Skip to content

qwen4exp : enable sparse flash attention in build_attn_qsa #28349

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
sammcj wants to merge 1 commit into from
+1 −4
Closed

qwen4exp : enable sparse flash attention in build_attn_qsa #28349

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 1 addition & 4 deletions src/models/qwen4exp.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -744,10 +744,7 @@ ggml_tensor * llama_model_qwen4exp::graph::build_attn_qsa(
ggml_tensor * k = mctx_cur->get_k(ctx0, il);
ggml_tensor * v = mctx_cur->get_v(ctx0, il);

// TODO: enable sparse attention when we are ready
// ref: https://github.com/ggml-org/llama.cpp/pull/27970
//ggml_tensor * cur = build_attn_mha(q, k, v, nullptr, kq_mask_top_k, nullptr, nullptr, top_k->ne[0], kq_scale, il);
ggml_tensor * cur = build_attn_mha(q, k, v, nullptr, kq_mask_top_k, nullptr, nullptr, 0, kq_scale, il);
ggml_tensor * cur = build_attn_mha(q, k, v, nullptr, kq_mask_top_k, nullptr, nullptr, top_k->ne[0], kq_scale, il);
cb(cur, "kqv_out", il);

// the rotation is its own inverse, so undo it on the value side of the output
Expand Down