matt/ollama37
1
0
Fork 0
mirror of https://github.com/dogkeeper886/ollama37.git synced 2025-12-18 03:37:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

Sync with upstream ollama/ollama and restore Tesla K80 (compute 3.7) support

Browse Source 
This commit represents a complete rework after pulling the latest changes from
official ollama/ollama repository and re-applying Tesla K80 compatibility patches.

## Key Changes

### CUDA Compute Capability 3.7 Support (Tesla K80)
- Added sm_37 (compute 3.7) to CMAKE_CUDA_ARCHITECTURES in CMakeLists.txt
- Updated CMakePresets.json to include compute 3.7 in "CUDA 11" preset
- Using 37-virtual (PTX with JIT compilation) for maximum compatibility

### Legacy Toolchain Compatibility
- **NVIDIA Driver**: 470.256.02 (last version supporting Kepler/K80)
- **CUDA Version**: 11.4.4 (last CUDA 11.x supporting compute 3.7)
- **GCC Version**: 10.5.0 (required by CUDA 11.4 host_config.h)

### CPU Architecture Trade-offs
Due to GCC 10.5 limitation, sacrificed newer CPU optimizations:
- Alderlake CPU variant enabled WITHOUT AVX_VNNI (requires GCC 11+)
- Still supports: SSE4.2, AVX, F16C, AVX2, BMI2, FMA
- Performance impact: ~3-7% on newer CPUs (acceptable for K80 compatibility)

### Build System Updates
- Modified ml/backend/ggml/ggml/src/ggml-cuda/CMakeLists.txt for compute 3.7
- Added -Wno-deprecated-gpu-targets flag to suppress warnings
- Updated ml/backend/ggml/ggml/src/CMakeLists.txt for Alderlake without AVX_VNNI

### Upstream Sync
Merged latest llama.cpp changes including:
- Enhanced KV cache management with ISWA and hybrid memory support
- Improved multi-modal support (mtmd framework)
- New model architectures (Gemma3, Llama4, Qwen3, etc.)
- GPU backend improvements for CUDA, Metal, and ROCm
- Updated quantization support and GGUF format handling

### Documentation
- Updated CLAUDE.md with comprehensive build instructions
- Documented toolchain constraints and CPU architecture trade-offs
- Removed outdated CI/CD workflows (tesla-k80-*.yml)
- Cleaned up temporary development artifacts

## Rationale

This fork maintains Tesla K80 GPU support (compute 3.7) which was dropped in
official Ollama due to legacy driver/CUDA requirements. The toolchain constraint
creates a deadlock:
- K80 → Driver 470 → CUDA 11.4 → GCC 10 → No AVX_VNNI

We accept the loss of cutting-edge CPU optimizations to enable running modern
LLMs on legacy but still capable Tesla K80 hardware (12GB VRAM per GPU).

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Shang Chieh Tseng
2025-11-05 14:03:05 +08:00
parent fabe2c5cb7
commit ef14fb5b26
817 changed files with 241634 additions and 70888 deletions
Download Patch File Download Diff File Expand all files Collapse all files

