llama.cpp
Fix GLM4 incoherence with fp16 accumulators
#13639
Merged
Overview
Commits
1
Changes
View On GitHub

Fix GLM4 incoherence with fp16 accumulators #13639

0cc4m merged 1 commit into master from 0cc4m/fix-vulkan-glm4
0cc4m
0cc4m Set GLM4 blk.*.attn_output.weight, kqv_out-* matmul to GGML_PREC_F32 …
adefa985
0cc4m 0cc4m requested a review from ggerganov ggerganov 1 year ago
ggerganov
ggerganov approved these changes on 2025-05-20
0cc4m 0cc4m merged c9c64dee into master 1 year ago
0cc4m 0cc4m deleted the 0cc4m/fix-vulkan-glm4 branch 1 year ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum