summaryrefslogtreecommitdiff
path: root/llama.cpp
AgeCommit message (Collapse)Author
2023-12-01build : fix build info generation and cleanup Makefile (#3920)Jared Van Bortel
* cmake : fix joining of REAL_GIT_DIR * fix includes with help from include-what-you-use * make : remove unneeded deps and add test-rope target * fix C includes in C++ source files * Revert "fix includes with help from include-what-you-use" This reverts commit 635e9fadfd516d4604a0fecf4a854bfb25ad17ae.
2023-11-30llama : fix alignment of general.name in print meta (#4254)Daniel Bevenius
* llama: fix alignment of general.name in print meta This commit fixes the alignment of the general.name field in the llm_load_print_meta function. Currently the output looks like this: ```console llm_load_print_meta: model ftype = mostly Q4_0 llm_load_print_meta: model params = 13.02 B llm_load_print_meta: model size = 6.86 GiB (4.53 BPW) llm_load_print_meta: general.name = LLaMA v2 ``` And with this commit it looks like this: ```console llm_load_print_meta: model ftype = mostly Q4_0 llm_load_print_meta: model params = 13.02 B llm_load_print_meta: model size = 6.86 GiB (4.53 BPW) llm_load_print_meta: general.name = LLaMA v2 ``` Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com> * llama: fix alignment of special tokens Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com> --------- Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
2023-11-30llama : fix typical sampling (#4261)tarcey
Typical sampling was broken because after copying new_candidates into canditates, the "sorted" bool is left at "true", but the new data is no longer sorted according to probability. Patch to set "sorted" to false. Test: Generating with temp=0.0001 (approx. argmax) should generate the same sequence at typical>=1.0 and typical=0.9999 (approx. disabled, but enters the typical sampling codepath).
2023-11-28ggml : re-enable BLAS for CPU when src0 != F32 + remove redundant full ↵Georgi Gerganov
offload checks in llama.cpp (#4240) * ggml : use blas even if src0 is not F32 * llama : use n_threads_batch only when n_tokens >= 32 ggml-ci * llama : revert n_threads_batch logic ggml-ci
2023-11-25llama : grammar `reserve` space in `decode_utf8` (#4210)Marcus Dunn
* reserve space for codepoints * improvement for the appended 0
2023-11-24llama : set metal log callback correctly (#4204)slaren
2023-11-24ggml-cuda : support stablelm rope (#4156)slaren
* ggml-cuda : support stablelm rope * remove unused freq_base kernel parameter * add n_dims parameter to llm_build_k_shift, default to n_rot via overload * llama : fix llm_build_k_shift args --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-11-23llama : KV cache view API + better KV cache management (#4170)Georgi Gerganov
* llama : keep track of used KV cells + better KV cache management * llama : zero KV cache used upon clear ggml-ci * llama : allow exporting a view of the KV cache (#4180) * Allow exporting a view of the KV cache * Allow dumping the sequences per cell in common * Track max contiguous cells value and position as well * Fix max contiguous empty cells index calculation Make dump functions deal with lengths or sequences counts > 10 better * Fix off by one error in dump_kv_cache_view * Add doc comments for KV cache view functions Eliminate cell sequence struct; use llama_seq_id directly Minor cleanups * common : add -dkvc arg for enabling kv cache dumps --------- Co-authored-by: Kerfuffle <44031344+KerfuffleV2@users.noreply.github.com>
2023-11-21stablelm : simplify + speedup generation (#4153)Galunid
2023-11-19gguf-py : export chat templates (#4125)slaren
* gguf-py : export chat templates * llama.cpp : escape new lines in gguf kv info prints * gguf-py : bump version * gguf-py : check chat_template type * gguf-py : initialize chat_template
2023-11-17llama : increase max nodes (#4115)slaren
2023-11-17llama : add functions to get the model's metadata (#4013)slaren
* llama : add functions to get the model's metadata * format -> std::to_string * better documentation
2023-11-17llama : fix data units (#4101)Georgi Gerganov
* llama : fix data units ggml-ci * Revert "llama : fix data units" This reverts commit f5feac831fe225ed7f3db938d115732a49dccfc4. * llama : disambiguate data units ggml-ci
2023-11-16Respect tokenizer.ggml.add_bos_token value when tokenizing (#4040)Kerfuffle
* gguf-py: gguf-dump: Respect --no-tensor flag in JSON mode. * Respect add_bos_token GGUF metadata value * gguf-py: Try to fix SpecialVocab giving up too easily for the Nth time
2023-11-15llama : restore prefix space in llama tokenizer (#4081)Jared Van Bortel
2023-11-14stablelm : StableLM support (#3586)Galunid
* Add support for stablelm-3b-4e1t * Supports GPU offloading of (n-1) layers
2023-11-13sync : ggml (backend v2) (#3912)Georgi Gerganov
* sync : ggml (backend v2) (wip) * sync : migrate examples and llama.cpp to dynamic graphs (wip) * sync : update tests + fix max op params to 64 ggml-ci * sync : ggml-cuda ggml-ci * llama : fix save/load state context size ggml-ci * sync : try to fix build on tvOS * sync : pass custom graph sizes in training examples * sync : update graph copies to new ggml API * sync : update sync-ggml.sh with new files * scripts : fix header in sync script * train : fix context size calculations * llama : increase inference graph size up to 4096 nodes * train : allocate grads for backward graphs * train : allocate grads for gb_tmp
2023-11-13Add ReLU and SQR CUDA ops to (partially) fix Persimmon offloading (#4041)Kerfuffle
* Add ReLU and SQR CUDA ops to fix Persimmon offloading * Persimmon loader: More helpful error on CUDA/ROCM when offloading too many layers
2023-11-10Unbreak persimmon after #3837 (#4010)Galunid
2023-11-07cuda : supports running on CPU for GGML_USE_CUBLAS=ON build (#3946)Meng Zhang
* protyping the idea that supports running on CPU for a GGML_USE_CUBLAS=on build * doc: add comments to ggml_cublas_loaded() * fix defined(...)
