Default Branch

f96eaddba8 · Revert DFlash SWA optimization (#2039) · Updated 2026-06-26 04:00:09 -05:00

Branches

5425749950 · Fix crash with GLM and MTP · Updated 2026-05-26 09:02:41 -05:00

124
1

e5732606c5 · Fix cache loading/saving for MLA models and split mode graph · Updated 2026-05-26 07:37:12 -05:00

124
1

f3e929c25e · Disable K Hadamard transform if K-head size is not a power of 2 · Updated 2026-05-26 02:19:08 -05:00

124
1

c7211cc500 · Minor logging cleanup · Updated 2026-05-23 11:05:54 -05:00    jdelony

129
1

e5abe3a86a · Per GPU fit margin · Updated 2026-05-23 10:24:19 -05:00    jdelony

129
1

d065b9f742 · It is actually not related to split mode graph · Updated 2026-05-23 05:47:57 -05:00    jdelony

130
2

516dfb39f3 · Fix split mode graph with ngl < n_layer · Updated 2026-05-23 01:43:52 -05:00    jdelony

131
1

8bf4e6ca50 · MTP tweaks 3 · Updated 2026-05-22 04:37:36 -05:00    jdelony

133
1

fa1f302d77 · Fix split mode graph for Qwen35-MoE + MTP · Updated 2026-05-22 01:11:14 -05:00    jdelony

134
1

0bcfde9518 · Disable split mode graph for Qwen35-MoE when MTP is enabled · Updated 2026-05-21 08:24:37 -05:00    jdelony

137
1

dd123f9f4f · Fix crash with split mode graph and partial offload · Updated 2026-05-21 05:31:51 -05:00    jdelony

139
1

f1e146859b · Fix Gemma4-E4B compute graph · Updated 2026-05-21 03:19:30 -05:00    jdelony

139
1

a3d46a963a · Fix MTP when -no-gr is used · Updated 2026-05-20 05:36:48 -05:00    jdelony

145
1

c2c12c987d · Fix mla = 1 / mla = 3 confusion · Updated 2026-05-20 02:02:00 -05:00    jdelony

150
2

8b9db5efcb · Remove Makefile · Updated 2026-05-20 01:12:59 -05:00    jdelony

147
1

301f8d9afd · Fix #1837 · Updated 2026-05-19 09:54:13 -05:00    jdelony

151
1

2575143637 · Enable split mode graph for MLA models and partial offload · Updated 2026-05-19 07:34:03 -05:00    jdelony

151
1

7dd19e197d · Some tweaks · Updated 2026-05-18 10:52:26 -05:00    jdelony

159
2

544fc08db2 · Check for output_extra.weight when loading Gemma4 assistant models · Updated 2026-05-17 09:24:04 -05:00    jdelony

160
1

d237d7b398 · Fix Gemma4 MTP · Updated 2026-05-17 09:16:10 -05:00    jdelony

160
2