| [AMD] Add fast_tanhf to libdevice (#7780) | 9 个月前 |
| Merge Triton-Ascend 425236de into release/3.5.x | 2 个月前 |
| fix(op): support fast_math semantics in dot_scaled lowering | 1 个月前 |
| Merge Triton-Ascend 62eb951f into release/3.5.x | 2 个月前 |
| Use 32-bit Philox consistently in randint4x even if offset is 64-bit (#6832) | 1 年前 |
| feature: Implement for dot scale fp4 | 1 个月前 |
| fix(intrusive): fix ravel to stay consistent with upstream | 1 个月前 |
| [LANG] Expand tl.target_info and fail gracefully on CPU (#7869) | 9 个月前 |