2023-11-05llama : mark LLM_ARCH_STARCODER as full offload supported (#3945)Meng Zhang
as done in https://github.com/ggerganov/llama.cpp/pull/3827
2023-11-03llama : change yarn_ext_factor placeholder to -1 (#3922)cebtenzzre
2023-11-02llm : prevent from 1-D tensors being GPU split (#3697)Georgi Gerganov
2023-11-01llama : fix llama_context_default_params after #2268 (#3893)cebtenzzre
2023-11-01llama : implement YaRN RoPE scaling (#2268)cebtenzzre
Co-authored-by: cebtenzzre <cebtenzzre@gmail.com> Co-authored-by: Jeffrey Quesnelle <jquesnelle@gmail.com>
2023-11-01llm : fix llm_build_kqv taking unused tensor (benign, #3837)Georgi Gerganov
2023-11-01llm : fix falcon norm after refactoring (#3837)Georgi Gerganov
2023-11-01llm : add llm_build_context (#3881)Georgi Gerganov
* llm : add llm_build_context * llm : deduce norm eps based on type + explict max_alibi_bias, clamp_kqv * llm : restore the non-graph llm_build_ functional API ggml-ci * llm : cleanup + comments
2023-11-01finetune : add -ngl parameter (#3762)Andrew Godfrey
* Add '-ngl' support to finetune.cpp * Add fprintf in ggml_cuda_op_add When I tried CUDA offloading during finetuning following the readme, I got an assert here. This probably isn't an important case because inference later gives a warning saying you should use f16 or f32 instead when using lora * Add 'finetune.sh', which currently fails when using GPU "error: operator (): Finetuning on tensors with type 'f16' is not yet supported" * tweak finetune.sh * Suppress some warnings in ggml.c * Add f16 implementation to ggml_compute_forward_add_f16_f32 * Add an f16 case to ggml_add_cast_impl and llama_build_lora_finetune_graphs * finetune.sh: Edit comments * Add "add_f16_f32_f32_cuda" * Tweak an error message * finetune.sh: Add an optional LLAMA_MODEL_DIR variable * finetune.sh: Add an optional LLAMA_TRAINING_DIR variable * train : minor * tabs to spaces --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> Co-authored-by: cebtenzzre <cebtenzzre@gmail.com>
2023-11-01llama : refactor graph build code (#3837)Georgi Gerganov
* llama : factor out ggml-alloc from graph graph build functions ggml-ci * metal : disable kernel load log * llama : factor out tensor offloading outside the build call (wip) ggml-ci * llama : offload rest of the models ggml-ci * llama : update offload log messages to print node index * llama : comments * llama : support offloading result_norm + comments * llama : factor graph input into a function * llama : do tensor offload only with CUDA * llama : fix res_norm offloading * llama : try to optimize offloading code * llama : fix non-CUDA build * llama : try to fix build * llama : move refact in correct place + optimize graph input * llama : refactor tensor offloading as callback * llama : add layer index to all tensor names * llama : add functional header * llama : comment ggml-ci * llama : remove obsolete map for layer counting * llama : add llm_build helper functions (#3848) * llama : add llm_build_norm helper function ggml-ci * llama : add llm_build_ffn helper function (#3849) ggml-ci * llama : add llm_build_k_shift helper ggml-ci * llama : fix offloading after recent changes * llama : add llm_build_kv_store helper ggml-ci * llama : remove obsolete offload names * llama : fix llm_build_k_shift to use n_head_kv instead of n_head * llama : simplify falcon Q, K, V computation * llama : remove obsolete comments in build graphs * llama : add llm_build_kqv helper ggml-ci * llama : minor * llama : add LLAMA_OFFLOAD_DEBUG + fix starcoder offloading * llama : fix input allocation logic * llama : update offload functions for KQ tensors * llama : normalize tensor names ggml-ci * llama : enable warning about not offloaded tensors * llama : remove extra ; + deduplicate gate_b logic * llama : add llm_build_inp_embd helper
2023-10-31samplers : Min-P sampler implementation [alternative to Top P/Top K] (#3841)kalomaze
* Introduce the new Min-P sampler by @kalomaze The Min-P sampling method was designed as an alternative to Top-P, and aims to ensure a balance of quality and variety. The parameter *p* represents the minimum probability for a token to be considered, relative to the probability of the most likely token. * Min-P enabled and set to 0.05 default --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> Co-authored-by: cebtenzzre <cebtenzzre@gmail.com>
