Skip to main content

Crate llama_cpp_sys_4

Crate llama_cpp_sys_4

Expand description

See llama-cpp-4 for a documented and safe API.

Modules§

common: Manual wrapper for values in llama.cpp/common/common.h

Structs§

Constants§

CHAT_SHIM_REASONING_AUTO
CHAT_SHIM_REASONING_DEEPSEEK
CHAT_SHIM_REASONING_DEEPSEEK_LEGACY
CHAT_SHIM_REASONING_NONE
CHAT_SHIM_TOOL_CHOICE_AUTO
CHAT_SHIM_TOOL_CHOICE_NONE
CHAT_SHIM_TOOL_CHOICE_REQUIRED
COMMON_PARAMS_FIT_STATUS_ERROR
COMMON_PARAMS_FIT_STATUS_FAILURE
COMMON_PARAMS_FIT_STATUS_SUCCESS
COMMON_SHIM_GRAMMAR_NONE
COMMON_SHIM_GRAMMAR_OUTPUT_FORMAT
COMMON_SHIM_GRAMMAR_TOOL_CALLS
COMMON_SHIM_GRAMMAR_USER
COMMON_SHIM_RBUDGET_COUNTING
COMMON_SHIM_RBUDGET_DONE
COMMON_SHIM_RBUDGET_FORCING
COMMON_SHIM_RBUDGET_IDLE
COMMON_SHIM_RBUDGET_WAITING_UTF8
GGML_BACKEND_BUFFER_USAGE_ANY
GGML_BACKEND_BUFFER_USAGE_COMPUTE
GGML_BACKEND_BUFFER_USAGE_WEIGHTS
GGML_BACKEND_DEVICE_TYPE_ACCEL
GGML_BACKEND_DEVICE_TYPE_CPU
GGML_BACKEND_DEVICE_TYPE_GPU
GGML_BACKEND_DEVICE_TYPE_IGPU
GGML_BACKEND_DEVICE_TYPE_META
GGML_BACKEND_SPLIT_AXIS_0
GGML_BACKEND_SPLIT_AXIS_1
GGML_BACKEND_SPLIT_AXIS_2
GGML_BACKEND_SPLIT_AXIS_3
GGML_BACKEND_SPLIT_AXIS_MIRRORED
GGML_BACKEND_SPLIT_AXIS_NONE
GGML_BACKEND_SPLIT_AXIS_PARTIAL
GGML_BACKEND_SPLIT_AXIS_UNKNOWN
GGML_FTYPE_ALL_F32
GGML_FTYPE_MOSTLY_BF16
GGML_FTYPE_MOSTLY_F8_E4M3
GGML_FTYPE_MOSTLY_F16
GGML_FTYPE_MOSTLY_IQ1_M
GGML_FTYPE_MOSTLY_IQ1_S
GGML_FTYPE_MOSTLY_IQ2_S
GGML_FTYPE_MOSTLY_IQ2_XS
GGML_FTYPE_MOSTLY_IQ2_XXS
GGML_FTYPE_MOSTLY_IQ3_S
GGML_FTYPE_MOSTLY_IQ3_XXS
GGML_FTYPE_MOSTLY_IQ4_NL
GGML_FTYPE_MOSTLY_IQ4_XS
GGML_FTYPE_MOSTLY_MXFP4
GGML_FTYPE_MOSTLY_NVFP4
GGML_FTYPE_MOSTLY_Q1_0
GGML_FTYPE_MOSTLY_Q2_0
GGML_FTYPE_MOSTLY_Q2_K
GGML_FTYPE_MOSTLY_Q3_K
GGML_FTYPE_MOSTLY_Q4_0
GGML_FTYPE_MOSTLY_Q4_1
GGML_FTYPE_MOSTLY_Q4_1_SOME_F16
GGML_FTYPE_MOSTLY_Q4_K
GGML_FTYPE_MOSTLY_Q5_0
GGML_FTYPE_MOSTLY_Q5_1
GGML_FTYPE_MOSTLY_Q5_K
GGML_FTYPE_MOSTLY_Q6_K
GGML_FTYPE_MOSTLY_Q8_0
GGML_FTYPE_UNKNOWN
GGML_GLU_OP_COUNT
GGML_GLU_OP_GEGLU
GGML_GLU_OP_GEGLU_ERF
GGML_GLU_OP_GEGLU_QUICK
GGML_GLU_OP_REGLU
GGML_GLU_OP_SWIGLU
GGML_GLU_OP_SWIGLU_CLAMP
GGML_GLU_OP_SWIGLU_OAI
GGML_HINT_NONE
GGML_HINT_SRC0_IS_HADAMARD
GGML_LOG_LEVEL_CONT
GGML_LOG_LEVEL_DEBUG
GGML_LOG_LEVEL_ERROR
GGML_LOG_LEVEL_INFO
GGML_LOG_LEVEL_NONE
GGML_LOG_LEVEL_WARN
GGML_MROPE_SECTIONS
GGML_NUMA_STRATEGY_COUNT
GGML_NUMA_STRATEGY_DISABLED
GGML_NUMA_STRATEGY_DISTRIBUTE
GGML_NUMA_STRATEGY_ISOLATE
GGML_NUMA_STRATEGY_MIRROR
GGML_NUMA_STRATEGY_NUMACTL
GGML_OBJECT_TYPE_GRAPH
GGML_OBJECT_TYPE_TENSOR
GGML_OBJECT_TYPE_WORK_BUFFER
GGML_OPT_BUILD_TYPE_FORWARD
GGML_OPT_BUILD_TYPE_GRAD
GGML_OPT_BUILD_TYPE_OPT
GGML_OPT_LOSS_TYPE_CROSS_ENTROPY
GGML_OPT_LOSS_TYPE_MEAN
GGML_OPT_LOSS_TYPE_MEAN_SQUARED_ERROR
GGML_OPT_LOSS_TYPE_SUM
GGML_OPT_OPTIMIZER_TYPE_ADAMW
GGML_OPT_OPTIMIZER_TYPE_COUNT
GGML_OPT_OPTIMIZER_TYPE_SGD
GGML_OP_ACC
GGML_OP_ADD
GGML_OP_ADD1
GGML_OP_ADD_ID
GGML_OP_ADD_REL_POS
GGML_OP_ARANGE
GGML_OP_ARGMAX
GGML_OP_ARGSORT
GGML_OP_CLAMP
GGML_OP_COL2IM_1D
GGML_OP_CONCAT
GGML_OP_CONT
GGML_OP_CONV_2D
GGML_OP_CONV_2D_DW
GGML_OP_CONV_3D
GGML_OP_CONV_TRANSPOSE_1D
GGML_OP_CONV_TRANSPOSE_2D
GGML_OP_COS
GGML_OP_COUNT
GGML_OP_COUNT_EQUAL
GGML_OP_CPY
GGML_OP_CROSS_ENTROPY_LOSS
GGML_OP_CROSS_ENTROPY_LOSS_BACK
GGML_OP_CUMSUM
GGML_OP_CUSTOM
GGML_OP_DIAG
GGML_OP_DIAG_MASK_INF
GGML_OP_DIAG_MASK_ZERO
GGML_OP_DIV
GGML_OP_DSV4_HC_COMB
GGML_OP_DSV4_HC_POST
GGML_OP_DSV4_HC_PRE
GGML_OP_DUP
GGML_OP_FILL
GGML_OP_FLASH_ATTN_BACK
GGML_OP_FLASH_ATTN_EXT
GGML_OP_GATED_DELTA_NET
GGML_OP_GATED_LINEAR_ATTN
GGML_OP_GET_REL_POS
GGML_OP_GET_ROWS
GGML_OP_GET_ROWS_BACK
GGML_OP_GLU
GGML_OP_GROUP_NORM
GGML_OP_IM2COL
GGML_OP_IM2COL_3D
GGML_OP_IM2COL_BACK
GGML_OP_L2_NORM
GGML_OP_LEAKY_RELU
GGML_OP_LIGHTNING_INDEXER
GGML_OP_LOG
GGML_OP_MAP_CUSTOM1
GGML_OP_MAP_CUSTOM2
GGML_OP_MAP_CUSTOM3
GGML_OP_MEAN
GGML_OP_MUL
GGML_OP_MUL_MAT
GGML_OP_MUL_MAT_ID
GGML_OP_NONE
GGML_OP_NORM
GGML_OP_OPT_STEP_ADAMW
GGML_OP_OPT_STEP_SGD
GGML_OP_OUT_PROD
GGML_OP_PAD
GGML_OP_PAD_REFLECT_1D
GGML_OP_PERMUTE
GGML_OP_POOL_1D
GGML_OP_POOL_2D
GGML_OP_POOL_2D_BACK
GGML_OP_POOL_AVG
GGML_OP_POOL_COUNT
GGML_OP_POOL_MAX
GGML_OP_REPEAT
GGML_OP_REPEAT_BACK
GGML_OP_RESHAPE
GGML_OP_RMS_NORM
GGML_OP_RMS_NORM_BACK
GGML_OP_ROLL
GGML_OP_ROPE
GGML_OP_ROPE_BACK
GGML_OP_RWKV_WKV6
GGML_OP_RWKV_WKV7
GGML_OP_SCALE
GGML_OP_SET
GGML_OP_SET_ROWS
GGML_OP_SILU_BACK
GGML_OP_SIN
GGML_OP_SOFT_MAX
GGML_OP_SOFT_MAX_BACK
GGML_OP_SOLVE_TRI
GGML_OP_SQR
GGML_OP_SQRT
GGML_OP_SSM_CONV
GGML_OP_SSM_SCAN
GGML_OP_SUB
GGML_OP_SUM
GGML_OP_SUM_ROWS
GGML_OP_TIMESTEP_EMBEDDING
GGML_OP_TOP_K
GGML_OP_TOP_K_MASK
GGML_OP_TRANSPOSE
GGML_OP_TRI
GGML_OP_UNARY
GGML_OP_UPSCALE
GGML_OP_VIEW
GGML_OP_WIN_PART
GGML_OP_WIN_UNPART
GGML_PREC_BF16
GGML_PREC_DEFAULT
GGML_PREC_F16
GGML_PREC_F32
GGML_PREC_Q4
GGML_PREC_Q8
GGML_PREC_UNDEFINED
GGML_SCALE_FLAG_ALIGN_CORNERS
GGML_SCALE_FLAG_ANTIALIAS
GGML_SCALE_MODE_BICUBIC
GGML_SCALE_MODE_BILINEAR
GGML_SCALE_MODE_COUNT
GGML_SCALE_MODE_NEAREST
GGML_SCHED_PRIO_HIGH
GGML_SCHED_PRIO_LOW
GGML_SCHED_PRIO_MEDIUM
GGML_SCHED_PRIO_NORMAL
GGML_SCHED_PRIO_REALTIME
GGML_SORT_ORDER_ASC
GGML_SORT_ORDER_DESC
GGML_STATUS_ABORTED
GGML_STATUS_ALLOC_FAILED
GGML_STATUS_FAILED
GGML_STATUS_SUCCESS
GGML_TENSOR_FLAG_COMPUTE
GGML_TENSOR_FLAG_INPUT
GGML_TENSOR_FLAG_LOSS
GGML_TENSOR_FLAG_OUTPUT
GGML_TENSOR_FLAG_PARAM
GGML_TRI_TYPE_LOWER
GGML_TRI_TYPE_LOWER_DIAG
GGML_TRI_TYPE_UPPER
GGML_TRI_TYPE_UPPER_DIAG
GGML_TYPE_BF16
GGML_TYPE_COUNT
GGML_TYPE_F8_E4M3
GGML_TYPE_F16
GGML_TYPE_F32
GGML_TYPE_F64
GGML_TYPE_I8
GGML_TYPE_I16
GGML_TYPE_I32
GGML_TYPE_I64
GGML_TYPE_IQ1_M
GGML_TYPE_IQ1_S
GGML_TYPE_IQ2_S
GGML_TYPE_IQ2_XS
GGML_TYPE_IQ2_XXS
GGML_TYPE_IQ3_S
GGML_TYPE_IQ3_XXS
GGML_TYPE_IQ4_NL
GGML_TYPE_IQ4_XS
GGML_TYPE_MXFP4
GGML_TYPE_NVFP4
GGML_TYPE_Q1_0
GGML_TYPE_Q2_0
GGML_TYPE_Q2_K
GGML_TYPE_Q3_K
GGML_TYPE_Q4_0
GGML_TYPE_Q4_1
GGML_TYPE_Q4_K
GGML_TYPE_Q5_0
GGML_TYPE_Q5_1
GGML_TYPE_Q5_K
GGML_TYPE_Q6_K
GGML_TYPE_Q8_0
GGML_TYPE_Q8_1
GGML_TYPE_Q8_K
GGML_TYPE_TQ1_0
GGML_TYPE_TQ2_0
GGML_UNARY_OP_ABS
GGML_UNARY_OP_CEIL
GGML_UNARY_OP_COUNT
GGML_UNARY_OP_ELU
GGML_UNARY_OP_EXP
GGML_UNARY_OP_EXPM1
GGML_UNARY_OP_FLOOR
GGML_UNARY_OP_GELU
GGML_UNARY_OP_GELU_ERF
GGML_UNARY_OP_GELU_QUICK
GGML_UNARY_OP_HARDSIGMOID
GGML_UNARY_OP_HARDSWISH
GGML_UNARY_OP_NEG
GGML_UNARY_OP_RELU
GGML_UNARY_OP_ROUND
GGML_UNARY_OP_SGN
GGML_UNARY_OP_SIGMOID
GGML_UNARY_OP_SILU
GGML_UNARY_OP_SOFTPLUS
GGML_UNARY_OP_STEP
GGML_UNARY_OP_TANH
GGML_UNARY_OP_TRUNC
GGML_UNARY_OP_XIELU
LLAMA_ATTENTION_TYPE_CAUSAL
