hugo/DS4Server
1
0 Fork 0
Code Issues 1 Pull Requests Actions Packages Projects Releases Wiki Activity

Match Qwen head-chunked attention execution

Browse Source
This commit is contained in:
Georg Bauer committed 2026-10-04 16:45:19 +02:00
1 parent 7b894fbfee
commit d9026aa7bf
70 files changed
+4979 -10
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

docs/issue113-head-chunk/mtplx_qsa_rows_gather_graph_matches_reference.log
+11
View File
@@ -0,0 +1,11 @@
MTPLX canonical qsa_rows_attention_h24_s64_t113_k17_offset: exact
MTPLX canonical qsa_rows_attention_h24_s64_t113_k17_strided: exact
MTPLX canonical qsa_rows_attention_h2_s3_t33_k8_plain: exact
MTPLX canonical qsa_rows_attention_h2_s3_t33_k8_offset: exact
MTPLX canonical qsa_rows_attention_h2_s3_t33_k8_strided: exact
MTPLX canonical qsa_rows_attention_h2_s16_t1057_k1025_plain: exact
MTPLX canonical qsa_rows_attention_h2_s16_t1057_k1025_offset: exact
MTPLX canonical qsa_rows_attention_h2_s16_t1057_k1025_strided: exact
ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 548 filtered out; finished in 0.25s