452
llama/llama.cpp/include/llama.h vendored
View File

@@ -61,59 +61,21 @@ extern "C" {
struct llama_model;
struct llama_context;
struct llama_sampler;
struct llama_kv_cache;
typedef struct llama_memory_i * llama_memory_t;
typedef int32_t llama_pos;
typedef int32_t llama_token;
typedef int32_t llama_seq_id;
enum llama_vocab_type {
LLAMA_VOCAB_TYPE_NONE = 0, // For models without vocab
LLAMA_VOCAB_TYPE_SPM = 1, // LLaMA tokenizer based on byte-level BPE with byte fallback
LLAMA_VOCAB_TYPE_BPE = 2, // GPT-2 tokenizer based on byte-level BPE
LLAMA_VOCAB_TYPE_WPM = 3, // BERT tokenizer based on WordPiece
LLAMA_VOCAB_TYPE_UGM = 4, // T5 tokenizer based on Unigram
LLAMA_VOCAB_TYPE_RWKV = 5, // RWKV tokenizer based on greedy tokenization
};
// pre-tokenization types
enum llama_vocab_pre_type {
LLAMA_VOCAB_PRE_TYPE_DEFAULT = 0,
LLAMA_VOCAB_PRE_TYPE_LLAMA3 = 1,
LLAMA_VOCAB_PRE_TYPE_DEEPSEEK_LLM = 2,
LLAMA_VOCAB_PRE_TYPE_DEEPSEEK_CODER = 3,
LLAMA_VOCAB_PRE_TYPE_FALCON = 4,
LLAMA_VOCAB_PRE_TYPE_MPT = 5,
LLAMA_VOCAB_PRE_TYPE_STARCODER = 6,
LLAMA_VOCAB_PRE_TYPE_GPT2 = 7,
LLAMA_VOCAB_PRE_TYPE_REFACT = 8,
LLAMA_VOCAB_PRE_TYPE_COMMAND_R = 9,
LLAMA_VOCAB_PRE_TYPE_STABLELM2 = 10,
LLAMA_VOCAB_PRE_TYPE_QWEN2 = 11,
LLAMA_VOCAB_PRE_TYPE_OLMO = 12,
LLAMA_VOCAB_PRE_TYPE_DBRX = 13,
LLAMA_VOCAB_PRE_TYPE_SMAUG = 14,
LLAMA_VOCAB_PRE_TYPE_PORO = 15,
LLAMA_VOCAB_PRE_TYPE_CHATGLM3 = 16,
LLAMA_VOCAB_PRE_TYPE_CHATGLM4 = 17,
LLAMA_VOCAB_PRE_TYPE_VIKING = 18,
LLAMA_VOCAB_PRE_TYPE_JAIS = 19,
LLAMA_VOCAB_PRE_TYPE_TEKKEN = 20,
LLAMA_VOCAB_PRE_TYPE_SMOLLM = 21,
LLAMA_VOCAB_PRE_TYPE_CODESHELL = 22,
LLAMA_VOCAB_PRE_TYPE_BLOOM = 23,
LLAMA_VOCAB_PRE_TYPE_GPT3_FINNISH = 24,
LLAMA_VOCAB_PRE_TYPE_EXAONE = 25,
LLAMA_VOCAB_PRE_TYPE_CHAMELEON = 26,
LLAMA_VOCAB_PRE_TYPE_MINERVA = 27,
LLAMA_VOCAB_PRE_TYPE_DEEPSEEK3_LLM = 28,
LLAMA_VOCAB_PRE_TYPE_GPT4O = 29,
LLAMA_VOCAB_PRE_TYPE_SUPERBPE = 30,
LLAMA_VOCAB_PRE_TYPE_TRILLION = 31,
LLAMA_VOCAB_PRE_TYPE_BAILINGMOE = 32,
LLAMA_VOCAB_PRE_TYPE_LLAMA4 = 33,
LLAMA_VOCAB_PRE_TYPE_PIXTRAL = 34,
LLAMA_VOCAB_PRE_TYPE_SEED_CODER = 35,
LLAMA_VOCAB_TYPE_NONE = 0, // For models without vocab
LLAMA_VOCAB_TYPE_SPM = 1, // LLaMA tokenizer based on byte-level BPE with byte fallback
LLAMA_VOCAB_TYPE_BPE = 2, // GPT-2 tokenizer based on byte-level BPE
LLAMA_VOCAB_TYPE_WPM = 3, // BERT tokenizer based on WordPiece
LLAMA_VOCAB_TYPE_UGM = 4, // T5 tokenizer based on Unigram
LLAMA_VOCAB_TYPE_RWKV = 5, // RWKV tokenizer based on greedy tokenization
LLAMA_VOCAB_TYPE_PLAMO2 = 6, // PLaMo-2 tokenizer based on Aho-Corasick with dynamic programming
};
enum llama_rope_type {
@@ -188,6 +150,7 @@ extern "C" {
//LLAMA_FTYPE_MOSTLY_Q4_0_8_8 = 35, // removed from gguf files, use Q4_0 and runtime repack
LLAMA_FTYPE_MOSTLY_TQ1_0 = 36, // except 1d tensors
LLAMA_FTYPE_MOSTLY_TQ2_0 = 37, // except 1d tensors
LLAMA_FTYPE_MOSTLY_MXFP4_MOE = 38, // except 1d tensors
LLAMA_FTYPE_GUESSED = 1024, // not specified in the model file
};
@@ -216,6 +179,14 @@ extern "C" {
LLAMA_ATTENTION_TYPE_NON_CAUSAL = 1,
};
enum llama_flash_attn_type {
LLAMA_FLASH_ATTN_TYPE_AUTO = -1,
LLAMA_FLASH_ATTN_TYPE_DISABLED = 0,
LLAMA_FLASH_ATTN_TYPE_ENABLED = 1,
};
LLAMA_API const char * llama_flash_attn_type_name(enum llama_flash_attn_type flash_attn_type);
enum llama_split_mode {
LLAMA_SPLIT_MODE_NONE = 0, // single GPU
LLAMA_SPLIT_MODE_LAYER = 1, // split layers and KV across GPUs
@@ -235,23 +206,26 @@ extern "C" {
llama_token_data * data;
size_t size;
int64_t selected; // this is the index in the data array (i.e. not the token id)
bool sorted;
bool sorted; // note: do not assume the data is sorted - always check this flag
} llama_token_data_array;
typedef bool (*llama_progress_callback)(float progress, void * user_data);
// Input data for llama_decode
// Input data for llama_encode/llama_decode
// A llama_batch object can contain input about one or many sequences
// The provided arrays (i.e. token, embd, pos, etc.) must have size of n_tokens
//
// - token : the token ids of the input (used when embd is NULL)
// - embd : token embeddings (i.e. float vector of size n_embd) (used when token is NULL)
// - pos : the positions of the respective token in the sequence