LLAMA_ATTENTION_TYPE_NON_CAUSAL
LLAMA_ATTENTION_TYPE_UNSPECIFIED
LLAMA_CONTEXT_TYPE_DEFAULT
LLAMA_CONTEXT_TYPE_MTP
LLAMA_DECISION_ORDER_NONE
LLAMA_DECISION_ORDER_OPTION
LLAMA_DECISION_ORDER_QUESTION_CHOICE
LLAMA_DECISION_ORDER_QUESTION_NOUL
LLAMA_DECISION_ORDER_QUESTION_SCORE
LLAMA_DEFAULT_SEED
LLAMA_DRY_MAX_SEQ_BREAKERS
LLAMA_EXAMPLE_BATCHED
LLAMA_EXAMPLE_BENCH
LLAMA_EXAMPLE_CLI
LLAMA_EXAMPLE_COMMON
LLAMA_EXAMPLE_COMPLETION
LLAMA_EXAMPLE_COUNT
LLAMA_EXAMPLE_CVECTOR_GENERATOR
LLAMA_EXAMPLE_DEBUG
LLAMA_EXAMPLE_DIFFUSION
LLAMA_EXAMPLE_DOWNLOAD
LLAMA_EXAMPLE_EMBEDDING
LLAMA_EXAMPLE_EXPORT_GRAPH_OPS
LLAMA_EXAMPLE_EXPORT_LORA
LLAMA_EXAMPLE_FINETUNE
LLAMA_EXAMPLE_FIT_PARAMS
LLAMA_EXAMPLE_IMATRIX
LLAMA_EXAMPLE_LOOKUP
LLAMA_EXAMPLE_MTMD
LLAMA_EXAMPLE_PARALLEL
LLAMA_EXAMPLE_PASSKEY
LLAMA_EXAMPLE_PERPLEXITY
LLAMA_EXAMPLE_RESULTS
LLAMA_EXAMPLE_RETRIEVAL
LLAMA_EXAMPLE_SERVER
LLAMA_EXAMPLE_SPECULATIVE
LLAMA_EXAMPLE_TOKENIZE
LLAMA_EXAMPLE_TTS
LLAMA_FILE_MAGIC_GGLA
LLAMA_FILE_MAGIC_GGSN
LLAMA_FILE_MAGIC_GGSQ
LLAMA_FLASH_ATTN_TYPE_AUTO
LLAMA_FLASH_ATTN_TYPE_DISABLED
LLAMA_FLASH_ATTN_TYPE_ENABLED
LLAMA_FTYPE_ALL_F32
LLAMA_FTYPE_GUESSED
LLAMA_FTYPE_MOSTLY_BF16
LLAMA_FTYPE_MOSTLY_F8_E4M3
LLAMA_FTYPE_MOSTLY_F16
LLAMA_FTYPE_MOSTLY_IQ1_M
LLAMA_FTYPE_MOSTLY_IQ1_S
LLAMA_FTYPE_MOSTLY_IQ2_M
LLAMA_FTYPE_MOSTLY_IQ2_S
LLAMA_FTYPE_MOSTLY_IQ2_XS
LLAMA_FTYPE_MOSTLY_IQ2_XXS
LLAMA_FTYPE_MOSTLY_IQ3_M
LLAMA_FTYPE_MOSTLY_IQ3_S
LLAMA_FTYPE_MOSTLY_IQ3_XS
LLAMA_FTYPE_MOSTLY_IQ3_XXS
LLAMA_FTYPE_MOSTLY_IQ4_NL
LLAMA_FTYPE_MOSTLY_IQ4_XS
LLAMA_FTYPE_MOSTLY_MXFP4_MOE
LLAMA_FTYPE_MOSTLY_NVFP4
LLAMA_FTYPE_MOSTLY_Q1_0
LLAMA_FTYPE_MOSTLY_Q2_0
LLAMA_FTYPE_MOSTLY_Q2_K
LLAMA_FTYPE_MOSTLY_Q2_K_S
LLAMA_FTYPE_MOSTLY_Q3_K_L
LLAMA_FTYPE_MOSTLY_Q3_K_M
LLAMA_FTYPE_MOSTLY_Q3_K_S
LLAMA_FTYPE_MOSTLY_Q4_0
LLAMA_FTYPE_MOSTLY_Q4_1
LLAMA_FTYPE_MOSTLY_Q4_K_M
LLAMA_FTYPE_MOSTLY_Q4_K_S
LLAMA_FTYPE_MOSTLY_Q5_0
LLAMA_FTYPE_MOSTLY_Q5_1
LLAMA_FTYPE_MOSTLY_Q5_K_M
LLAMA_FTYPE_MOSTLY_Q5_K_S
LLAMA_FTYPE_MOSTLY_Q6_K
LLAMA_FTYPE_MOSTLY_Q8_0
LLAMA_FTYPE_MOSTLY_TQ1_0
LLAMA_FTYPE_MOSTLY_TQ2_0
LLAMA_KV_OVERRIDE_TYPE_BOOL
LLAMA_KV_OVERRIDE_TYPE_FLOAT
LLAMA_KV_OVERRIDE_TYPE_INT
LLAMA_KV_OVERRIDE_TYPE_STR
LLAMA_LAZY_MODE_AUTO
LLAMA_LAZY_MODE_OFF
LLAMA_LAZY_MODE_ON
LLAMA_LOAD_MODE_AUTO
LLAMA_LOAD_MODE_DIRECT_IO
LLAMA_LOAD_MODE_MLOCK
LLAMA_LOAD_MODE_MMAP
LLAMA_LOAD_MODE_MMAP_MLOCK
LLAMA_LOAD_MODE_NONE
LLAMA_MODEL_META_KEY_SAMPLING_MIN_P
LLAMA_MODEL_META_KEY_SAMPLING_MIROSTAT
LLAMA_MODEL_META_KEY_SAMPLING_MIROSTAT_ETA
LLAMA_MODEL_META_KEY_SAMPLING_MIROSTAT_TAU
LLAMA_MODEL_META_KEY_SAMPLING_PENALTY_LAST_N
LLAMA_MODEL_META_KEY_SAMPLING_PENALTY_REPEAT
LLAMA_MODEL_META_KEY_SAMPLING_SEQUENCE
LLAMA_MODEL_META_KEY_SAMPLING_TEMP
LLAMA_MODEL_META_KEY_SAMPLING_TOP_K
LLAMA_MODEL_META_KEY_SAMPLING_TOP_P
LLAMA_MODEL_META_KEY_SAMPLING_XTC_PROBABILITY
LLAMA_MODEL_META_KEY_SAMPLING_XTC_THRESHOLD
LLAMA_POOLING_TYPE_CLS
LLAMA_POOLING_TYPE_LAST
LLAMA_POOLING_TYPE_MEAN
LLAMA_POOLING_TYPE_NONE
LLAMA_POOLING_TYPE_RANK
LLAMA_POOLING_TYPE_UNSPECIFIED
LLAMA_PROCESS_TYPE_DECODE
LLAMA_PROCESS_TYPE_ENCODE
LLAMA_ROPE_SCALING_TYPE_LINEAR
LLAMA_ROPE_SCALING_TYPE_LONGROPE
LLAMA_ROPE_SCALING_TYPE_MAX_VALUE
LLAMA_ROPE_SCALING_TYPE_NONE
LLAMA_ROPE_SCALING_TYPE_UNSPECIFIED
LLAMA_ROPE_SCALING_TYPE_YARN
LLAMA_ROPE_TYPE_IMROPE
LLAMA_ROPE_TYPE_MROPE
LLAMA_ROPE_TYPE_NEOX
LLAMA_ROPE_TYPE_NONE
LLAMA_ROPE_TYPE_NORM
LLAMA_ROPE_TYPE_VISION
LLAMA_SESSION_MAGIC
LLAMA_SESSION_VERSION
LLAMA_SHIM_BAD_JSON
LLAMA_SHIM_BUFFER_TOO_SMALL
LLAMA_SHIM_INVALID_ARG
LLAMA_SHIM_OK
LLAMA_SHIM_THROWN
LLAMA_SPLIT_MODE_LAYER
LLAMA_SPLIT_MODE_NONE
LLAMA_SPLIT_MODE_ROW
LLAMA_SPLIT_MODE_TENSOR
LLAMA_STATE_SEQ_FLAGS_NONE
LLAMA_STATE_SEQ_FLAGS_ON_DEVICE
LLAMA_STATE_SEQ_FLAGS_PARTIAL_ONLY
LLAMA_STATE_SEQ_FLAGS_SWA_ONLY
LLAMA_STATE_SEQ_MAGIC
LLAMA_STATE_SEQ_VERSION
LLAMA_TOKEN_ATTR_BYTE
LLAMA_TOKEN_ATTR_CONTROL
LLAMA_TOKEN_ATTR_LSTRIP
LLAMA_TOKEN_ATTR_NORMAL
LLAMA_TOKEN_ATTR_NORMALIZED
LLAMA_TOKEN_ATTR_RSTRIP
LLAMA_TOKEN_ATTR_SINGLE_WORD
LLAMA_TOKEN_ATTR_UNDEFINED
LLAMA_TOKEN_ATTR_UNKNOWN
LLAMA_TOKEN_ATTR_UNUSED
LLAMA_TOKEN_ATTR_USER_DEFINED
LLAMA_TOKEN_NULL
LLAMA_TOKEN_TYPE_BYTE
LLAMA_TOKEN_TYPE_CONTROL
LLAMA_TOKEN_TYPE_NORMAL
LLAMA_TOKEN_TYPE_UNDEFINED
LLAMA_TOKEN_TYPE_UNKNOWN
LLAMA_TOKEN_TYPE_UNUSED
LLAMA_TOKEN_TYPE_USER_DEFINED
LLAMA_VOCAB_TYPE_BPE
LLAMA_VOCAB_TYPE_NONE
LLAMA_VOCAB_TYPE_PLAMO2
LLAMA_VOCAB_TYPE_PLAMO3
LLAMA_VOCAB_TYPE_RWKV
LLAMA_VOCAB_TYPE_SPM
LLAMA_VOCAB_TYPE_TEST
LLAMA_VOCAB_TYPE_UGM
LLAMA_VOCAB_TYPE_WPM
LLM_ARCH_AFMOE
LLM_ARCH_APERTUS
LLM_ARCH_ARCEE
LLM_ARCH_ARCTIC
LLM_ARCH_ARWKV7
LLM_ARCH_BAICHUAN
LLM_ARCH_BAILINGMOE
LLM_ARCH_BAILINGMOE2
LLM_ARCH_BAILINGMOE3
LLM_ARCH_BEAM
LLM_ARCH_BERT
LLM_ARCH_BITNET
LLM_ARCH_BLOOM
LLM_ARCH_CHAMELEON
LLM_ARCH_CHATGLM
LLM_ARCH_CLEF
LLM_ARCH_CLIP
LLM_ARCH_CODESHELL
LLM_ARCH_COGVLM
LLM_ARCH_COHERE2
LLM_ARCH_COHERE2MOE
LLM_ARCH_COMMAND_R
LLM_ARCH_DBRX
LLM_ARCH_DECI
LLM_ARCH_DEEPSEEK
LLM_ARCH_DEEPSEEK2
LLM_ARCH_DEEPSEEK4
LLM_ARCH_DEEPSEEK2OCR
LLM_ARCH_DEEPSEEK32
LLM_ARCH_DEEPSEEK41
LLM_ARCH_DFLASH
LLM_ARCH_DOTS1
LLM_ARCH_DOTS3NOTE
LLM_ARCH_DREAM
LLM_ARCH_EAGLE3
LLM_ARCH_ERNIE4_5
LLM_ARCH_ERNIE4_5_MOE
LLM_ARCH_EUROBERT
LLM_ARCH_EXAONE
LLM_ARCH_EXAONE4
LLM_ARCH_EXAONE_MOE
LLM_ARCH_FALCON
LLM_ARCH_FALCON_H1
LLM_ARCH_FLINT
LLM_ARCH_GEMMA
LLM_ARCH_GEMMA2
LLM_ARCH_GEMMA3
LLM_ARCH_GEMMA4
LLM_ARCH_GEMMA3N
LLM_ARCH_GEMMA4_ASSISTANT
LLM_ARCH_GEMMA_EMBEDDING
LLM_ARCH_GEMMA_EMBEDDING2
LLM_ARCH_GLM4
LLM_ARCH_GLM4_MOE
LLM_ARCH_GLM5_NEXT
LLM_ARCH_GLM_DSA
LLM_ARCH_GPT2
LLM_ARCH_GPTJ
LLM_ARCH_GPTNEOX
LLM_ARCH_GRANITE
LLM_ARCH_GRANITE_HYBRID
LLM_ARCH_GRANITE_MOE
LLM_ARCH_GRANITE_SWA
LLM_ARCH_GRANITE_SWITCH
LLM_ARCH_GROK
LLM_ARCH_GROVEMOE
LLM_ARCH_HRM_TEXT
LLM_ARCH_HUNYUAN_DENSE
LLM_ARCH_HUNYUAN_MOE
LLM_ARCH_HUNYUAN_VL
LLM_ARCH_HY_V3
LLM_ARCH_HY_V4
LLM_ARCH_INTERNLM2
LLM_ARCH_JAIS
LLM_ARCH_JAIS2
LLM_ARCH_JAMBA
LLM_ARCH_JINA_BERT_V2
LLM_ARCH_JINA_BERT_V3
LLM_ARCH_K2_HORIZON
LLM_ARCH_KIMI_K3
LLM_ARCH_KIMI_LINEAR
LLM_ARCH_KOLIBRI1
LLM_ARCH_LAGUNA
LLM_ARCH_LFM2
LLM_ARCH_LFM3
LLM_ARCH_LFM2MOE
LLM_ARCH_LIMITE
LLM_ARCH_LLADA
LLM_ARCH_LLADA_MOE
LLM_ARCH_LLAMA
LLM_ARCH_LLAMA4
LLM_ARCH_LLAMA_EMBED
LLM_ARCH_MAINCODER
LLM_ARCH_MAION_CODER
LLM_ARCH_MAMBA
LLM_ARCH_MAMBA2
LLM_ARCH_MAPLE
LLM_ARCH_MELLUM
LLM_ARCH_MIMO2
LLM_ARCH_MINICPM
LLM_ARCH_MINICPM3
LLM_ARCH_MINIMAX_01
LLM_ARCH_MINIMAX_M2
LLM_ARCH_MINIMAX_M3
LLM_ARCH_MISTRAL3
LLM_ARCH_MISTRAL4
LLM_ARCH_MISTRAL_LARGE4
LLM_ARCH_MODERN_BERT
LLM_ARCH_MPT
LLM_ARCH_MUSE_GLIMMER
