matt/ollama37
1
0
Fork 0
mirror of https://github.com/dogkeeper886/ollama37.git synced 2025-12-13 17:27:05 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama: update to commit 71e90e88 (#10192)

Browse Source
This commit is contained in:
Jeffrey Morgan
2025-04-16 18:14:01 -04:00
committed by GitHub
parent 369de832cd
commit 943464ccb8
160 changed files with 42219 additions and 33080 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
llama/llama.cpp/src/llama-arch.h vendored
View File

@@ -10,6 +10,7 @@
enum llm_arch {
LLM_ARCH_LLAMA,
LLM_ARCH_LLAMA4,
LLM_ARCH_MLLAMA,
LLM_ARCH_DECI,
LLM_ARCH_FALCON,
@@ -30,6 +31,8 @@ enum llm_arch {
LLM_ARCH_QWEN2,
LLM_ARCH_QWEN2MOE,
LLM_ARCH_QWEN2VL,
LLM_ARCH_QWEN3,
LLM_ARCH_QWEN3MOE,
LLM_ARCH_PHI2,
LLM_ARCH_PHI3,
LLM_ARCH_PHIMOE,
@@ -56,6 +59,7 @@ enum llm_arch {
LLM_ARCH_DEEPSEEK,
LLM_ARCH_DEEPSEEK2,
LLM_ARCH_CHATGLM,
LLM_ARCH_GLM4,
LLM_ARCH_BITNET,
LLM_ARCH_T5,
LLM_ARCH_T5ENCODER,
@@ -64,12 +68,16 @@ enum llm_arch {
LLM_ARCH_EXAONE,
LLM_ARCH_RWKV6,
LLM_ARCH_RWKV6QWEN2,
LLM_ARCH_RWKV7,
LLM_ARCH_ARWKV7,
LLM_ARCH_GRANITE,
LLM_ARCH_GRANITE_MOE,
LLM_ARCH_CHAMELEON,
LLM_ARCH_SOLAR,
LLM_ARCH_WAVTOKENIZER_DEC,
LLM_ARCH_MISTRAL3,
LLM_ARCH_PLM,
LLM_ARCH_BAILINGMOE,
LLM_ARCH_UNKNOWN,
};
@@ -78,6 +86,7 @@ enum llm_kv {
LLM_KV_GENERAL_ARCHITECTURE,
LLM_KV_GENERAL_QUANTIZATION_VERSION,
LLM_KV_GENERAL_ALIGNMENT,
LLM_KV_GENERAL_FILE_TYPE,
LLM_KV_GENERAL_NAME,
LLM_KV_GENERAL_AUTHOR,
LLM_KV_GENERAL_VERSION,
@@ -116,6 +125,7 @@ enum llm_kv {
LLM_KV_RESIDUAL_SCALE,
LLM_KV_EMBEDDING_SCALE,
LLM_KV_TOKEN_SHIFT_COUNT,
LLM_KV_INTERLEAVE_MOE_LAYER_STEP,
LLM_KV_ATTENTION_HEAD_COUNT,
LLM_KV_ATTENTION_HEAD_COUNT_KV,
@@ -130,6 +140,10 @@ enum llm_kv {
LLM_KV_ATTENTION_CAUSAL,
LLM_KV_ATTENTION_Q_LORA_RANK,
LLM_KV_ATTENTION_KV_LORA_RANK,
LLM_KV_ATTENTION_DECAY_LORA_RANK,
LLM_KV_ATTENTION_ICLR_LORA_RANK,
LLM_KV_ATTENTION_VALUE_RESIDUAL_MIX_LORA_RANK,
LLM_KV_ATTENTION_GATE_LORA_RANK,
LLM_KV_ATTENTION_RELATIVE_BUCKETS_COUNT,
LLM_KV_ATTENTION_SLIDING_WINDOW,
LLM_KV_ATTENTION_SCALE,
@@ -248,6 +262,8 @@ enum llm_tensor {
LLM_TENSOR_ATTN_Q_NORM,
LLM_TENSOR_ATTN_K_NORM,
LLM_TENSOR_LAYER_OUT_NORM,
LLM_TENSOR_POST_ATTN_NORM,
LLM_TENSOR_POST_MLP_NORM,
LLM_TENSOR_SSM_IN,
LLM_TENSOR_SSM_CONV1D,
LLM_TENSOR_SSM_X,
@@ -255,8 +271,20 @@ enum llm_tensor {
LLM_TENSOR_SSM_A,
LLM_TENSOR_SSM_D,
LLM_TENSOR_SSM_OUT,
LLM_TENSOR_TIME_MIX_W0,
LLM_TENSOR_TIME_MIX_W1,
LLM_TENSOR_TIME_MIX_W2,
LLM_TENSOR_TIME_MIX_A0,
LLM_TENSOR_TIME_MIX_A1,
LLM_TENSOR_TIME_MIX_A2,
LLM_TENSOR_TIME_MIX_V0,
LLM_TENSOR_TIME_MIX_V1,
LLM_TENSOR_TIME_MIX_V2,
LLM_TENSOR_TIME_MIX_G1,
LLM_TENSOR_TIME_MIX_G2,
LLM_TENSOR_TIME_MIX_K_K,
LLM_TENSOR_TIME_MIX_K_A,
LLM_TENSOR_TIME_MIX_R_K,
LLM_TENSOR_TIME_MIX_LERP_X,
LLM_TENSOR_TIME_MIX_LERP_W,
LLM_TENSOR_TIME_MIX_LERP_K,