branch: master
Commits on master
- ddeec24 Cleanup & fix llama.py (#2524) 2 years ago
- 7d26452 call ruff with --preview (#2522) 2 years ago
- 5db0cdf support list of ints (or other Tensorable) in tensor indices (#2520) 2 years ago
- bd941a0 first version of test_indexing (#2515) 2 years ago
- d210f6a minor device.py cleanups (#2510) 2 years ago
- 370cfbb Cleanup vectorized hip renders (#2497) 2 years ago
- abfc991 cleanup realize (#2505) 2 years ago
- 3dedeaa rebalance tests (#2504) 2 years ago
- 065aff7 make webgpu test reliable (#2502) 2 years ago
- 6707f25 use copyin (#2500) 2 years ago
- 947711a split metal and webgpu tests (#2501) 2 years ago
- 3eb3c74 metal ci tests everything (#2499) 2 years ago
- 889acef Support weird loads in Image (#2498) 2 years ago
- e333672 realize cleanup (#2496) 2 years ago
- 5629fc3 Use Buffer.STORE at the end of ASTs (#2494) 2 years ago
- cf0c909 Removing METAL Skips as CI works (#2488) 2 years ago
- 5588922 Update cuda_matmul.py (#2495) 2 years ago
- cdc3b95 if you don't appreciate a 15 second timeout, you get a 10 second timeout 2 years ago
- d87a246 move to new cached fetch (#2493) 2 years ago
- ab5d14d MEM -> LOAD (#2492) 2 years ago
- a739c66 fp16 in gpt2 attention (#2491) 2 years ago
- 847f0a0 non-simplifiable mod should result in ModNode (#2490) 2 years ago
- 3f137b1 jax parallel matmul example 2 years ago
- ddb6a33 improve test assertions for jit cache len with graph executor (#2476) 2 years ago
- 28a6710 enable symbolic ops tests for hip (#2485) 2 years ago
- 7f01dd0 Apply ruff linting rules to tests (#2473) 2 years ago
- 136dbd8 HIP CI that compiles (to RDNA3) but doesn't have to run (#2482) 2 years ago
- 756b01f why were these ever called buffer (#2483) 2 years ago
- acbe6d1 Revert "HIP compilation on CI targeting RDNA3 (#2459)" (#2481) 2 years ago
- cb507a9 Remove the toCPU copy (#2445) 2 years ago