Skip to content

[ExecuTorch][WebGPU] Test coverage for the f16 KV cache #20773

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

Open
JCNTH wants to merge 2 commits into
base: gh/JCNTH/11/base
Choose a base branch
from
+10 −1
Open

[ExecuTorch][WebGPU] Test coverage for the f16 KV cache #20773

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
dd4538b
Update
JCNTH Jul 7, 2026
1479030
Update
JCNTH Jul 7, 2026
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
11 changes: 10 additions & 1 deletion backends/webgpu/test/test_webgpu_native.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -229,14 +229,23 @@ bool sdpa_within_tol(
int n,
float* ma,
float* mr) {
float atol = 1e-4f, rtol = 1e-3f;
#ifdef WGPU_BACKEND_KV_F16
// f16 KV opt-in: looser tol for f16 K/V read precision (shader-f16 device).
const WebGPUContext* kv_ctx = get_default_webgpu_context();
if (kv_ctx != nullptr && kv_ctx->shader_f16_supported) {
atol = 2e-3f;
rtol = 1e-2f;
}
#endif
float max_abs = 0.0f, max_rel = 0.0f;
bool ok = true;
for (int i = 0; i < n; i++) {
const float ae = std::abs(out[i] - golden[i]);
const float re = ae / std::max(std::abs(golden[i]), 1e-6f);
max_abs = std::max(max_abs, ae);
max_rel = std::max(max_rel, re);
if (ae > 1e-4f && re > 1e-3f) {
if (ae > atol && re > rtol) {
ok = false;
}
}
Expand Down
Loading