// (if set to NULL, the token position will be tracked automatically by llama_decode)
// (if set to NULL, the token position will be tracked automatically by llama_encode/llama_decode)
// - seq_id : the sequence to which the respective token belongs
// (if set to NULL, the sequence ID will be assumed to be 0)
// - logits : if zero, the logits (and/or the embeddings) for the respective token will not be output
// (if set to NULL, only the logits for last token will be returned)
// (if set to NULL:
// - if embeddings: all tokens are output
// - if not: only the last token is output
// )
//
typedef struct llama_batch {
int32_t n_tokens;
@@ -261,7 +235,7 @@ extern "C" {
llama_pos * pos;
int32_t * n_seq_id;
llama_seq_id ** seq_id;
int8_t * logits; // TODO: rename this to "output"
int8_t * logits; // TODO: rename this to "output"
} llama_batch;
enum llama_model_kv_override_type {
@@ -317,10 +291,12 @@ extern "C" {
const struct llama_model_kv_override * kv_overrides;
// Keep the booleans together to avoid misalignment during copy-by-value.
bool vocab_only; // only load the vocabulary, no weights
bool use_mmap; // use mmap if possible
bool use_mlock; // force system to keep model in RAM
bool check_tensors; // validate model tensor data
bool vocab_only; // only load the vocabulary, no weights
bool use_mmap; // use mmap if possible
bool use_mlock; // force system to keep model in RAM
bool check_tensors; // validate model tensor data
bool use_extra_bufts; // use extra buffer types (used for weight repacking)
bool no_host; // bypass host buffer allowing extra buffers to be used
};
// NOTE: changing the default values of parameters marked as [EXPERIMENTAL] may cause crashes or incorrect results in certain configurations
@@ -336,6 +312,7 @@ extern "C" {
enum llama_rope_scaling_type rope_scaling_type; // RoPE scaling type, from `enum llama_rope_scaling_type`
enum llama_pooling_type pooling_type; // whether to pool (sum) embedding results by sequence id
enum llama_attention_type attention_type; // attention type to use for embeddings
enum llama_flash_attn_type flash_attn_type; // when to enable Flash Attention
// ref: https://github.com/ggml-org/llama.cpp/pull/2054
float rope_freq_base; // RoPE base frequency, 0 = from model
@@ -345,7 +322,7 @@ extern "C" {
float yarn_beta_fast; // YaRN low correction dim
float yarn_beta_slow; // YaRN high correction dim
uint32_t yarn_orig_ctx; // YaRN original context size
float defrag_thold; // defragment the KV cache if holes/size > thold, < 0 disabled (default)
float defrag_thold; // [DEPRECATED] defragment the KV cache if holes/size > thold, <= 0 disabled (default)
ggml_backend_sched_eval_callback cb_eval;
void * cb_eval_user_data;
@@ -361,10 +338,15 @@ extern "C" {
// Keep the booleans together and at the end of the struct to avoid misalignment during copy-by-value.
bool embeddings; // if true, extract embeddings (together with logits)
bool offload_kqv; // whether to offload the KQV ops (including the KV cache) to GPU
bool flash_attn; // whether to use flash attention [EXPERIMENTAL]
bool no_perf; // whether to measure performance timings
bool op_offload; // whether to offload host tensor operations to device
bool offload_kqv; // offload the KQV ops (including the KV cache) to GPU
bool no_perf; // measure performance timings
bool op_offload; // offload host tensor operations to device
bool swa_full; // use full-size SWA cache (https://github.com/ggml-org/llama.cpp/pull/13194#issuecomment-2868343055)
// NOTE: setting to false when n_seq_max > 1 can cause bad performance in some cases
// ref: https://github.com/ggml-org/llama.cpp/pull/13845#issuecomment-2924800573
bool kv_unified; // use a unified buffer across the input sequences when computing the attention
// try to disable when n_seq_max > 1 for improved performance when the sequences do not share a large prefix
// ref: https://github.com/ggml-org/llama.cpp/pull/14363
};
// model quantization parameters
@@ -381,6 +363,7 @@ extern "C" {
void * imatrix; // pointer to importance matrix data
void * kv_overrides; // pointer to vector containing overrides
void * tensor_types; // pointer to vector containing tensor types
void * prune_layers; // pointer to vector containing layer indices to prune