LLM_ARCH_MUSE_SPARK
LLM_ARCH_NANBEIGE
LLM_ARCH_NEMOTRON
LLM_ARCH_NEMOTRON_H
LLM_ARCH_NEMOTRON_H_MOE
LLM_ARCH_NEO_BERT
LLM_ARCH_NOMIC_BERT
LLM_ARCH_NOMIC_BERT_MOE
LLM_ARCH_OLMO
LLM_ARCH_OLMO2
LLM_ARCH_OLMOE
LLM_ARCH_OPENAI_MOE
LLM_ARCH_OPENELM
LLM_ARCH_ORION
LLM_ARCH_PADDLEOCR
LLM_ARCH_PANGU_EMBED
LLM_ARCH_PHI2
LLM_ARCH_PHI3
LLM_ARCH_PHIMOE
LLM_ARCH_PLAMO
LLM_ARCH_PLAMO2
LLM_ARCH_PLAMO3
LLM_ARCH_PLM
LLM_ARCH_POCKETTTS
LLM_ARCH_QWEN
LLM_ARCH_QWEN2
LLM_ARCH_QWEN3
LLM_ARCH_QWEN2MOE
LLM_ARCH_QWEN2VL
LLM_ARCH_QWEN3MOE
LLM_ARCH_QWEN3NEXT
LLM_ARCH_QWEN3TTS
LLM_ARCH_QWEN3VL
LLM_ARCH_QWEN3VLMOE
LLM_ARCH_QWEN4EXP
LLM_ARCH_QWEN35
LLM_ARCH_QWEN35MOE
LLM_ARCH_REFACT
LLM_ARCH_RND1
LLM_ARCH_RWKV6
LLM_ARCH_RWKV7
LLM_ARCH_RWKV6QWEN2
LLM_ARCH_SEED_OSS
LLM_ARCH_SENSENOVA_U1
LLM_ARCH_SMALLTHINKER
LLM_ARCH_SMOLLM3
LLM_ARCH_SPARK2_5
LLM_ARCH_STABLELM
LLM_ARCH_STARCODER
LLM_ARCH_STARCODER2
LLM_ARCH_STEP35
LLM_ARCH_T5
LLM_ARCH_T5ENCODER
LLM_ARCH_TALKIE
LLM_ARCH_UNKNOWN
LLM_ARCH_VLLM_DECISION2
LLM_ARCH_WAVTOKENIZER_DEC
LLM_ARCH_XVERSE
LLM_KV_ACTIVATION_SITU_BETA
LLM_KV_ACTIVATION_SITU_LINEAR_BETA
LLM_KV_ADAPTER_ALORA_INVOCATION_TOKENS
LLM_KV_ADAPTER_COUNT
LLM_KV_ADAPTER_LORA_ALPHA
LLM_KV_ADAPTER_LORA_PROMPT_PREFIX
LLM_KV_ADAPTER_LORA_RANK
LLM_KV_ADAPTER_LORA_TASK_NAME
LLM_KV_ADAPTER_ROUTER_GAIN
LLM_KV_ADAPTER_TOKEN_IDS_ACTIVATE
LLM_KV_ADAPTER_TOKEN_IDS_SUBSTITUTE
LLM_KV_ADAPTER_TYPE
LLM_KV_ATTENTION_CAUSAL
LLM_KV_ATTENTION_CLAMP_KQV
LLM_KV_ATTENTION_COMPRESS_RATIOS
LLM_KV_ATTENTION_COMPRESS_ROPE_FREQ_BASE
LLM_KV_ATTENTION_DECAY_LORA_RANK
LLM_KV_ATTENTION_GATE_LORA_RANK
LLM_KV_ATTENTION_GROUPNORM_EPS
LLM_KV_ATTENTION_GROUPNORM_GROUPS
LLM_KV_ATTENTION_HEAD_COUNT
LLM_KV_ATTENTION_HEAD_COUNT_KV
LLM_KV_ATTENTION_ICLR_LORA_RANK
LLM_KV_ATTENTION_INDEXER_BLOCK_SIZE
LLM_KV_ATTENTION_INDEXER_CANDIDATE_BLOCK_SIZE
LLM_KV_ATTENTION_INDEXER_CANDIDATE_SOURCE_LAYER
LLM_KV_ATTENTION_INDEXER_CANDIDATE_TOP_K_BLOCKS
LLM_KV_ATTENTION_INDEXER_HEAD_COUNT
LLM_KV_ATTENTION_INDEXER_KEY_LENGTH
LLM_KV_ATTENTION_INDEXER_KPOOL
LLM_KV_ATTENTION_INDEXER_KPOOL_SELECT_TAIL
LLM_KV_ATTENTION_INDEXER_LOCAL_BLOCKS
LLM_KV_ATTENTION_INDEXER_SOURCE_LAYERS
LLM_KV_ATTENTION_INDEXER_TOP_K
LLM_KV_ATTENTION_INDEXER_TYPES
LLM_KV_ATTENTION_KEY_LENGTH
LLM_KV_ATTENTION_KEY_LENGTH_MLA
LLM_KV_ATTENTION_KEY_LENGTH_MLA_SWA
LLM_KV_ATTENTION_KEY_LENGTH_SWA
LLM_KV_ATTENTION_KV_LORA_RANK
LLM_KV_ATTENTION_KV_LORA_RANK_SWA
LLM_KV_ATTENTION_KV_SOURCE_LAYERS
LLM_KV_ATTENTION_KV_SOURCE_MAP
LLM_KV_ATTENTION_LAYERNORM_EPS
LLM_KV_ATTENTION_LAYERNORM_RMS_EPS
LLM_KV_ATTENTION_MAX_ALIBI_BIAS
LLM_KV_ATTENTION_OUTPUT_GROUP_COUNT
LLM_KV_ATTENTION_OUTPUT_LORA_RANK
LLM_KV_ATTENTION_OUTPUT_SCALE
LLM_KV_ATTENTION_Q_LORA_RANK
LLM_KV_ATTENTION_RECURRENT_LAYERS
LLM_KV_ATTENTION_RELATIVE_BUCKETS_COUNT
LLM_KV_ATTENTION_ROPE_PATTERN
LLM_KV_ATTENTION_SCALE
LLM_KV_ATTENTION_SHARED_KV_LAYERS
LLM_KV_ATTENTION_SLIDING_WINDOW
LLM_KV_ATTENTION_SLIDING_WINDOW_PATTERN
LLM_KV_ATTENTION_TEMPERATURE_LENGTH
LLM_KV_ATTENTION_TEMPERATURE_SCALE
LLM_KV_ATTENTION_VALUE_EXPERT_COUNT
LLM_KV_ATTENTION_VALUE_EXPERT_USED_COUNT
LLM_KV_ATTENTION_VALUE_LENGTH
LLM_KV_ATTENTION_VALUE_LENGTH_MLA
LLM_KV_ATTENTION_VALUE_LENGTH_MLA_SWA
LLM_KV_ATTENTION_VALUE_LENGTH_SWA
LLM_KV_ATTENTION_VALUE_RESIDUAL_MIX_LORA_RANK
LLM_KV_ATTENTION_VALUE_SCALE
LLM_KV_ATTENTION_XSA_EPS
LLM_KV_ATTN_LOGIT_SOFTCAPPING
LLM_KV_ATTN_RES_BLOCK_SIZE
LLM_KV_BLOCK_COUNT
LLM_KV_CLASSIFIER_ACTIVATION
LLM_KV_CLASSIFIER_OUTPUT_LABELS
LLM_KV_CLASSIFIER_POOLING_TYPE
LLM_KV_CONTEXT_LENGTH
LLM_KV_CONVNEXT_BLOCK_COUNT
LLM_KV_CONVNEXT_EMBEDDING_LENGTH
LLM_KV_DECISION_BLOCK_COUNT
LLM_KV_DECISION_HEAD_COUNT
LLM_KV_DECISION_LABELS
LLM_KV_DECISION_ROUTING_BLOCK_COUNT
LLM_KV_DECODER_BLOCK_COUNT
LLM_KV_DECODER_START_TOKEN_ID
LLM_KV_DEEPSTACK_MAPPING
LLM_KV_DENSE_2_FEAT_IN
LLM_KV_DENSE_2_FEAT_OUT
LLM_KV_DENSE_3_FEAT_IN
LLM_KV_DENSE_3_FEAT_OUT
LLM_KV_DFLASH_BLOCK_SIZE
LLM_KV_DFLASH_CONV_GROUP_SIZE
LLM_KV_DFLASH_CONV_KERNEL_SIZE
LLM_KV_DFLASH_SELECTOR_RANK
LLM_KV_DFLASH_SELECTOR_TOP_K
LLM_KV_EMBEDDING_LENGTH
LLM_KV_EMBEDDING_LENGTH_OUT
LLM_KV_EMBEDDING_LENGTH_PER_LAYER
LLM_KV_EMBEDDING_SCALE
LLM_KV_ENGRAM_PAD_TOKEN_ID
LLM_KV_ENGRAM_TOKEN_MAP
LLM_KV_EXPERTS_PER_GROUP
LLM_KV_EXPERT_CHUNK_FEED_FORWARD_LENGTH
LLM_KV_EXPERT_COUNT
LLM_KV_EXPERT_FEED_FORWARD_LENGTH
LLM_KV_EXPERT_GATING_FUNC
LLM_KV_EXPERT_GROUP_COUNT
LLM_KV_EXPERT_GROUP_SCALE
LLM_KV_EXPERT_GROUP_USED_COUNT
LLM_KV_EXPERT_LATENT_LENGTH
LLM_KV_EXPERT_SHARED_COUNT
LLM_KV_EXPERT_SHARED_FEED_FORWARD_LENGTH
LLM_KV_EXPERT_USED_COUNT
LLM_KV_EXPERT_WEIGHTS_NORM
LLM_KV_EXPERT_WEIGHTS_SCALE
LLM_KV_FEATURES_LENGTH
LLM_KV_FEED_FORWARD_LENGTH
LLM_KV_FINAL_LOGIT_SIGMOID_CAPPING
LLM_KV_FINAL_LOGIT_SOFTCAPPING
LLM_KV_FULL_ATTENTION_INTERVAL
LLM_KV_GENERAL_ALIGNMENT
LLM_KV_GENERAL_ARCHITECTURE
LLM_KV_GENERAL_AUTHOR
LLM_KV_GENERAL_DESCRIPTION
LLM_KV_GENERAL_FILE_TYPE
LLM_KV_GENERAL_LICENSE
LLM_KV_GENERAL_NAME
LLM_KV_GENERAL_QUANTIZATION_VERSION
LLM_KV_GENERAL_SAMPLING_MIN_P
LLM_KV_GENERAL_SAMPLING_MIROSTAT
LLM_KV_GENERAL_SAMPLING_MIROSTAT_ETA
LLM_KV_GENERAL_SAMPLING_MIROSTAT_TAU
LLM_KV_GENERAL_SAMPLING_PENALTY_LAST_N
LLM_KV_GENERAL_SAMPLING_PENALTY_REPEAT
LLM_KV_GENERAL_SAMPLING_SEQUENCE
LLM_KV_GENERAL_SAMPLING_TEMP
LLM_KV_GENERAL_SAMPLING_TOP_K
LLM_KV_GENERAL_SAMPLING_TOP_P
LLM_KV_GENERAL_SAMPLING_XTC_PROBABILITY
LLM_KV_GENERAL_SAMPLING_XTC_THRESHOLD
LLM_KV_GENERAL_SOURCE_HF_REPO
LLM_KV_GENERAL_SOURCE_URL
LLM_KV_GENERAL_TENSOR_EXTRA_NAME
LLM_KV_GENERAL_TENSOR_EXTRA_PREC_A4
LLM_KV_GENERAL_TYPE
LLM_KV_GENERAL_URL
LLM_KV_GENERAL_VERSION
LLM_KV_HASH_LAYER_COUNT
LLM_KV_HIDDEN_ACT
LLM_KV_HRM_H_CYCLES
LLM_KV_HRM_LAYERS_PER_STACK
LLM_KV_HRM_L_CYCLES
LLM_KV_HRM_PREFIX_LM
LLM_KV_HYPER_CONNECTION_COUNT
LLM_KV_HYPER_CONNECTION_EPSILON
LLM_KV_HYPER_CONNECTION_LOW_RANK
LLM_KV_HYPER_CONNECTION_MAGNITUDE
LLM_KV_HYPER_CONNECTION_SINGLE_PASS
LLM_KV_HYPER_CONNECTION_SINKHORN_ITERATIONS
LLM_KV_INTERLEAVE_MOE_LAYER_STEP
LLM_KV_KDA_GATE_LOWER_BOUND
LLM_KV_KDA_HEAD_DIM
LLM_KV_KDA_SAFE_GATE
LLM_KV_LEADING_DENSE_BLOCK_COUNT
LLM_KV_LILICORR_CANDIDATE_TOPK
LLM_KV_LILICORR_FACTOR_DIM
LLM_KV_LILICORR_HIDDEN_SIZE
LLM_KV_LILICORR_LOGIT_SCALE
LLM_KV_LILICORR_MLP_HIDDEN_SIZE
LLM_KV_LILICORR_NUM_HEADS
LLM_KV_LILICORR_NUM_LAYERS
LLM_KV_LILICORR_VECTOR_EPS
LLM_KV_LOGIT_SCALE
LLM_KV_MOE_EVERY_N_LAYERS
LLM_KV_MOE_LATENT_SIZE
LLM_KV_MUDD_FEED_FORWARD_LENGTH
LLM_KV_MUDD_TAP_COUNT
LLM_KV_MUDD_TAP_INDICES
LLM_KV_NEXTN_PREDICT_LAYERS
LLM_KV_NORM_BEFORE_FC
LLM_KV_NORM_BEFORE_RESIDUAL
