git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit

]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit

overview / pkg / ggml / sources / llama.cpp / commit

author	Georgi Gerganov <redacted>
	Fri, 5 Jun 2026 08:09:36 +0000 (11:09 +0300)
committer	GitHub <redacted>
	Fri, 5 Jun 2026 08:09:36 +0000 (11:09 +0300)
commit	7acb4e8cd2ce21f457d1298e75fad729520d263c
tree	5a77024ea13397365f3be1610bee5573b0c42f1f	tree
parent	3ecfb150a4bd2d92b2a7974bb1af954c8a5e2985	commit \| diff

hparams : refactor `hparams.n_layer` (#24060)

* hparams : refactor hparams.n_layer

* cont : remove `n_layer_kv()`, use n_layer_all instead

* cont : type consistency

* pi : update SYSTEM.md

* models : fix Step3.5 MTP

* cont : remove duplicate switch cases

* cont : explicitly set `false` to extra layers for `is_swa` and `is_recr`

* cont : fix nextn layer count handling

Co-authored-by: Sigbjørn Skjæret <redacted>
---------

Co-authored-by: Sigbjørn Skjæret <redacted>

129 files changed:

.pi/gg/SYSTEM.md		diff \| blob \| history
src/llama-adapter.cpp		diff \| blob \| history
src/llama-context.cpp		diff \| blob \| history
src/llama-graph.cpp		diff \| blob \| history
src/llama-hparams.cpp		diff \| blob \| history
src/llama-hparams.h		diff \| blob \| history
src/llama-kv-cache.cpp		diff \| blob \| history
src/llama-memory-recurrent.cpp		diff \| blob \| history
src/llama-model-loader.cpp		diff \| blob \| history
src/llama-model-saver.cpp		diff \| blob \| history
src/llama-model.cpp		diff \| blob \| history
src/llama-model.h		diff \| blob \| history
src/llama-quant.cpp		diff \| blob \| history
src/models/afmoe.cpp		diff \| blob \| history
src/models/apertus.cpp		diff \| blob \| history
src/models/arcee.cpp		diff \| blob \| history
src/models/arctic.cpp		diff \| blob \| history
src/models/arwkv7.cpp		diff \| blob \| history
src/models/baichuan.cpp		diff \| blob \| history
src/models/bailingmoe.cpp		diff \| blob \| history
src/models/bailingmoe2.cpp		diff \| blob \| history
src/models/bert.cpp		diff \| blob \| history
src/models/bitnet.cpp		diff \| blob \| history
src/models/bloom.cpp		diff \| blob \| history
src/models/chameleon.cpp		diff \| blob \| history
src/models/chatglm.cpp		diff \| blob \| history
src/models/codeshell.cpp		diff \| blob \| history
src/models/cogvlm.cpp		diff \| blob \| history
src/models/cohere2.cpp		diff \| blob \| history
src/models/command-r.cpp		diff \| blob \| history
src/models/dbrx.cpp		diff \| blob \| history
src/models/deci.cpp		diff \| blob \| history
src/models/deepseek2.cpp		diff \| blob \| history
src/models/deepseek2ocr.cpp		diff \| blob \| history
src/models/deepseek32.cpp		diff \| blob \| history
src/models/dots1.cpp		diff \| blob \| history
src/models/dream.cpp		diff \| blob \| history
src/models/ernie4-5.cpp		diff \| blob \| history
src/models/eurobert.cpp		diff \| blob \| history
src/models/exaone-moe.cpp		diff \| blob \| history
src/models/exaone.cpp		diff \| blob \| history
src/models/exaone4.cpp		diff \| blob \| history
src/models/falcon-h1.cpp		diff \| blob \| history
src/models/falcon.cpp		diff \| blob \| history
src/models/gemma-embedding.cpp		diff \| blob \| history
src/models/gemma.cpp		diff \| blob \| history
src/models/gemma2.cpp		diff \| blob \| history
src/models/gemma3.cpp		diff \| blob \| history
src/models/gemma3n.cpp		diff \| blob \| history
src/models/gemma4.cpp		diff \| blob \| history
src/models/glm-dsa.cpp		diff \| blob \| history
src/models/glm4-moe.cpp		diff \| blob \| history
src/models/glm4.cpp		diff \| blob \| history
src/models/gpt2.cpp		diff \| blob \| history
src/models/gptneox.cpp		diff \| blob \| history