} llama_model_quantize_params;
typedef struct llama_logit_bias {
@@ -470,6 +453,7 @@ extern "C" {
LLAMA_API int64_t llama_time_us(void);
LLAMA_API size_t llama_max_devices(void);
LLAMA_API size_t llama_max_parallel_sequences(void);
LLAMA_API bool llama_supports_mmap (void);
LLAMA_API bool llama_supports_mlock (void);
@@ -489,7 +473,7 @@ extern "C" {
DEPRECATED(LLAMA_API int32_t llama_n_vocab (const struct llama_vocab * vocab), "use llama_vocab_n_tokens instead");
LLAMA_API const struct llama_model * llama_get_model (const struct llama_context * ctx);
LLAMA_API struct llama_kv_cache * llama_get_kv_self ( struct llama_context * ctx);
LLAMA_API llama_memory_t llama_get_memory (const struct llama_context * ctx);
LLAMA_API enum llama_pooling_type llama_pooling_type(const struct llama_context * ctx); // TODO: rename to llama_get_pooling_type
LLAMA_API const struct llama_vocab * llama_model_get_vocab(const struct llama_model * model);
@@ -500,10 +484,18 @@ extern "C" {
LLAMA_API int32_t llama_model_n_layer (const struct llama_model * model);
LLAMA_API int32_t llama_model_n_head (const struct llama_model * model);
LLAMA_API int32_t llama_model_n_head_kv (const struct llama_model * model);
LLAMA_API int32_t llama_model_n_swa (const struct llama_model * model);
// Get the model's RoPE frequency scaling factor
LLAMA_API float llama_model_rope_freq_scale_train(const struct llama_model * model);
// Returns the number of classifier outputs (only valid for classifier models)
// Undefined behavior for non-classifier models
LLAMA_API uint32_t llama_model_n_cls_out(const struct llama_model * model);
// Returns label of classifier output by index (<n_cls_out). Returns nullptr if no label provided
LLAMA_API const char * llama_model_cls_label(const struct llama_model * model, uint32_t i);
LLAMA_API enum llama_vocab_type llama_vocab_type(const struct llama_vocab * vocab);
LLAMA_API int32_t llama_vocab_n_tokens(const struct llama_vocab * vocab);
@@ -552,6 +544,12 @@ extern "C" {
// Returns true if the model is recurrent (like Mamba, RWKV, etc.)
LLAMA_API bool llama_model_is_recurrent(const struct llama_model * model);
// Returns true if the model is hybrid (like Jamba, Granite, etc.)
LLAMA_API bool llama_model_is_hybrid(const struct llama_model * model);
// Returns true if the model is diffusion-based (like LLaDA, Dream, etc.)
LLAMA_API bool llama_model_is_diffusion(const struct llama_model * model);
// Returns 0 on success
LLAMA_API uint32_t llama_model_quantize(
const char * fname_inp,
@@ -567,10 +565,32 @@ extern "C" {
struct llama_model * model,
const char * path_lora);
// Functions to access the adapter's GGUF metadata scalar values
// - The functions return the length of the string on success, or -1 on failure
// - The output string is always null-terminated and cleared on failure
// - When retrieving a string, an extra byte must be allocated to account for the null terminator
// - GGUF array values are not supported by these functions
// Get metadata value as a string by key name
LLAMA_API int32_t llama_adapter_meta_val_str(const struct llama_adapter_lora * adapter, const char * key, char * buf, size_t buf_size);
// Get the number of metadata key/value pairs
LLAMA_API int32_t llama_adapter_meta_count(const struct llama_adapter_lora * adapter);
// Get metadata key name by index
LLAMA_API int32_t llama_adapter_meta_key_by_index(const struct llama_adapter_lora * adapter, int32_t i, char * buf, size_t buf_size);
// Get metadata value as a string by index
LLAMA_API int32_t llama_adapter_meta_val_str_by_index(const struct llama_adapter_lora * adapter, int32_t i, char * buf, size_t buf_size);
// Manually free a LoRA adapter
// Note: loaded adapters will be free when the associated model is deleted
LLAMA_API void llama_adapter_lora_free(struct llama_adapter_lora * adapter);
// Get the invocation tokens if the current lora is an alora
LLAMA_API uint64_t llama_adapter_get_alora_n_invocation_tokens(const struct llama_adapter_lora * adapter);
LLAMA_API const llama_token * llama_adapter_get_alora_invocation_tokens (const struct llama_adapter_lora * adapter);
// The following functions operate on a llama_context, hence the naming: llama_verb_...
