furyhawk/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-07-21 02:05:35 +00:00
Code Issues Packages Projects Releases Wiki Activity
4,735 Commits 952 Branches 1 Tag

Default Branch

main

9d07d8681e · perplexity: fix int overflows in large context x vocab buffer sizing (#2150) · Updated 2026-07-18 06:13:47 +00:00
 ZIP  TAR.GZ

Branches

ik/dsa_loop_hadamard_blend

06e6530d32 · Various · Updated 2026-07-01 13:31:54 +00:00
61
19
 ZIP  TAR.GZ
ik/dsa_loop

d536359431 · Perform indexer score calculation in a loop over attention heads · Updated 2026-06-29 17:18:25 +00:00
74
11
 ZIP  TAR.GZ
ik/max_src_16

6ceac17021 · Bump GGML_MAX_SRC to 16 · Updated 2026-06-29 13:32:41 +00:00
64
1
 ZIP  TAR.GZ
ik/ctx_for_layer_split

13b732bc2c · Adjust split context for layer · Updated 2026-06-29 13:04:52 +00:00
65
1
 ZIP  TAR.GZ
ik/pr_2047_a

fa7fe4eaaf · cuda-graph: GLM4_MOE - Don't load layer.nextn.embed_tokens · Updated 2026-06-27 23:55:55 +00:00
69
3
 ZIP  TAR.GZ
ik/revert_dflash_swa_opt

0440345ba9 · Revert DFlash SWA optimization · Updated 2026-06-26 08:58:50 +00:00
70
1
 ZIP  TAR.GZ
ik/dflash_tweaks

a4e408611d · Minor DFlash tweaks · Updated 2026-06-25 15:10:16 +00:00
74
1
 ZIP  TAR.GZ
ik/qwen35_mtp_smgraph

e1670f6c6c · Merge remote-tracking branch 'origin/main' into ik/qwen35_mtp_smgraph · Updated 2026-06-24 16:32:10 +00:00
78
7
 ZIP  TAR.GZ
ik/g4_assistant_smgraph

1f5828eaa4 · It is better to use llama_context pointers as keys · Updated 2026-06-24 13:53:59 +00:00
80
5
 ZIP  TAR.GZ
ik/tensor_names

9283af5ed8 · Avoid Gemma4 assistant strange tensor name warnings · Updated 2026-06-24 09:20:41 +00:00
80
1
 ZIP  TAR.GZ
ik/purge_blas

3cf0f5468f · Also these · Updated 2026-06-19 15:24:24 +00:00
98
2
 ZIP  TAR.GZ
ik/gemma4_mtp_last_device

e734b76632 · Force Gemma4 assistant to be loaded on last GPU · Updated 2026-06-19 13:51:11 +00:00
98
2
 ZIP  TAR.GZ
ik/gemma4_mtp_graph_reuse

d1692e1951 · Allow graph reuse for Gemma4 MTP · Updated 2026-06-19 09:34:45 +00:00
98
1
 ZIP  TAR.GZ
ik/compat_g4_assistant

25d91dea44 · Add compatibility for llama.cpp Gemma4 assistant GGUFs · Updated 2026-06-19 07:50:26 +00:00
99
1
 ZIP  TAR.GZ
ik/fix_gemma4_mtp

67b0b22760 · Fix Gemma4 MTP compute graph · Updated 2026-06-18 15:51:22 +00:00
103
2
 ZIP  TAR.GZ
ik/glm_mtp_warmup

2c1dc8781b · Fix MTP warmup for GLM models · Updated 2026-06-18 13:15:10 +00:00
103
1
 ZIP  TAR.GZ
ik/fix_qwen_mtp_warmup

dc81d79cb6 · Provide API to gtet the model arch string · Updated 2026-06-17 16:18:32 +00:00
108
4
 ZIP  TAR.GZ
ik/dflash_fix_smgraph

5b9c3bbc3b · Fix DFlash oerformance with split mode graph · Updated 2026-06-17 05:46:05 +00:00
108
1
 ZIP  TAR.GZ
ik/dflash_fix_cpu

6f45163a95 · Fix DFlash on the CPU · Updated 2026-06-16 13:22:36 +00:00
111
0
 Included
 ZIP  TAR.GZ
ik/fattn_mma_gqa_16

6be3a488d3 · CUDA FA: faster TG when GQA is 16 and head size is 128 · Updated 2026-06-15 11:46:02 +00:00
132
0
 Included
 ZIP  TAR.GZ