HappyZ happyz
0 Followers · 0 Following
Repositories
18
 Projects Packages Public Activity Starred Repositories
happyz synced commits to refs/pull/19743/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:26 -07:00
82b49a5dd3 Merge ef80e915e5 into b7ad48ebda
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
Compare 21 commits »
happyz synced commits to refs/pull/19441/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:25 -07:00
d66447f506 Merge eefcfeed7a into 650bf14eb9
650bf14eb9 llama-model: read final_logit_softcapping for Gemma 4 (#21390)
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
Compare 3 commits »
happyz synced commits to refs/pull/19691/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:25 -07:00
98d5b8fa51 Merge c82e486fdb into d006858316
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
f1f793ad06 common/parser: fix call ID detection (Mistral parser mostly) + atomicity for tag-json parsers (#21230)
Compare 26 commits »
happyz synced commits to refs/pull/19493/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:25 -07:00
56919a38e8 Merge 00c0c4c10e into b7ad48ebda
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
Compare 7 commits »
happyz synced commits to refs/pull/19434/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:24 -07:00
a705051cc7 Merge 05dfc18d55 into d006858316
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
f1f793ad06 common/parser: fix call ID detection (Mistral parser mostly) + atomicity for tag-json parsers (#21230)
Compare 22 commits »
happyz synced commits to refs/pull/19341/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:24 -07:00
b1115ec82d Merge 1ea10cd774 into d006858316
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
f1f793ad06 common/parser: fix call ID detection (Mistral parser mostly) + atomicity for tag-json parsers (#21230)
Compare 47 commits »
happyz synced commits to refs/pull/19294/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:24 -07:00
0fd431372c Merge 820ad0d191 into d006858316
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
f1f793ad06 common/parser: fix call ID detection (Mistral parser mostly) + atomicity for tag-json parsers (#21230)
Compare 22 commits »
happyz synced commits to refs/pull/19254/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:24 -07:00
bcec16b50f Merge 40f23b341c into b7ad48ebda
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
Compare 12 commits »
happyz synced commits to refs/pull/19196/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:23 -07:00
664eac81e7 Merge 2675beead6 into d006858316
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
f1f793ad06 common/parser: fix call ID detection (Mistral parser mostly) + atomicity for tag-json parsers (#21230)
Compare 72 commits »
happyz synced commits to refs/pull/19171/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:23 -07:00
f1d33b3ef0 Merge 18ad28ca0a into b7ad48ebda
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
Compare 33 commits »
happyz synced commits to refs/pull/18923/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:23 -07:00
c7637dbffb Merge 7f14ce6344 into b7ad48ebda
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
Compare 82 commits »
happyz synced commits to refs/pull/18432/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:22 -07:00
a4216575c6 Merge 7e9bea7f1c into b7ad48ebda
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
Compare 3 commits »
happyz synced commits to refs/pull/18742/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:22 -07:00
0adde3b5f9 Merge 9882f21ab9 into b7ad48ebda
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
Compare 125 commits »
happyz synced commits to refs/pull/18588/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:22 -07:00
baa1c6e5c8 Merge f32778568a into b7ad48ebda
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
Compare 15 commits »
happyz synced commits to refs/pull/18465/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:22 -07:00
97faa7f8fd Merge 4d56e8abe2 into d006858316
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
f1f793ad06 common/parser: fix call ID detection (Mistral parser mostly) + atomicity for tag-json parsers (#21230)
Compare 58 commits »
happyz synced commits to refs/pull/18150/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:21 -07:00
876844b91e Merge be23f5f74a into 650bf14eb9
650bf14eb9 llama-model: read final_logit_softcapping for Gemma 4 (#21390)
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
Compare 13 commits »
happyz synced commits to refs/pull/17791/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:21 -07:00
1274ed6e21 Merge fd94e4cdca into b7ad48ebda
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
Compare 2 commits »
happyz synced commits to refs/pull/18373/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:21 -07:00
2cd28a3ad1 Merge 95d2017d28 into b7ad48ebda
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
Compare 23 commits »
happyz synced commits to refs/pull/17330/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:20 -07:00
a3d04feea6 Merge 69447f879f into d01f6274c0
d01f6274c0 common : respect specified tag, only fallback when tag is empty (#21413)
650bf14eb9 llama-model: read final_logit_softcapping for Gemma 4 (#21390)
b7ad48ebda llama: add custom newline split for Gemma 4 (#21406)
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
Compare 5 commits »
happyz synced commits to refs/pull/16948/merge at happyz/llama.cpp from mirror 2026-04-04 07:02:20 -07:00
bd7e7af7de Merge d6d24487c2 into d006858316
d006858316 ggml-webgpu: move from parameter buffer pool to single buffer with offsets (#21278)
e439700992 ci: Add Windows Vulkan backend testing on Intel (#21292)
50e0ad08fb server: save and clear idle slots on new task (`--clear-idle`) (#20993)
f1f793ad06 common/parser: fix call ID detection (Mistral parser mostly) + atomicity for tag-json parsers (#21230)
Compare 6 commits »