// Add a loaded LoRA adapter to given context
@@ -604,210 +624,85 @@ extern "C" {
int32_t il_end);
//
// KV cache
// Memory
//
// TODO: start using struct llama_kv_cache
// Information associated with an individual cell in the KV cache view.
struct llama_kv_cache_view_cell {
// The position for this cell. Takes KV cache shifts into account.
// May be negative if the cell is not populated.
llama_pos pos;
};
// An updateable view of the KV cache.
struct llama_kv_cache_view {
// Number of KV cache cells. This will be the same as the context size.
int32_t n_cells;
// Maximum number of sequences that can exist in a cell. It's not an error
// if there are more sequences in a cell than this value, however they will
// not be visible in the view cells_sequences.
int32_t n_seq_max;
// Number of tokens in the cache. For example, if there are two populated
// cells, the first with 1 sequence id in it and the second with 2 sequence
// ids then you'll have 3 tokens.
int32_t token_count;
// Number of populated cache cells.
int32_t used_cells;
// Maximum contiguous empty slots in the cache.
int32_t max_contiguous;
// Index to the start of the max_contiguous slot range. Can be negative
// when cache is full.
int32_t max_contiguous_idx;
// Information for an individual cell.
struct llama_kv_cache_view_cell * cells;
// The sequences for each cell. There will be n_seq_max items per cell.
llama_seq_id * cells_sequences;
};
// Create an empty KV cache view. (use only for debugging purposes)
LLAMA_API struct llama_kv_cache_view llama_kv_cache_view_init(const struct llama_context * ctx, int32_t n_seq_max);
// Free a KV cache view. (use only for debugging purposes)
LLAMA_API void llama_kv_cache_view_free(struct llama_kv_cache_view * view);
// Update the KV cache view structure with the current state of the KV cache. (use only for debugging purposes)
// TODO: change signature to llama_kv_cache_view_update(struct llama_kv_cache_view * view, const struct llama_context * ctx)
LLAMA_API void llama_kv_cache_view_update(const struct llama_context * ctx, struct llama_kv_cache_view * view);
///
// Returns the number of tokens in the KV cache (slow, use only for debug)
// If a KV cell has multiple sequences assigned to it, it will be counted multiple times
LLAMA_API int32_t llama_kv_self_n_tokens(const struct llama_context * ctx);
DEPRECATED(LLAMA_API int32_t llama_get_kv_cache_token_count(const struct llama_context * ctx),
"use llama_kv_self_n_tokens instead");
// Returns the number of used KV cells (i.e. have at least one sequence assigned to them)
LLAMA_API int32_t llama_kv_self_used_cells(const struct llama_context * ctx);
DEPRECATED(LLAMA_API int32_t llama_get_kv_cache_used_cells(const struct llama_context * ctx),
"use llama_kv_self_used_cells instead");
// Clear the KV cache - both cell info is erased and KV data is zeroed
LLAMA_API void llama_kv_self_clear(
struct llama_context * ctx);
// Clear the memory contents
// If data == true, the data buffers will also be cleared together with the metadata
LLAMA_API void llama_memory_clear(
llama_memory_t mem,
bool data);
// Removes all tokens that belong to the specified sequence and have positions in [p0, p1)
// Returns false if a partial sequence cannot be removed. Removing a whole sequence never fails
// seq_id < 0 : match any sequence
// p0 < 0 : [0, p1]
// p1 < 0 : [p0, inf)
LLAMA_API bool llama_kv_self_seq_rm(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1);
LLAMA_API bool llama_memory_seq_rm(
llama_memory_t mem,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1);
// Copy all tokens that belong to the specified sequence to another sequence
// Note that this does not allocate extra KV cache memory - it simply assigns the tokens to the new sequence
// p0 < 0 : [0, p1]
// p1 < 0 : [p0, inf)
LLAMA_API void llama_kv_self_seq_cp(
struct llama_context * ctx,
llama_seq_id seq_id_src,
llama_seq_id seq_id_dst,
llama_pos p0,
llama_pos p1);
LLAMA_API void llama_memory_seq_cp(
llama_memory_t mem,
llama_seq_id seq_id_src,
llama_seq_id seq_id_dst,
llama_pos p0,
llama_pos p1);
// Removes all tokens that do not belong to the specified sequence
LLAMA_API void llama_kv_self_seq_keep(