src/models/granite-hybrid.cpp		diff \| blob \| history
src/models/granite-moe.cpp		diff \| blob \| history
src/models/granite.cpp		diff \| blob \| history
src/models/grok.cpp		diff \| blob \| history
src/models/grovemoe.cpp		diff \| blob \| history
src/models/hunyuan-moe.cpp		diff \| blob \| history
src/models/internlm2.cpp		diff \| blob \| history
src/models/jais.cpp		diff \| blob \| history
src/models/jais2.cpp		diff \| blob \| history
src/models/jamba.cpp		diff \| blob \| history
src/models/jina-bert-v2.cpp		diff \| blob \| history
src/models/jina-bert-v3.cpp		diff \| blob \| history
src/models/kimi-linear.cpp		diff \| blob \| history
src/models/lfm2.cpp		diff \| blob \| history
src/models/lfm2moe.cpp		diff \| blob \| history
src/models/llada-moe.cpp		diff \| blob \| history
src/models/llada.cpp		diff \| blob \| history
src/models/llama.cpp		diff \| blob \| history
src/models/llama4.cpp		diff \| blob \| history
src/models/maincoder.cpp		diff \| blob \| history
src/models/mamba.cpp		diff \| blob \| history
src/models/mamba2.cpp		diff \| blob \| history
src/models/mellum.cpp		diff \| blob \| history
src/models/mimo2.cpp		diff \| blob \| history
src/models/minicpm.cpp		diff \| blob \| history
src/models/minicpm3.cpp		diff \| blob \| history
src/models/minimax-m2.cpp		diff \| blob \| history
src/models/mistral3.cpp		diff \| blob \| history
src/models/modern-bert.cpp		diff \| blob \| history
src/models/mpt.cpp		diff \| blob \| history
src/models/nemotron-h.cpp		diff \| blob \| history
src/models/nemotron.cpp		diff \| blob \| history
src/models/neo-bert.cpp		diff \| blob \| history
src/models/nomic-bert-moe.cpp		diff \| blob \| history
src/models/nomic-bert.cpp		diff \| blob \| history
src/models/olmo.cpp		diff \| blob \| history
src/models/olmo2.cpp		diff \| blob \| history
src/models/olmoe.cpp		diff \| blob \| history
src/models/openai-moe.cpp		diff \| blob \| history
src/models/openelm.cpp		diff \| blob \| history
src/models/orion.cpp		diff \| blob \| history
src/models/pangu-embed.cpp		diff \| blob \| history
src/models/phi2.cpp		diff \| blob \| history
src/models/phi3.cpp		diff \| blob \| history
src/models/phimoe.cpp		diff \| blob \| history
src/models/plamo.cpp		diff \| blob \| history
src/models/plamo2.cpp		diff \| blob \| history
src/models/plamo3.cpp		diff \| blob \| history
src/models/plm.cpp		diff \| blob \| history
src/models/qwen.cpp		diff \| blob \| history
src/models/qwen2.cpp		diff \| blob \| history
src/models/qwen2moe.cpp		diff \| blob \| history
src/models/qwen3.cpp		diff \| blob \| history
src/models/qwen35.cpp		diff \| blob \| history
src/models/qwen35moe.cpp		diff \| blob \| history
src/models/qwen3moe.cpp		diff \| blob \| history
src/models/qwen3next.cpp		diff \| blob \| history
src/models/qwen3vl.cpp		diff \| blob \| history
src/models/qwen3vlmoe.cpp		diff \| blob \| history
src/models/refact.cpp		diff \| blob \| history
src/models/rnd1.cpp		diff \| blob \| history
src/models/rwkv6.cpp		diff \| blob \| history
src/models/rwkv6qwen2.cpp		diff \| blob \| history
src/models/rwkv7.cpp		diff \| blob \| history
src/models/seed-oss.cpp		diff \| blob \| history
src/models/smallthinker.cpp		diff \| blob \| history
src/models/smollm3.cpp		diff \| blob \| history
src/models/stablelm.cpp		diff \| blob \| history
src/models/starcoder.cpp		diff \| blob \| history
src/models/starcoder2.cpp		diff \| blob \| history
src/models/step35.cpp		diff \| blob \| history
src/models/t5.cpp		diff \| blob \| history
src/models/talkie.cpp		diff \| blob \| history
src/models/xverse.cpp		diff \| blob \| history

Packaging of ggml-org/llama.cpp

RSS Atom