LLM_KV_NUM_DEEPSTACK_LAYERS
LLM_KV_NUM_LOOPS
LLM_KV_PLE_CONV_KERNEL
LLM_KV_PLE_EOS_TOKEN_ID
LLM_KV_PLE_HEADS_PER_NGRAM
LLM_KV_PLE_HEAD_OFFSETS
LLM_KV_PLE_HEAD_VOCAB_SIZES
LLM_KV_PLE_IMAGE_TOKEN_ID
LLM_KV_PLE_LAYERS
LLM_KV_PLE_LAYER_MULTIPLIERS
LLM_KV_PLE_NGRAM_SIZE
LLM_KV_POOLING_TYPE
LLM_KV_POSNET_BLOCK_COUNT
LLM_KV_POSNET_EMBEDDING_LENGTH
LLM_KV_POST_NORM_EPS
LLM_KV_PRISM_HADAMARD_AXIS
LLM_KV_PRISM_HADAMARD_BLOCK_SIZE
LLM_KV_PRISM_HADAMARD_GDN_V_GROUPED
LLM_KV_PRISM_HADAMARD_INVERSE_WEIGHT_NAMES
LLM_KV_PRISM_HADAMARD_SIGN_MODE
LLM_KV_PRISM_HADAMARD_SIGN_VALUES
LLM_KV_PRISM_HADAMARD_SIGN_WIDTHS
LLM_KV_PRISM_HADAMARD_TIED_OUTPUT
LLM_KV_PRISM_HADAMARD_TRANSFORM
LLM_KV_PRISM_HADAMARD_VERSION
LLM_KV_PRISM_HADAMARD_WEIGHT_NAMES
LLM_KV_REALTIME_DELAY_TOKENS
LLM_KV_RESCALE_EVERY_N_LAYERS
LLM_KV_RESIDUAL_SCALE
LLM_KV_ROPE_DIMENSION_COUNT
LLM_KV_ROPE_DIMENSION_COUNT_SWA
LLM_KV_ROPE_DIMENSION_SECTIONS
LLM_KV_ROPE_FREQ_BASE
LLM_KV_ROPE_FREQ_BASE_SPATIAL
LLM_KV_ROPE_FREQ_BASE_SWA
LLM_KV_ROPE_SCALE_LINEAR
LLM_KV_ROPE_SCALING_ALPHA
LLM_KV_ROPE_SCALING_ATTN_FACTOR
LLM_KV_ROPE_SCALING_FACTOR
LLM_KV_ROPE_SCALING_FINETUNED
LLM_KV_ROPE_SCALING_ORIG_CTX_LEN
LLM_KV_ROPE_SCALING_TYPE
LLM_KV_ROPE_SCALING_YARN_ATTN_FACTOR
LLM_KV_ROPE_SCALING_YARN_BETA_FAST
LLM_KV_ROPE_SCALING_YARN_BETA_SLOW
LLM_KV_ROPE_SCALING_YARN_EXT_FACTOR
LLM_KV_ROPE_SCALING_YARN_LOG_MUL
LLM_KV_ROPE_SECTION_ORDER
LLM_KV_ROUTER_LOGIT_SOFTCAPPING
LLM_KV_SHORTCONV_L_CACHE
LLM_KV_SKIP_LOOP_FINAL_NORM
LLM_KV_SPLIT_COUNT
LLM_KV_SPLIT_NO
LLM_KV_SPLIT_TENSORS_COUNT
LLM_KV_SSM_CONV_KERNEL
LLM_KV_SSM_DT_B_C_RMS
LLM_KV_SSM_GROUP_COUNT
LLM_KV_SSM_INNER_SIZE
LLM_KV_SSM_STATE_SIZE
LLM_KV_SSM_TIME_STEP_RANK
LLM_KV_SWIGLU_CLAMP_EXP
LLM_KV_SWIGLU_CLAMP_SHEXP
LLM_KV_SWIN_NORM
LLM_KV_TARGET_HIDDEN_SIZE
LLM_KV_TARGET_LAYERS
LLM_KV_TENSOR_DATA_LAYOUT
LLM_KV_TIME_DECAY_EXTRA_DIM
LLM_KV_TIME_MIX_EXTRA_DIM
LLM_KV_TOKENIZER_ADD_BOS
LLM_KV_TOKENIZER_ADD_EOS
LLM_KV_TOKENIZER_ADD_PREFIX
LLM_KV_TOKENIZER_ADD_SEP
LLM_KV_TOKENIZER_BOS_ID
LLM_KV_TOKENIZER_CHAT_TEMPLATE
LLM_KV_TOKENIZER_CLEAN_SPACES
LLM_KV_TOKENIZER_CLS_ID
LLM_KV_TOKENIZER_DNA_AUTO_TAGS
LLM_KV_TOKENIZER_EOM_ID
LLM_KV_TOKENIZER_EOS_ID
LLM_KV_TOKENIZER_EOT_ID
LLM_KV_TOKENIZER_FIM_MID_ID
LLM_KV_TOKENIZER_FIM_PAD_ID
LLM_KV_TOKENIZER_FIM_PRE_ID
LLM_KV_TOKENIZER_FIM_REP_ID
LLM_KV_TOKENIZER_FIM_SEP_ID
LLM_KV_TOKENIZER_FIM_SUF_ID
LLM_KV_TOKENIZER_HF_JSON
LLM_KV_TOKENIZER_LIST
LLM_KV_TOKENIZER_MASK_ID
LLM_KV_TOKENIZER_MERGES
LLM_KV_TOKENIZER_METASPACE_PREPEND_SCHEME
LLM_KV_TOKENIZER_MIDDLE_ID
LLM_KV_TOKENIZER_MODEL
LLM_KV_TOKENIZER_NORMALIZER_LOWERCASE
LLM_KV_TOKENIZER_NORMALIZER_NFC
LLM_KV_TOKENIZER_NORMALIZER_STRIP_ACCENTS
LLM_KV_TOKENIZER_PAD_ID
LLM_KV_TOKENIZER_PRE
LLM_KV_TOKENIZER_PRECOMPILED_CHARSMAP
LLM_KV_TOKENIZER_PREFIX_ID
LLM_KV_TOKENIZER_REMOVE_EXTRA_WS
LLM_KV_TOKENIZER_RWKV
LLM_KV_TOKENIZER_SCORES
LLM_KV_TOKENIZER_SEP_ID
LLM_KV_TOKENIZER_SUFFIX_ID
LLM_KV_TOKENIZER_SUPPRESS_TOKENS
LLM_KV_TOKENIZER_TOKEN_TYPE
LLM_KV_TOKENIZER_TOKEN_TYPE_COUNT
LLM_KV_TOKENIZER_UNK_ID
LLM_KV_TOKEN_SHIFT_COUNT
LLM_KV_USE_PARALLEL_RESIDUAL
LLM_KV_VOCAB_SIZE
LLM_KV_WKV_HEAD_SIZE
LLM_KV_XIELU_ALPHA_N
LLM_KV_XIELU_ALPHA_P
LLM_KV_XIELU_BETA
LLM_KV_XIELU_EPS
MTP_SPEC_TYPE_DFLASH
MTP_SPEC_TYPE_EAGLE3
MTP_SPEC_TYPE_MTP
MTP_STATE_STATUS_BAD_SEQUENCE
MTP_STATE_STATUS_BUFFER_SMALL
MTP_STATE_STATUS_EXCEPTION
MTP_STATE_STATUS_INVALID
MTP_STATE_STATUS_NOT_QUIESCENT
MTP_STATE_STATUS_NULL
MTP_STATE_STATUS_OK
MTP_STATE_STATUS_OVERFLOW
MTP_STATE_STATUS_UNAVAILABLE

Functions§

chat_shim_format_name^⚠
chat_shim_msgs_parse_oaicompat^⚠
chat_shim_parse^⚠
chat_shim_parser_free^⚠
chat_shim_templates_apply^⚠
chat_shim_templates_free^⚠
chat_shim_templates_get_caps^⚠
chat_shim_templates_init^⚠
chat_shim_templates_source^⚠
chat_shim_templates_support_enable_thinking^⚠
chat_shim_templates_was_explicit^⚠
chat_shim_tool_choice_parse_oaicompat^⚠
chat_shim_tools_parse_oaicompat^⚠
common_device_memory_collect^⚠
common_fit_params^⚠
common_fit_print^⚠
common_json_schema_to_grammar_c^⚠
common_memory_breakdown_print^⚠
common_shim_docker_resolve_model^⚠
common_shim_download_remove^⚠
common_shim_download_resolve_path^⚠
common_shim_download_split_repo_tag^⚠
common_shim_list_cached_models^⚠
common_shim_log_get_verbosity^⚠
common_shim_log_pause^⚠
common_shim_log_resume^⚠
common_shim_log_set_colors^⚠
common_shim_log_set_file^⚠
common_shim_log_set_jsonl^⚠
common_shim_log_set_prefix^⚠
common_shim_log_set_timestamps^⚠
common_shim_log_set_verbosity^⚠
common_shim_ngram_cache_draft^⚠
common_shim_ngram_cache_free^⚠
common_shim_ngram_cache_init^⚠
common_shim_ngram_cache_load^⚠
common_shim_ngram_cache_merge^⚠
common_shim_ngram_cache_save^⚠
common_shim_ngram_cache_size^⚠
common_shim_ngram_cache_update^⚠
common_shim_ngram_map_accept^⚠
common_shim_ngram_map_begin^⚠
common_shim_ngram_map_draft^⚠
common_shim_ngram_map_free^⚠
common_shim_ngram_map_init^⚠
common_shim_ngram_simple_draft^⚠
common_shim_reasoning_budget_force^⚠
common_shim_reasoning_budget_get_state^⚠
common_shim_reasoning_budget_init^⚠
common_shim_sampler_accept^⚠
common_shim_sampler_accept_raw^⚠
common_shim_sampler_clone^⚠
common_shim_sampler_free^⚠
common_shim_sampler_get^⚠
common_shim_sampler_get_seed^⚠
common_shim_sampler_init^⚠
common_shim_sampler_last^⚠
common_shim_sampler_params_add_grammar_trigger^⚠
common_shim_sampler_params_add_logit_bias^⚠
common_shim_sampler_params_free^⚠
common_shim_sampler_params_get_scalars^⚠
common_shim_sampler_params_init^⚠
common_shim_sampler_params_set_dry_breakers^⚠
common_shim_sampler_params_set_generation_prompt^⚠
common_shim_sampler_params_set_grammar^⚠
common_shim_sampler_params_set_reasoning_budget^⚠
common_shim_sampler_params_set_samplers^⚠
common_shim_sampler_params_set_scalars^⚠
common_shim_sampler_prev_str^⚠
common_shim_sampler_print^⚠
common_shim_sampler_reasoning_budget_force^⚠
common_shim_sampler_reset^⚠
common_shim_sampler_sample^⚠
common_shim_sampler_sample_and_accept_n^⚠
common_shim_sampler_sample_raw^⚠
common_shim_sampler_type_to_str^⚠
common_shim_sampler_types_from_names^⚠
common_shim_speculative_all_types_str^⚠
common_shim_speculative_type_from_name^⚠
common_shim_speculative_type_to_str^⚠
common_shim_speculative_types_from_gguf^⚠
common_token_to_piece^⚠
common_token_to_piece1^⚠
common_tokenize^⚠
common_tokenize1^⚠
ggml_abort^⚠
ggml_abs^⚠
ggml_abs_inplace^⚠
ggml_acc^⚠
ggml_acc_inplace^⚠
ggml_add^⚠
ggml_add1^⚠
ggml_add1_inplace^⚠
ggml_add_cast^⚠
ggml_add_id^⚠
ggml_add_inplace^⚠