struct llama_context * ctx,
llama_seq_id seq_id);
LLAMA_API void llama_memory_seq_keep(
llama_memory_t mem,
llama_seq_id seq_id);
// Adds relative position "delta" to all tokens that belong to the specified sequence and have positions in [p0, p1)
// If the KV cache is RoPEd, the KV data is updated accordingly:
// - lazily on next llama_decode()
// - explicitly with llama_kv_self_update()
// p0 < 0 : [0, p1]
// p1 < 0 : [p0, inf)
LLAMA_API void llama_kv_self_seq_add(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1,
llama_pos delta);
LLAMA_API void llama_memory_seq_add(
llama_memory_t mem,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1,
llama_pos delta);
// Integer division of the positions by factor of `d > 1`
// If the KV cache is RoPEd, the KV data is updated accordingly:
// - lazily on next llama_decode()
// - explicitly with llama_kv_self_update()
// p0 < 0 : [0, p1]
// p1 < 0 : [p0, inf)
LLAMA_API void llama_kv_self_seq_div(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1,
int d);
LLAMA_API void llama_memory_seq_div(
llama_memory_t mem,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1,
int d);
// Returns the largest position present in the KV cache for the specified sequence
LLAMA_API llama_pos llama_kv_self_seq_pos_max(
struct llama_context * ctx,
llama_seq_id seq_id);
// Returns the smallest position present in the memory for the specified sequence
// This is typically non-zero only for SWA caches
// Note that all positions in the range [pos_min, pos_max] are guaranteed to be present in the memory
// Return -1 if the sequence is empty
LLAMA_API llama_pos llama_memory_seq_pos_min(
llama_memory_t mem,
llama_seq_id seq_id);
// Defragment the KV cache
// This will be applied:
// - lazily on next llama_decode()
// - explicitly with llama_kv_self_update()
LLAMA_API void llama_kv_self_defrag(struct llama_context * ctx);
// Check if the context supports KV cache shifting
LLAMA_API bool llama_kv_self_can_shift(const struct llama_context * ctx);
// Apply the KV cache updates (such as K-shifts, defragmentation, etc.)
LLAMA_API void llama_kv_self_update(struct llama_context * ctx);
DEPRECATED(LLAMA_API void llama_kv_cache_clear(
struct llama_context * ctx),
"use llama_kv_self_clear instead");
DEPRECATED(LLAMA_API bool llama_kv_cache_seq_rm(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1),
"use llama_kv_self_seq_rm instead");
DEPRECATED(LLAMA_API void llama_kv_cache_seq_cp(
struct llama_context * ctx,
llama_seq_id seq_id_src,
llama_seq_id seq_id_dst,
llama_pos p0,
llama_pos p1),
"use llama_kv_self_seq_cp instead");
DEPRECATED(LLAMA_API void llama_kv_cache_seq_keep(
struct llama_context * ctx,
llama_seq_id seq_id),
"use llama_kv_self_seq_keep instead");
DEPRECATED(LLAMA_API void llama_kv_cache_seq_add(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1,
llama_pos delta),
"use llama_kv_self_seq_add instead");
DEPRECATED(LLAMA_API void llama_kv_cache_seq_div(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1,
int d),
"use llama_kv_self_seq_div instead");
DEPRECATED(LLAMA_API llama_pos llama_kv_cache_seq_pos_max(
struct llama_context * ctx,
llama_seq_id seq_id),
"use llama_kv_self_seq_pos_max instead");
DEPRECATED(LLAMA_API void llama_kv_cache_defrag(struct llama_context * ctx),
"use llama_kv_self_defrag instead");
DEPRECATED(LLAMA_API bool llama_kv_cache_can_shift(const struct llama_context * ctx),
"use llama_kv_self_can_shift instead");
DEPRECATED(LLAMA_API void llama_kv_cache_update(struct llama_context * ctx),
"use llama_kv_self_update instead");
// Returns the largest position present in the memory for the specified sequence
// Note that all positions in the range [pos_min, pos_max] are guaranteed to be present in the memory
// Return -1 if the sequence is empty
LLAMA_API llama_pos llama_memory_seq_pos_max(
llama_memory_t mem,
llama_seq_id seq_id);
// Check if the memory supports shifting
LLAMA_API bool llama_memory_can_shift(llama_memory_t mem);
//
// State / sessions
//
// Returns the *actual* size in bytes of the state
// (logits, embedding and kv_cache)
// (logits, embedding and memory)
// Only use when saving the state, not when restoring it, otherwise the size may be too small.
