanup [pr] (#8046) 1 year ago
- 5933ec8 use argfix in smax/smin and remove if [pr] (#8045) 1 year ago
- 4e51833 minor get_grouped_dims cleanup [pr] (#8044) 1 year ago
- 5ce8090 simple onnx_ops cleanups (#8003) 1 year ago
- 70db1ba Fold nested div with const (#8010) 1 year ago
- 0693158 lower v_theoretical gemv on red (#8042) 1 year ago
- 5c2b108 vectorized input in div_and_mod_folding returns None [pr] (#8041) 1 year ago
- ff6def9 simple contiguous_while_contiguous prereqs [pr] (#8038) 1 year ago
- c9e7701 Fast YoloV8 on WebGPU (#8036) 1 year ago
- b116e15 make device on uop optional [pr] (#8034) 1 year ago
- 13eedd3 Run WebGPU tests on ubuntu (#8033) 1 year ago
- fb89971 use BufferedReader (#8032) 1 year ago
- 08657cb hotfix: bump expectations in speed_v_theoretical 1 year ago
- ea65c79 hotfix: don't spam BEAM debug in speed_v_theoretical 1 year ago
- 09b00b1 hotfix: use kernel timings instead of python timings in speed_v_theoretical 1 year ago
- 8f65c1f simpler block reorder function [pr] (#8031) 1 year ago
- f0401e1 tar_extract with Tensors (#7853) 1 year ago