ggml_add_rel_pos^⚠
ggml_add_rel_pos_inplace^⚠
ggml_arange^⚠
ggml_are_same_shape^⚠
ggml_are_same_stride^⚠
ggml_argmax^⚠
ggml_argsort^⚠
ggml_argsort_top_k^⚠
ggml_backend_alloc_buffer^⚠
ggml_backend_alloc_ctx_tensors^⚠
ggml_backend_alloc_ctx_tensors_from_buft^⚠
ggml_backend_alloc_ctx_tensors_from_buft_size^⚠
ggml_backend_buffer_clear^⚠
ggml_backend_buffer_free^⚠
ggml_backend_buffer_get_alignment^⚠
ggml_backend_buffer_get_alloc_size^⚠
ggml_backend_buffer_get_base^⚠
ggml_backend_buffer_get_max_size^⚠
ggml_backend_buffer_get_size^⚠
ggml_backend_buffer_get_type^⚠
ggml_backend_buffer_get_usage^⚠
ggml_backend_buffer_init_tensor^⚠
ggml_backend_buffer_is_host^⚠
ggml_backend_buffer_name^⚠
ggml_backend_buffer_reset^⚠
ggml_backend_buffer_set_usage^⚠
ggml_backend_buft_alloc_buffer^⚠
ggml_backend_buft_alloc_buffer_n^⚠
ggml_backend_buft_get_alignment^⚠
ggml_backend_buft_get_alloc_size^⚠
ggml_backend_buft_get_alloc_size_n^⚠
ggml_backend_buft_get_device^⚠
ggml_backend_buft_get_max_size^⚠
ggml_backend_buft_is_host^⚠
ggml_backend_buft_name^⚠
ggml_backend_compare_graph_backend^⚠
ggml_backend_cpu_buffer_from_ptr^⚠
ggml_backend_cpu_buffer_type^⚠
ggml_backend_cpu_init^⚠
ggml_backend_cpu_reg^⚠
ggml_backend_cpu_set_abort_callback^⚠
ggml_backend_cpu_set_n_threads^⚠
ggml_backend_cpu_set_threadpool^⚠
ggml_backend_cpu_set_use_ref^⚠
ggml_backend_dev_backend_reg^⚠
ggml_backend_dev_buffer_from_host_ptr^⚠
ggml_backend_dev_buffer_type^⚠
ggml_backend_dev_by_name^⚠
ggml_backend_dev_by_type^⚠
ggml_backend_dev_count^⚠
ggml_backend_dev_description^⚠
ggml_backend_dev_get^⚠
ggml_backend_dev_get_props^⚠
ggml_backend_dev_host_buffer_type^⚠
ggml_backend_dev_init^⚠
ggml_backend_dev_memory^⚠
ggml_backend_dev_name^⚠
ggml_backend_dev_offload_op^⚠
ggml_backend_dev_supports_buft^⚠
ggml_backend_dev_supports_op^⚠
ggml_backend_dev_type^⚠
ggml_backend_device_register^⚠
ggml_backend_event_free^⚠
ggml_backend_event_new^⚠
ggml_backend_event_record^⚠
ggml_backend_event_synchronize^⚠
ggml_backend_event_wait^⚠
ggml_backend_free^⚠
ggml_backend_get_alignment^⚠
ggml_backend_get_default_buffer_type^⚠
ggml_backend_get_device^⚠
ggml_backend_get_max_size^⚠
ggml_backend_graph_compute^⚠
ggml_backend_graph_compute_async^⚠
ggml_backend_graph_copy^⚠
ggml_backend_graph_copy_free^⚠
ggml_backend_graph_plan_compute^⚠
ggml_backend_graph_plan_create^⚠
ggml_backend_graph_plan_free^⚠
ggml_backend_guid^⚠
ggml_backend_init_best^⚠
ggml_backend_init_by_name^⚠
ggml_backend_init_by_type^⚠
ggml_backend_is_cpu^⚠
ggml_backend_load^⚠
ggml_backend_load_all^⚠
ggml_backend_load_all_from_path^⚠
ggml_backend_meta_device^⚠
ggml_backend_meta_split_axis_name^⚠
ggml_backend_name^⚠
ggml_backend_offload_op^⚠
ggml_backend_reg_by_name^⚠
ggml_backend_reg_count^⚠
ggml_backend_reg_dev_count^⚠
ggml_backend_reg_dev_get^⚠
ggml_backend_reg_get^⚠
ggml_backend_reg_get_proc_address^⚠
ggml_backend_reg_name^⚠
ggml_backend_register^⚠
ggml_backend_sched_alloc_graph^⚠
ggml_backend_sched_free^⚠
ggml_backend_sched_get_backend^⚠
ggml_backend_sched_get_buffer_size^⚠
ggml_backend_sched_get_buffer_type^⚠
ggml_backend_sched_get_n_backends^⚠
ggml_backend_sched_get_n_copies^⚠
ggml_backend_sched_get_n_splits^⚠
ggml_backend_sched_get_tensor_backend^⚠
ggml_backend_sched_graph_compute^⚠
ggml_backend_sched_graph_compute_async^⚠
ggml_backend_sched_new^⚠
ggml_backend_sched_reserve^⚠
ggml_backend_sched_reserve_size^⚠
ggml_backend_sched_reset^⚠
ggml_backend_sched_set_copy_callback^⚠
ggml_backend_sched_set_eval_callback^⚠
ggml_backend_sched_set_realloc_callback^⚠
ggml_backend_sched_set_tensor_backend^⚠
ggml_backend_sched_split_graph^⚠
ggml_backend_sched_synchronize^⚠
ggml_backend_supports_buft^⚠
ggml_backend_supports_op^⚠
ggml_backend_synchronize^⚠
ggml_backend_tensor_alloc^⚠
ggml_backend_tensor_copy^⚠
ggml_backend_tensor_copy_async^⚠
ggml_backend_tensor_copy_n^⚠
ggml_backend_tensor_get^⚠
ggml_backend_tensor_get_2d^⚠
ggml_backend_tensor_get_2d_async^⚠
ggml_backend_tensor_get_async^⚠
ggml_backend_tensor_get_n^⚠
ggml_backend_tensor_memset^⚠
ggml_backend_tensor_set^⚠
ggml_backend_tensor_set_2d^⚠
ggml_backend_tensor_set_2d_async^⚠
ggml_backend_tensor_set_async^⚠
ggml_backend_tensor_set_n^⚠
ggml_backend_unload^⚠
ggml_backend_view_init^⚠
ggml_bf16_to_fp32^⚠
ggml_bf16_to_fp32_row^⚠
ggml_blck_size^⚠
ggml_build_backward_expand^⚠
ggml_build_forward_expand^⚠
ggml_build_forward_order^⚠
ggml_build_forward_select^⚠
ggml_can_repeat^⚠
ggml_cast^⚠
ggml_ceil^⚠
ggml_ceil_inplace^⚠
ggml_clamp^⚠
ggml_clamp_inplace^⚠
ggml_col2im_1d^⚠
ggml_commit^⚠
ggml_concat^⚠
ggml_cont^⚠
ggml_cont_1d^⚠
ggml_cont_2d^⚠
ggml_cont_3d^⚠
ggml_cont_4d^⚠
ggml_conv_1d^⚠
ggml_conv_1d_dw^⚠
ggml_conv_1d_dw_ph^⚠
ggml_conv_1d_ph^⚠
ggml_conv_2d^⚠
ggml_conv_2d_direct^⚠
ggml_conv_2d_dw^⚠
ggml_conv_2d_dw_direct^⚠
ggml_conv_2d_s1_ph^⚠
ggml_conv_2d_sk_p0^⚠
ggml_conv_3d^⚠
ggml_conv_3d_direct^⚠
ggml_conv_transpose_1d^⚠
ggml_conv_transpose_2d_p0^⚠
ggml_cos^⚠
ggml_cos_inplace^⚠
ggml_count_equal^⚠
ggml_cpu_bf16_to_fp32^⚠
ggml_cpu_fp16_to_fp32^⚠
ggml_cpu_fp32_to_bf16^⚠
ggml_cpu_fp32_to_fp16^⚠
ggml_cpu_fp32_to_fp32^⚠
ggml_cpu_fp32_to_i32^⚠
ggml_cpu_get_rvv_vlen^⚠
ggml_cpu_get_sve_cnt^⚠
ggml_cpu_has_amx_int8^⚠
ggml_cpu_has_arm_fma^⚠
ggml_cpu_has_avx^⚠
ggml_cpu_has_avx2^⚠
ggml_cpu_has_avx512^⚠
ggml_cpu_has_avx512_bf16^⚠
ggml_cpu_has_avx512_vbmi^⚠
ggml_cpu_has_avx512_vnni^⚠
ggml_cpu_has_avx_vnni^⚠
ggml_cpu_has_bmi2^⚠
ggml_cpu_has_dotprod^⚠
ggml_cpu_has_f16c^⚠
ggml_cpu_has_fma^⚠
ggml_cpu_has_fp16_va^⚠
ggml_cpu_has_llamafile^⚠
ggml_cpu_has_matmul_int8^⚠
ggml_cpu_has_neon^⚠
ggml_cpu_has_riscv_v^⚠
ggml_cpu_has_sme^⚠
ggml_cpu_has_sme2^⚠
ggml_cpu_has_sse3^⚠
ggml_cpu_has_ssse3^⚠
ggml_cpu_has_sve^⚠
ggml_cpu_has_vsx^⚠
ggml_cpu_has_vxe^⚠
ggml_cpu_has_wasm_simd^⚠
ggml_cpu_init^⚠
ggml_cpy^⚠
ggml_cross_entropy_loss^⚠
ggml_cross_entropy_loss_back^⚠
ggml_cumsum^⚠
ggml_custom_4d^⚠
ggml_custom_inplace^⚠
ggml_cycles^⚠
ggml_cycles_per_ms^⚠
ggml_diag^⚠
ggml_diag_mask_inf^⚠
ggml_diag_mask_inf_inplace^⚠
ggml_diag_mask_zero^⚠
ggml_diag_mask_zero_inplace^⚠
ggml_div^⚠
ggml_div_inplace^⚠
ggml_dsv4_hc_comb^⚠
ggml_dsv4_hc_post^⚠
ggml_dsv4_hc_pre^⚠
ggml_dsv4_hc_pre_gated^⚠
ggml_dup^⚠
ggml_dup_inplace^⚠
ggml_dup_tensor^⚠
ggml_element_size^⚠
ggml_elu^⚠
ggml_elu_inplace^⚠
ggml_exp^⚠
ggml_exp_inplace^⚠
ggml_expm1^⚠
ggml_expm1_inplace^⚠
ggml_fill^⚠
ggml_fill_inplace^⚠
ggml_flash_attn_back^⚠
ggml_flash_attn_ext^⚠
ggml_flash_attn_ext_add_sinks^⚠
ggml_flash_attn_ext_get_prec^⚠
ggml_flash_attn_ext_set_n_kv_max^⚠
ggml_flash_attn_ext_set_prec^⚠