LLAMA_API size_t llama_state_get_size(struct llama_context * ctx);
LLAMA_API DEPRECATED(size_t llama_get_state_size(struct llama_context * ctx),
@@ -863,12 +758,12 @@ extern "C" {
size_t n_token_count),
"use llama_state_save_file instead");
// Get the exact size needed to copy the KV cache of a single sequence
// Get the exact size needed to copy the state of a single sequence
LLAMA_API size_t llama_state_seq_get_size(
struct llama_context * ctx,
llama_seq_id seq_id);
// Copy the KV cache of a single sequence into the specified buffer
// Copy the state of a single sequence into the specified buffer
LLAMA_API size_t llama_state_seq_get_data(
struct llama_context * ctx,
uint8_t * dst,
@@ -900,6 +795,33 @@ extern "C" {
size_t n_token_capacity,
size_t * n_token_count_out);
// for backwards-compat
#define LLAMA_STATE_SEQ_FLAGS_SWA_ONLY 1
// work only with partial states, such as SWA KV cache or recurrent cache (e.g. Mamba)
#define LLAMA_STATE_SEQ_FLAGS_PARTIAL_ONLY 1
typedef uint32_t llama_state_seq_flags;
LLAMA_API size_t llama_state_seq_get_size_ext(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_state_seq_flags flags);
LLAMA_API size_t llama_state_seq_get_data_ext(
struct llama_context * ctx,
uint8_t * dst,
size_t size,
llama_seq_id seq_id,
llama_state_seq_flags flags);
LLAMA_API size_t llama_state_seq_set_data_ext(
struct llama_context * ctx,
const uint8_t * src,
size_t size,
llama_seq_id dest_seq_id,
llama_state_seq_flags flags);
//
// Decoding
//
@@ -934,18 +856,23 @@ extern "C" {
// For encode-decoder contexts, processes the batch using the encoder.
// Can store the encoder output internally for later use by the decoder's cross-attention layers.
// 0 - success
// < 0 - error. the KV cache state is restored to the state before this call
// < 0 - error. the memory state is restored to the state before this call
LLAMA_API int32_t llama_encode(
struct llama_context * ctx,
struct llama_batch batch);
// Process a batch of tokens.
// Requires KV cache.
// Requires the context to have a memory.
// For encode-decoder contexts, processes the batch using the decoder.
// Positive return values does not mean a fatal error, but rather a warning.
// 0 - success
// 1 - could not find a KV slot for the batch (try reducing the size of the batch or increase the context)
// < 0 - error. the KV cache state is restored to the state before this call
// Upon fatal-error or abort, the ubatches that managed to be been processed will remain in the memory state of the context
// To handle this correctly, query the memory state using llama_memory_seq_pos_min() and llama_memory_seq_pos_max()
// Upon other return values, the memory state is restored to the state before this call
// 0 - success
// 1 - could not find a KV slot for the batch (try reducing the size of the batch or increase the context)
// 2 - aborted (processed ubatches will remain in the context's memory)
// -1 - invalid input batch
// < -1 - fatal error (processed ubatches will remain in the context's memory)
LLAMA_API int32_t llama_decode(
struct llama_context * ctx,
struct llama_batch batch);
@@ -961,8 +888,8 @@ extern "C" {
// Get the number of threads used for prompt and batch processing (multiple token).
LLAMA_API int32_t llama_n_threads_batch(struct llama_context * ctx);
// Set whether the model is in embeddings mode or not
// If true, embeddings will be returned but logits will not
// Set whether the context outputs embeddings or not
// TODO: rename to avoid confusion with llama_get_embeddings()
LLAMA_API void llama_set_embeddings(struct llama_context * ctx, bool embeddings);
// Set whether to use causal attention or not
@@ -986,6 +913,7 @@ extern "C" {
// in the order they have appeared in the batch.
// Rows: number of tokens for which llama_batch.logits[i] != 0
// Cols: n_vocab
// TODO: deprecate in favor of llama_get_logits_ith() (ref: https://github.com/ggml-org/llama.cpp/pull/14853#issuecomment-3113143522)
LLAMA_API float * llama_get_logits(struct llama_context * ctx);
// Logits for the ith token. For positive indices, Equivalent to:
@@ -1000,6 +928,7 @@ extern "C" {
// in the order they have appeared in the batch.
// shape: [n_outputs*n_embd]
// Otherwise, returns NULL.
