Files
nano-vllm/nanovllm
Anai-Guo bf99453d90 fix RowParallelLinear weight_loader crash when bias is enabled
When RowParallelLinear has bias=True, the weight_loader crashes with an
IndexError because it calls param_data.size(tp_dim) where tp_dim=1, but
the bias tensor is 1D and only has dimension 0.

The bias in RowParallelLinear is not sharded (all ranks hold the full
bias, only rank 0 applies it), so skip the sharding logic for 1D params.

Fixes GeeeekExplorer/nano-vllm#125
2026-04-12 23:10:52 -07:00
..
2025-11-04 01:44:42 +08:00
2025-11-04 01:44:42 +08:00
2025-06-21 17:19:15 +08:00
2025-06-15 10:36:45 +08:00
2025-06-27 01:51:57 +08:00
2025-06-15 01:31:24 +08:00
2025-08-31 22:55:34 +08:00