ggml_floor^⚠
ggml_floor_inplace^⚠
ggml_fopen^⚠
ggml_format_name^⚠
ggml_fp16_to_fp32^⚠
ggml_fp16_to_fp32_row^⚠
ggml_fp32_to_bf16^⚠
ggml_fp32_to_bf16_row^⚠
ggml_fp32_to_bf16_row_ref^⚠
ggml_fp32_to_fp16^⚠
ggml_fp32_to_fp16_row^⚠
ggml_free^⚠
ggml_ftype_to_ggml_type^⚠
ggml_gallocr_alloc_graph^⚠
ggml_gallocr_free^⚠
ggml_gallocr_get_buffer_size^⚠
ggml_gallocr_new^⚠
ggml_gallocr_new_n^⚠
ggml_gallocr_reserve^⚠
ggml_gallocr_reserve_n^⚠
ggml_gallocr_reserve_n_size^⚠
ggml_gated_delta_net^⚠
ggml_gated_linear_attn^⚠
ggml_geglu^⚠
ggml_geglu_erf^⚠
ggml_geglu_erf_split^⚠
ggml_geglu_erf_swapped^⚠
ggml_geglu_quick^⚠
ggml_geglu_quick_split^⚠
ggml_geglu_quick_swapped^⚠
ggml_geglu_split^⚠
ggml_geglu_swapped^⚠
ggml_gelu^⚠
ggml_gelu_erf^⚠
ggml_gelu_erf_inplace^⚠
ggml_gelu_inplace^⚠
ggml_gelu_quick^⚠
ggml_gelu_quick_inplace^⚠
ggml_get_data^⚠
ggml_get_data_f32^⚠
ggml_get_f32_1d^⚠
ggml_get_f32_nd^⚠
ggml_get_first_tensor^⚠
ggml_get_glu_op^⚠
ggml_get_i32_1d^⚠
ggml_get_i32_nd^⚠
ggml_get_max_tensor_size^⚠
ggml_get_mem_buffer^⚠
ggml_get_mem_size^⚠
ggml_get_name^⚠
ggml_get_next_tensor^⚠
ggml_get_no_alloc^⚠
ggml_get_rel_pos^⚠
ggml_get_rows^⚠
ggml_get_rows_back^⚠
ggml_get_tensor^⚠
ggml_get_type_traits^⚠
ggml_get_type_traits_cpu^⚠
ggml_get_unary_op^⚠
ggml_glu^⚠
ggml_glu_op_name^⚠
ggml_glu_split^⚠
ggml_graph_add_node^⚠
ggml_graph_clear^⚠
ggml_graph_compute^⚠
ggml_graph_compute_with_ctx^⚠
ggml_graph_cpy^⚠
ggml_graph_dump_dot^⚠
ggml_graph_dup^⚠
ggml_graph_get_grad^⚠
ggml_graph_get_grad_acc^⚠
ggml_graph_get_tensor^⚠
ggml_graph_n_nodes^⚠
ggml_graph_node^⚠
ggml_graph_nodes^⚠
ggml_graph_overhead^⚠
ggml_graph_overhead_custom^⚠
ggml_graph_plan^⚠
ggml_graph_print^⚠
ggml_graph_reset^⚠
ggml_graph_size^⚠
ggml_group_norm^⚠
ggml_group_norm_inplace^⚠
ggml_guid_matches^⚠
ggml_hardsigmoid^⚠
ggml_hardswish^⚠
ggml_im2col^⚠
ggml_im2col_3d^⚠
ggml_im2col_back^⚠
ggml_init^⚠
ggml_interpolate^⚠
ggml_is_3d^⚠
ggml_is_contiguous^⚠
ggml_is_contiguous_0^⚠
ggml_is_contiguous_1^⚠
ggml_is_contiguous_2^⚠
ggml_is_contiguous_channels^⚠
ggml_is_contiguous_rows^⚠
ggml_is_contiguous_to_1^⚠
ggml_is_contiguous_to_2^⚠
ggml_is_contiguous_to_3^⚠
ggml_is_contiguously_allocated^⚠
ggml_is_empty^⚠
ggml_is_matrix^⚠
ggml_is_numa^⚠
ggml_is_permuted^⚠
ggml_is_quantized^⚠
ggml_is_scalar^⚠
ggml_is_transposed^⚠
ggml_is_vector^⚠
ggml_is_view^⚠
ggml_l2_norm^⚠
ggml_l2_norm_inplace^⚠
ggml_leaky_relu^⚠
ggml_lightning_indexer^⚠
ggml_log^⚠
ggml_log_get^⚠
ggml_log_inplace^⚠
ggml_log_set^⚠
ggml_map_custom1^⚠
ggml_map_custom2^⚠
ggml_map_custom3^⚠
ggml_map_custom1_inplace^⚠
ggml_map_custom2_inplace^⚠
ggml_map_custom3_inplace^⚠
ggml_mean^⚠
ggml_mul^⚠
ggml_mul_inplace^⚠
ggml_mul_mat^⚠
ggml_mul_mat_id^⚠
ggml_mul_mat_set_hint^⚠
ggml_mul_mat_set_prec^⚠
ggml_n_dims^⚠
ggml_nbytes^⚠
ggml_nbytes_pad^⚠
ggml_neg^⚠
ggml_neg_inplace^⚠
ggml_nelements^⚠
ggml_new_buffer^⚠
ggml_new_f32^⚠
ggml_new_graph^⚠
ggml_new_graph_custom^⚠
ggml_new_i32^⚠
ggml_new_tensor^⚠
ggml_new_tensor_1d^⚠
ggml_new_tensor_2d^⚠
ggml_new_tensor_3d^⚠
ggml_new_tensor_4d^⚠
ggml_norm^⚠
ggml_norm_inplace^⚠
ggml_nrows^⚠
ggml_numa_init^⚠
ggml_op_desc^⚠
ggml_op_name^⚠
ggml_op_symbol^⚠
ggml_opt_alloc^⚠
ggml_opt_context_optimizer_type^⚠
ggml_opt_dataset_data^⚠
ggml_opt_dataset_free^⚠
ggml_opt_dataset_get_batch^⚠
ggml_opt_dataset_get_batch_host^⚠
ggml_opt_dataset_init^⚠
ggml_opt_dataset_labels^⚠
ggml_opt_dataset_ndata^⚠
ggml_opt_dataset_shuffle^⚠
ggml_opt_default_params^⚠
ggml_opt_epoch^⚠
ggml_opt_epoch_callback_progress_bar^⚠
ggml_opt_eval^⚠
ggml_opt_fit^⚠
ggml_opt_free^⚠
ggml_opt_get_constant_optimizer_params^⚠
ggml_opt_get_default_optimizer_params^⚠
ggml_opt_grad_acc^⚠
ggml_opt_init^⚠
ggml_opt_inputs^⚠
ggml_opt_labels^⚠
ggml_opt_loss^⚠
ggml_opt_ncorrect^⚠
ggml_opt_optimizer_name^⚠
ggml_opt_outputs^⚠
ggml_opt_pred^⚠
ggml_opt_prepare_alloc^⚠
ggml_opt_reset^⚠
ggml_opt_result_accuracy^⚠
ggml_opt_result_free^⚠
ggml_opt_result_init^⚠
ggml_opt_result_loss^⚠
ggml_opt_result_ndata^⚠
ggml_opt_result_pred^⚠
ggml_opt_result_reset^⚠
ggml_opt_static_graphs^⚠
ggml_opt_step_adamw^⚠
ggml_opt_step_sgd^⚠
ggml_out_prod^⚠
ggml_pad^⚠
ggml_pad_circular^⚠
ggml_pad_ext^⚠
ggml_pad_ext_circular^⚠
ggml_pad_reflect_1d^⚠
ggml_permute^⚠
ggml_pool_1d^⚠
ggml_pool_2d^⚠
ggml_pool_2d_back^⚠
ggml_prec_get_acc^⚠
ggml_prec_get_src^⚠
ggml_prec_set_acc^⚠
ggml_prec_set_src^⚠
ggml_print_object^⚠
ggml_print_objects^⚠
ggml_quantize_chunk^⚠
ggml_quantize_free^⚠
ggml_quantize_init^⚠
ggml_quantize_requires_imatrix^⚠
ggml_reglu^⚠
ggml_reglu_split^⚠
ggml_reglu_swapped^⚠
ggml_relu^⚠
ggml_relu_inplace^⚠
ggml_repeat^⚠
ggml_repeat_4d^⚠
ggml_repeat_back^⚠
ggml_reset^⚠
ggml_reshape^⚠
ggml_reshape_1d^⚠
ggml_reshape_2d^⚠
ggml_reshape_3d^⚠
ggml_reshape_4d^⚠
ggml_rms_norm^⚠
ggml_rms_norm_back^⚠
ggml_rms_norm_inplace^⚠
ggml_roll^⚠
ggml_rope^⚠
ggml_rope_custom^⚠
ggml_rope_custom_inplace^⚠
ggml_rope_ext^⚠
ggml_rope_ext_back^⚠
ggml_rope_ext_inplace^⚠
ggml_rope_inplace^⚠
ggml_rope_multi^⚠
ggml_rope_multi_back^⚠
ggml_rope_multi_inplace^⚠
ggml_rope_set_offset^⚠
ggml_rope_yarn_corr_dims^⚠
ggml_round^⚠
ggml_round_inplace^⚠
ggml_row_size^⚠
ggml_rwkv_wkv6^⚠
ggml_rwkv_wkv7^⚠
ggml_scale^⚠
ggml_scale_bias^⚠
ggml_scale_bias_inplace^⚠
ggml_scale_inplace^⚠
ggml_set^⚠
ggml_set_1d^⚠
ggml_set_1d_inplace^⚠
ggml_set_2d^⚠
ggml_set_2d_inplace^⚠
ggml_set_abort_callback^⚠
ggml_set_f32^⚠
ggml_set_f32_1d^⚠
ggml_set_f32_nd^⚠
ggml_set_i32^⚠
ggml_set_i32_1d^⚠
ggml_set_i32_nd^⚠
ggml_set_inplace^⚠
ggml_set_input^⚠
ggml_set_loss^⚠
ggml_set_name^⚠
ggml_set_no_alloc^⚠
ggml_set_output^⚠
ggml_set_param^⚠
ggml_set_rows^⚠
ggml_set_zero^⚠
ggml_sgn^⚠
ggml_sgn_inplace^⚠
ggml_sigmoid^⚠
ggml_sigmoid_inplace^⚠
ggml_silu^⚠
ggml_silu_back^⚠
ggml_silu_inplace^⚠
ggml_sin^⚠
ggml_sin_inplace^⚠
ggml_soft_max^⚠
ggml_soft_max_add_sinks^⚠
ggml_soft_max_ext^⚠
ggml_soft_max_ext_back^⚠
ggml_soft_max_ext_back_inplace^⚠
ggml_soft_max_ext_inplace^⚠
ggml_soft_max_inplace^⚠
ggml_softplus^⚠
ggml_softplus_inplace^⚠
ggml_solve_tri^⚠
ggml_sqr^⚠
ggml_sqr_inplace^⚠
ggml_sqrt^⚠
ggml_sqrt_inplace^⚠
ggml_ssm_conv^⚠
ggml_ssm_scan^⚠
ggml_status_to_string^⚠
ggml_step^⚠
ggml_step_inplace^⚠
ggml_sub^⚠
ggml_sub_inplace^⚠
ggml_sum^⚠
ggml_sum_rows^⚠
ggml_swiglu^⚠
ggml_swiglu_clamp^⚠
ggml_swiglu_oai^⚠
ggml_swiglu_split^⚠
ggml_swiglu_swapped^⚠
ggml_tallocr_alloc^⚠
ggml_tallocr_new^⚠
ggml_tanh^⚠
ggml_tanh_inplace^⚠
ggml_tensor_overhead^⚠
ggml_threadpool_free^⚠
ggml_threadpool_get_n_threads^⚠
ggml_threadpool_new^⚠
ggml_threadpool_params_default^⚠
ggml_threadpool_params_from_cpu_params^⚠