// TODO: deprecate in favor of llama_get_embeddings_ith() (ref: https://github.com/ggml-org/llama.cpp/pull/14853#issuecomment-3113143522)
LLAMA_API float * llama_get_embeddings(struct llama_context * ctx);
// Get the embeddings for the ith token. For positive indices, Equivalent to:
@@ -1011,7 +940,7 @@ extern "C" {
// Get the embeddings for a sequence id
// Returns NULL if pooling_type is LLAMA_POOLING_TYPE_NONE
// when pooling_type == LLAMA_POOLING_TYPE_RANK, returns float[1] with the rank of the sequence
// when pooling_type == LLAMA_POOLING_TYPE_RANK, returns float[n_cls_out] with the rank(s) of the sequence
// otherwise: float[n_embd] (1-dimensional)
LLAMA_API float * llama_get_embeddings_seq(struct llama_context * ctx, llama_seq_id seq_id);
@@ -1038,9 +967,11 @@ extern "C" {
LLAMA_API llama_token llama_vocab_sep(const struct llama_vocab * vocab); // sentence separator
LLAMA_API llama_token llama_vocab_nl (const struct llama_vocab * vocab); // next-line
LLAMA_API llama_token llama_vocab_pad(const struct llama_vocab * vocab); // padding
LLAMA_API llama_token llama_vocab_mask(const struct llama_vocab * vocab); // mask
LLAMA_API bool llama_vocab_get_add_bos(const struct llama_vocab * vocab);
LLAMA_API bool llama_vocab_get_add_eos(const struct llama_vocab * vocab);
LLAMA_API bool llama_vocab_get_add_sep(const struct llama_vocab * vocab);
LLAMA_API llama_token llama_vocab_fim_pre(const struct llama_vocab * vocab);
LLAMA_API llama_token llama_vocab_fim_suf(const struct llama_vocab * vocab);
@@ -1084,6 +1015,7 @@ extern "C" {
/// @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.
@@ -1236,11 +1168,6 @@ extern "C" {
LLAMA_API struct llama_sampler * llama_sampler_init_greedy(void);
LLAMA_API struct llama_sampler * llama_sampler_init_dist (uint32_t seed);
/// @details Sorts candidate tokens by their logits in descending order and calculate probabilities based on logits.
/// NOTE: Avoid using on the full vocabulary as the sorting can become slow. For example, apply top-k or top-p sampling first.
DEPRECATED(LLAMA_API struct llama_sampler * llama_sampler_init_softmax (void),
"will be removed in the future (see https://github.com/ggml-org/llama.cpp/pull/9896#discussion_r1800920915)");
/// @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_API struct llama_sampler * llama_sampler_init_top_k (int32_t k);
@@ -1410,23 +1337,25 @@ extern "C" {
//
// Performance utils
//
// NOTE: Used by llama.cpp examples, avoid using in third-party apps. Instead, do your own performance measurements.
// NOTE: Used by llama.cpp examples/tools, avoid using in third-party apps. Instead, do your own performance measurements.
//
struct llama_perf_context_data {
double t_start_ms;
double t_load_ms;
double t_p_eval_ms;
double t_eval_ms;
// ms == milliseconds
double t_start_ms; // absolute start time
double t_load_ms; // time needed for loading the model
double t_p_eval_ms; // time needed for processing the prompt
double t_eval_ms; // time needed for generating tokens
int32_t n_p_eval;
int32_t n_eval;
int32_t n_p_eval; // number of prompt tokens
int32_t n_eval; // number of generated tokens
int32_t n_reused; // number of times a ggml compute graph had been reused
};
struct llama_perf_sampler_data {
double t_sample_ms;
double t_sample_ms; // time needed for sampling in ms
int32_t n_sample;
int32_t n_sample; // number of sampled tokens
};
LLAMA_API struct llama_perf_context_data llama_perf_context (const struct llama_context * ctx);
@@ -1438,6 +1367,9 @@ extern "C" {
LLAMA_API void llama_perf_sampler_print(const struct llama_sampler * chain);
LLAMA_API void llama_perf_sampler_reset( struct llama_sampler * chain);
// print a breakdown of per-device memory use via LLAMA_LOG:
LLAMA_API void llama_memory_breakdown_print(const struct llama_context * ctx);
//
// training
//
@@ -1456,6 +1388,8 @@ extern "C" {
ggml_opt_get_optimizer_params get_opt_pars; // callback for calculating optimizer parameters
void * get_opt_pars_ud; // userdata for calculating optimizer parameters
enum ggml_opt_optimizer_type optimizer_type;
};
LLAMA_API void llama_opt_init(struct llama_context * lctx, struct llama_model * model, struct llama_opt_params lopt_params);