[Dependency] Upgrade FlashInfer version to 0.7.0 (#58069)

Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Kevin H. Luu <khluu000@gmail.com>
This commit is contained in:
Wei Zhao
2026-09-23 22:45:23 -07:00
committed by GitHub
co-authored by Codex Yongye Zhu Kevin H. Luu
parent a1b6763df6
commit 44bf011f44
15 changed files with 197 additions and 31 deletions
+5 -2
View File
@@ -25,7 +25,7 @@ def _flashinfer_topk_topp_supported() -> bool:
Mirrors the gate in `TopKTopPSampler.__init__`: CUDA + flashinfer
importable + GPU compute capability supported by the FlashInfer
backend.
backend + more than 16 SMs.
"""
if not current_platform.is_cuda():
return False
@@ -38,7 +38,10 @@ def _flashinfer_topk_topp_supported() -> bool:
capability = current_platform.get_device_capability()
if capability is None:
return False
return FlashInferBackend.supports_compute_capability(capability)
return FlashInferBackend.supports_compute_capability(capability) and (
current_platform.num_compute_units(torch.accelerator.current_device_index())
> 16
)
FLASHINFER_TOPK_TOPP_SUPPORTED = _flashinfer_topk_topp_supported()