ggml_threadpool_params_init^⚠
ggml_threadpool_params_match^⚠
ggml_threadpool_pause^⚠
ggml_threadpool_resume^⚠
ggml_time_init^⚠
ggml_time_ms^⚠
ggml_time_us^⚠
ggml_timestep_embedding^⚠
ggml_top_k^⚠
ggml_top_k_mask^⚠
ggml_transpose^⚠
ggml_tri^⚠
ggml_trunc^⚠: Truncates the fractional part of each element in the tensor (towards zero). For example: trunc(3.7) = 3.0, trunc(-2.9) = -2.0 Similar to std::trunc in C/C++.
ggml_trunc_inplace^⚠
ggml_type_name^⚠
ggml_type_size^⚠
ggml_type_sizef^⚠
ggml_unary^⚠
ggml_unary_inplace^⚠
ggml_unary_op_name^⚠
ggml_unravel_index^⚠
ggml_upscale^⚠
ggml_upscale_ext^⚠
ggml_used_mem^⚠
ggml_validate_row_data^⚠
ggml_version^⚠
ggml_view_1d^⚠
ggml_view_2d^⚠
ggml_view_3d^⚠
ggml_view_4d^⚠
ggml_view_tensor^⚠
ggml_win_part^⚠
ggml_win_unpart^⚠
ggml_xielu^⚠
llama_adapter_get_alora_invocation_tokens^⚠
llama_adapter_get_alora_n_invocation_tokens^⚠
llama_adapter_lora_free^⚠
llama_adapter_lora_init^⚠
llama_adapter_lora_init_from_file_ptr^⚠
llama_adapter_meta_count^⚠
llama_adapter_meta_key_by_index^⚠
llama_adapter_meta_val_str^⚠
llama_adapter_meta_val_str_by_index^⚠
llama_add_bos_token^⚠
llama_add_eos_token^⚠
llama_arch_name^⚠
llama_arch_supports_mixed_batch^⚠
llama_arch_supports_save^⚠
llama_arch_supports_sm_tensor^⚠
llama_attach_threadpool^⚠
llama_backend_free^⚠
llama_backend_init^⚠
llama_batch_ext_add^⚠
llama_batch_ext_add_embd^⚠
llama_batch_ext_add_seq^⚠
llama_batch_ext_add_token^⚠
llama_batch_ext_clear^⚠
llama_batch_ext_free^⚠
llama_batch_ext_init^⚠
llama_batch_ext_set_decision_order^⚠
llama_batch_ext_set_embd_state^⚠
llama_batch_ext_set_embd_token^⚠
llama_batch_ext_set_output_embd^⚠
llama_batch_ext_set_output_logits^⚠
llama_batch_ext_set_pos^⚠
llama_batch_free^⚠
llama_batch_get_one^⚠
llama_batch_init^⚠
llama_chat_apply_template^⚠: Apply chat template. Inspired by hf apply_chat_template() on python.
llama_chat_builtin_templates^⚠
llama_context_default_params^⚠
llama_copy_state_data^⚠
llama_cpp_rs_decode_hooks_abi_v1^⚠
llama_decode^⚠
llama_detach_threadpool^⚠
llama_detokenize^⚠: @details Convert the provided tokens into text (inverse of llama_tokenize()). @param text The char pointer must be large enough to hold the resulting text. @return Returns the number of chars/bytes on success, no more than text_len_max. @return Returns a negative number on failure - the number of chars/bytes that would have been returned. @param remove_special Allow to remove BOS and EOS tokens if model is configured to do so. @param unparse_special If true, special tokens are rendered in the output.
llama_encode^⚠
llama_flash_attn_type_name^⚠
llama_free^⚠
llama_free_model^⚠
llama_ftype_get_default_type^⚠
llama_ftype_name^⚠
llama_get_causal_attn^⚠
llama_get_ctx_other^⚠
llama_get_embeddings^⚠
llama_get_embeddings_ith^⚠
llama_get_embeddings_layer_inp^⚠
llama_get_embeddings_nextn^⚠
llama_get_embeddings_nextn_ith^⚠
llama_get_embeddings_seq^⚠
llama_get_logits^⚠
llama_get_logits_ith^⚠
llama_get_memory^⚠
llama_get_memory_breakdown^⚠
llama_get_model^⚠
llama_get_sampled_candidates_count_ith^⚠
llama_get_sampled_candidates_ith^⚠
llama_get_sampled_logits_count_ith^⚠
llama_get_sampled_logits_ith^⚠
llama_get_sampled_probs_count_ith^⚠
llama_get_sampled_probs_ith^⚠
llama_get_sampled_token_ith^⚠
llama_get_state_size^⚠
llama_graph_reserve^⚠
llama_init_from_model^⚠
llama_load_mode_from_str^⚠
llama_load_mode_name^⚠
llama_load_model_from_file^⚠
llama_load_session_file^⚠
llama_log_get^⚠
llama_log_set^⚠
llama_max_devices^⚠
llama_max_parallel_sequences^⚠
llama_max_tensor_buft_overrides^⚠
llama_memory_breakdown_collect^⚠
llama_memory_buffers_collect^⚠
llama_memory_can_shift^⚠
llama_memory_clear^⚠
llama_memory_seq_add^⚠
llama_memory_seq_cp^⚠
llama_memory_seq_div^⚠
llama_memory_seq_keep^⚠
llama_memory_seq_pos_max^⚠
llama_memory_seq_pos_min^⚠
llama_memory_seq_rm^⚠
llama_model_chat_template^⚠
llama_model_cls_label^⚠
llama_model_decision_label^⚠
llama_model_decoder_start_token^⚠
llama_model_default_params^⚠
llama_model_desc^⚠
llama_model_dflash_selector_top_k^⚠
llama_model_free^⚠
llama_model_ftype^⚠
llama_model_get_device^⚠
llama_model_get_tok_embd^⚠
llama_model_get_tok_embd_rows^⚠
llama_model_get_vocab^⚠
llama_model_has_decoder^⚠
llama_model_has_encoder^⚠
llama_model_init_from_user^⚠
llama_model_is_diffusion^⚠
llama_model_is_hybrid^⚠
llama_model_is_recurrent^⚠
llama_model_load_from_file^⚠
llama_model_load_from_file_ptr^⚠
llama_model_load_from_splits^⚠
llama_model_meta_count^⚠
llama_model_meta_key_by_index^⚠
llama_model_meta_key_str^⚠
llama_model_meta_val_str^⚠
llama_model_meta_val_str_by_index^⚠
llama_model_n_cls_out^⚠
llama_model_n_ctx_train^⚠
llama_model_n_devices^⚠
llama_model_n_embd^⚠
llama_model_n_embd_inp^⚠
llama_model_n_embd_out^⚠
llama_model_n_expert^⚠
llama_model_n_head^⚠
llama_model_n_head_kv^⚠
llama_model_n_layer^⚠
llama_model_n_layer_nextn^⚠
llama_model_n_params^⚠
llama_model_n_swa^⚠
llama_model_n_tokens_delay^⚠
llama_model_quantize^⚠
llama_model_quantize_default_params^⚠
llama_model_rope_freq_scale_train^⚠
llama_model_rope_type^⚠
llama_model_save_to_file^⚠
llama_model_set_attn_rot^⚠
llama_model_size^⚠
llama_model_target_layer_ids^⚠
llama_model_target_layer_ids_n^⚠
llama_n_batch^⚠
llama_n_ctx^⚠
llama_n_ctx_seq^⚠
llama_n_ctx_train^⚠
llama_n_embd^⚠
llama_n_head^⚠
llama_n_layer^⚠
llama_n_rs_seq^⚠
llama_n_seq_max^⚠
llama_n_threads^⚠
llama_n_threads_batch^⚠
llama_n_ubatch^⚠
llama_n_vocab^⚠
llama_new_context_with_model^⚠
llama_numa_init^⚠
llama_opt_epoch^⚠
llama_opt_init^⚠
llama_opt_param_filter_all^⚠
llama_perf_context^⚠
llama_perf_context_print^⚠
llama_perf_context_reset^⚠
llama_perf_sampler^⚠
llama_perf_sampler_print^⚠
llama_perf_sampler_reset^⚠
llama_pooling_type^⚠
llama_print_system_info^⚠
llama_process^⚠
llama_quant_compute_types^⚠
llama_quant_compute_types_guarded^⚠
llama_quant_free^⚠
llama_quant_init^⚠
llama_quant_init_guarded^⚠
llama_quant_model_from_metadata^⚠
llama_quant_model_from_metadata_guarded^⚠
llama_quant_tensor_allows_quantization^⚠
llama_quant_tensor_allows_quantization_guarded^⚠
llama_sampler_accept^⚠
llama_sampler_apply^⚠
llama_sampler_chain_add^⚠
llama_sampler_chain_default_params^⚠
llama_sampler_chain_get^⚠
llama_sampler_chain_init^⚠
llama_sampler_chain_n^⚠
llama_sampler_chain_remove^⚠
llama_sampler_clone^⚠
llama_sampler_copy^⚠
llama_sampler_free^⚠
llama_sampler_get_seed^⚠
llama_sampler_init^⚠
llama_sampler_init_adaptive_p^⚠: adaptive-p: select tokens near a configurable target probability over time.
llama_sampler_init_dist^⚠: seed == LLAMA_DEFAULT_SEED to use a random seed.
llama_sampler_init_dry^⚠: @details DRY sampler, designed by p-e-w, as described in: https://github.com/oobabooga/text-generation-webui/pull/5677, porting Koboldcpp implementation authored by pi6am: https://github.com/LostRuins/koboldcpp/pull/982
llama_sampler_init_grammar^⚠: @details Initializes a GBNF grammar, see grammars/README.md for details. @param vocab The vocabulary that this grammar will be used with. @param grammar_str The production rules for the grammar, encoded as a string. Returns an empty grammar if empty. Returns NULL if parsing of grammar_str fails. @param grammar_root The name of the start symbol for the grammar.
llama_sampler_init_grammar_lazy^⚠
llama_sampler_init_grammar_lazy_patterns^⚠: @details Lazy grammar sampler, introduced in https://github.com/ggml-org/llama.cpp/pull/9639 @param trigger_patterns A list of patterns that will trigger the grammar sampler. Pattern will be matched from the start of the generation output, and grammar sampler will be fed content starting from its first match group. @param trigger_tokens A list of tokens that will trigger the grammar sampler. Grammar sampler will be fed content starting from the trigger token included.
llama_sampler_init_greedy^⚠
llama_sampler_init_infill^⚠
llama_sampler_init_logit_bias^⚠
llama_sampler_init_min_p^⚠: @details Minimum P sampling as described in https://github.com/ggml-org/llama.cpp/pull/3841
llama_sampler_init_mirostat^⚠: @details Mirostat 1.0 algorithm described in the paper https://arxiv.org/abs/2007.14966. Uses tokens instead of words. @param candidates A vector of llama_token_data containing the candidate tokens, their probabilities (p), and log-odds (logit) for the current position in the generated text. @param tau The target cross-entropy (or surprise) value you want to achieve for the generated text. A higher value corresponds to more surprising or less predictable text, while a lower value corresponds to less surprising or more predictable text. @param eta The learning rate used to update mu based on the error between the target and observed surprisal of the sampled word. A larger learning rate will cause mu to be updated more quickly, while a smaller learning rate will result in slower updates. @param m The number of tokens considered in the estimation of s_hat. This is an arbitrary value that is used to calculate s_hat, which in turn helps to calculate the value of k. In the paper, they use m = 100, but you can experiment with different values to see how it affects the performance of the algorithm. @param mu Maximum cross-entropy. This value is initialized to be twice the target cross-entropy (2 * tau) and is updated in the algorithm based on the error between the target and observed surprisal.
llama_sampler_init_mirostat_v2^⚠: @details Mirostat 2.0 algorithm described in the paper https://arxiv.org/abs/2007.14966. Uses tokens instead of words. @param candidates A vector of llama_token_data containing the candidate tokens, their probabilities (p), and log-odds (logit) for the current position in the generated text. @param tau The target cross-entropy (or surprise) value you want to achieve for the generated text. A higher value corresponds to more surprising or less predictable text, while a lower value corresponds to less surprising or more predictable text. @param eta The learning rate used to update mu based on the error between the target and observed surprisal of the sampled word. A larger learning rate will cause mu to be updated more quickly, while a smaller learning rate will result in slower updates. @param mu Maximum cross-entropy. This value is initialized to be twice the target cross-entropy (2 * tau) and is updated in the algorithm based on the error between the target and observed surprisal.
llama_sampler_init_penalties^⚠: NOTE: Avoid using on the full vocabulary as searching for repeated tokens can become slow. For example, apply top-k or top-p sampling first.
llama_sampler_init_temp^⚠: #details Updates the logits l_i` = l_i/t. When t <= 0.0f, the maximum logit is kept at it’s original value, the rest are set to -inf
llama_sampler_init_temp_ext^⚠: @details Dynamic temperature implementation (a.k.a. entropy) described in the paper https://arxiv.org/abs/2309.02772.
llama_sampler_init_top_k^⚠: @details Top-K sampling described in academic paper “The Curious Case of Neural Text Degeneration” https://arxiv.org/abs/1904.09751 Setting k <= 0 makes this a noop
llama_sampler_init_top_n_sigma^⚠: @details Top n sigma sampling as described in academic paper “Top-nσ: Not All Logits Are You Need” https://arxiv.org/pdf/2411.07641
llama_sampler_init_top_p^⚠: @details Nucleus sampling described in academic paper “The Curious Case of Neural Text Degeneration” https://arxiv.org/abs/1904.09751
llama_sampler_init_typical^⚠: @details Locally Typical Sampling implementation described in the paper https://arxiv.org/abs/2202.00666.
llama_sampler_init_xtc^⚠: @details XTC sampler as described in https://github.com/oobabooga/text-generation-webui/pull/6335
llama_sampler_name^⚠
llama_sampler_reset^⚠
llama_sampler_sample^⚠
llama_save_session_file^⚠
llama_set_abort_callback^⚠
llama_set_adapter_cvec^⚠
llama_set_adapters_lora^⚠
llama_set_causal_attn^⚠
llama_set_embeddings^⚠
llama_set_embeddings_layer_inp^⚠
llama_set_embeddings_nextn^⚠
llama_set_n_threads^⚠
llama_set_nextn_layer_offset^⚠
llama_set_sampler^⚠
llama_set_state_data^⚠
llama_set_warmup^⚠
llama_shim_last_error^⚠
llama_split_path^⚠: @details Build a split GGUF final path for this chunk. llama_split_path(split_path, sizeof(split_path), “/models/ggml-model-q4_0”, 2, 4) => split_path = “/models/ggml-model-q4_0-00002-of-00004.gguf”
llama_split_prefix^⚠: @details Extract the path prefix from the split_path if and only if the split_no and split_count match. llama_split_prefix(split_prefix, 64, “/models/ggml-model-q4_0-00002-of-00004.gguf”, 2, 4) => split_prefix = “/models/ggml-model-q4_0”
llama_state_get_data^⚠
llama_state_get_size^⚠
llama_state_load_file^⚠
llama_state_save_file^⚠
llama_state_seq_get_data^⚠
llama_state_seq_get_data_ext^⚠
llama_state_seq_get_size^⚠
llama_state_seq_get_size_ext^⚠
llama_state_seq_load_file^⚠
llama_state_seq_save_file^⚠
llama_state_seq_set_data^⚠
llama_state_seq_set_data_ext^⚠
llama_state_set_data^⚠
llama_supports_gpu_offload^⚠
llama_supports_mlock^⚠
llama_supports_mmap^⚠
llama_supports_rpc^⚠
llama_synchronize^⚠
llama_time_us^⚠
llama_token_bos^⚠
llama_token_cls^⚠
llama_token_eos^⚠
llama_token_eot^⚠
llama_token_fim_mid^⚠
llama_token_fim_pad^⚠
llama_token_fim_pre^⚠
llama_token_fim_rep^⚠
llama_token_fim_sep^⚠
llama_token_fim_suf^⚠
llama_token_get_attr^⚠
llama_token_get_score^⚠
llama_token_get_text^⚠
llama_token_is_control^⚠
llama_token_is_eog^⚠
llama_token_nl^⚠
llama_token_pad^⚠
llama_token_sep^⚠
llama_token_to_piece^⚠
llama_tokenize^⚠: @details Convert the provided text into tokens. @param tokens The tokens pointer must be large enough to hold the resulting tokens. @return Returns the number of tokens on success, no more than n_tokens_max @return Returns a negative number on failure - the number of tokens that would have been returned @return Returns INT32_MIN on overflow (e.g., tokenization result size exceeds int32_t limit) @param add_special Allow to add BOS and EOS tokens if model is configured to do so. @param parse_special Allow tokenizing special and/or control tokens which otherwise are not exposed and treated as plaintext. Does not insert a leading space.
llama_version^⚠
llama_vocab_bos^⚠
llama_vocab_cls^⚠
llama_vocab_eos^⚠
llama_vocab_eot^⚠
llama_vocab_fim_mid^⚠
llama_vocab_fim_pad^⚠
llama_vocab_fim_pre^⚠
llama_vocab_fim_rep^⚠
llama_vocab_fim_sep^⚠
llama_vocab_fim_suf^⚠
llama_vocab_get_add_bos^⚠
llama_vocab_get_add_eos^⚠
llama_vocab_get_add_sep^⚠
llama_vocab_get_attr^⚠
llama_vocab_get_score^⚠
llama_vocab_get_suppress_tokens^⚠
llama_vocab_get_text^⚠
llama_vocab_is_control^⚠
llama_vocab_is_eog^⚠
llama_vocab_mask^⚠
llama_vocab_n_tokens^⚠
llama_vocab_nl^⚠
llama_vocab_pad^⚠
llama_vocab_sep^⚠
llama_vocab_type^⚠
mtp_session_accept^⚠
mtp_session_begin^⚠
mtp_session_draft^⚠
mtp_session_free^⚠
mtp_session_has_pending_proposal^⚠
mtp_session_is_quiescent^⚠
mtp_session_n_max^⚠
mtp_session_need_embd^⚠
mtp_session_need_embd_pre_norm^⚠
mtp_session_new^⚠
mtp_session_print_stats^⚠
mtp_session_process^⚠
mtp_session_state_get^⚠
mtp_session_state_set^⚠
mtp_session_state_size^⚠

Type Aliases§

Unions§

llama_model_kv_override__bindgen_ty_1
std_vector__Temporary_value__Storage