2023-10-30ggml : move FP16 <-> FP32 code to ggml-impl.h (#3861)Georgi Gerganov
* ggml : move FP16 <-> FP32 stuff to ggml-impl.h ggml-ci * tests : fix ARM build * ggml : explicitly initialize deprecated type traits * ggml : add math.h to ggml-impl.h * ggml : remove duplicate static assert macros * ggml : prefix lookup tables with ggml_ ggml-ci * ggml-impl : move extern "C" to start of file
2023-10-29Extend llama_kv_cache_seq_rm to allow matching any sequence (#3843)Kerfuffle
* Extend llama_kv_cache_seq_rm to allow matichng any sequence * Replace llama_kv_cache_tokens_rm with llama_kv_cache_clear Use llama_kv_cache_clear for cache clearing Change calls to llama_kv_cache_tokens_rm that want to delete by position to use llama_kv_cache_seq_rm functionality
2023-10-29llama : fix kv shift bug (#3835)Georgi Gerganov
ggml-ci
2023-10-29ggml : quantization refactoring (#3833)Georgi Gerganov
* ggml : factor all quantization code in ggml-quants ggml-ci * ggml-quants : fix Zig and Swift builds + quantize tool ggml-ci * quantize : --pure option for disabling k-quant mixtures --------- Co-authored-by: cebtenzzre <cebtenzzre@gmail.com>
2023-10-28llama : allow quantizing k-quants to fall back when tensor size incompatible ↵Kerfuffle
(#3747) * Allow quantizing k-quants to fall back when tensor size incompatible * quantizing: Add warning when tensors were incompatible with k-quants Clean up k-quants state passing a bit
2023-10-28starcoder : add GPU offloading (#3827)Georgi Gerganov
* starcoder : do not GPU split 1D bias tensors * starcoder : offload layers to GPU ggml-ci
2023-10-27llama : correctly report GGUFv3 format (#3818)cebtenzzre
2023-10-27cuda : improve text-generation and batched decoding performance (#3776)Georgi Gerganov
* cuda : prints wip * cuda : new cublas gemm branch for multi-batch quantized src0 * cuda : add F32 sgemm branch * cuda : fine-tune >= VOLTA params + use MMQ only for small batches * cuda : remove duplicated cuBLAS GEMM code * cuda : add CUDA_USE_TENSOR_CORES and GGML_CUDA_FORCE_MMQ macros * build : add compile option to force use of MMQ kernels
2023-10-23llama : remove token functions with `context` args in favor of `model` (#3720)Marcus Dunn
* added `llama_model_token_*` variants to all the `llama_token_*` functions. * added `LLAMA_API` * formatting Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * removed old `llama_token` functions * changed 3 more functions to take in model - `llama_token_get_text` - `llama_token_get_score` - `llama_token_get_type` * added back docs * fixed main.cpp * changed token functions to use new model variants * changed token functions to use new model variants --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-10-22Add test for MPT tokenization (#3728)goerch
* Add test for MPT tokenization * Revert code motion * Remove unnecessary restriction in test case * Clarify logic in conversion
2023-10-22llama : validate special token ids are in range when loading GGUF model (#3635)Kerfuffle
* Add validation for special token ids to llama.cpp Small optimization for llama_byte_to_token SPM mode * Fix BPE newline check, only I could break something so simple * Killll meeeeee * Account for GGUF_KEY_KEY only setting when the key exists * Minor code cleanups. * Fix convert.py error msg when added tokens are out of range * Make gguf SpecialVocab vocab size-aware Update conversion scripts accordingly * Avoid a string copy Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-10-20sampling : refactor init to use llama_sampling_params (#3696)Georgi Gerganov
* sampling : refactor init to use llama_sampling_params * llama : combine repetition, frequency and presence penalties in 1 call * examples : remove embd-input and gptneox-wip * sampling : rename penalty params + reduce size of "prev" vector * sampling : add llama_sampling_print helper * sampling : hide prev behind API and apply #3661 ggml-ci
2023-10-20ggml : fix rope + llama minor optimizations (#3560)Herman Semenov
* Minor fixes and fixed memleak * Using const auto references in range-based loop C++17
2023-10-18speculative : add tree-based sampling example (#3624)Georgi Gerganov
* sampling : one sequence per sampling context ggml-ci * speculative : add tree-based sampling support ggml-ci * speculative : reuse the n_parallel CLI param * speculative : refactor sampling * examples : fix build after sampling refactoring ggml-ci * batched : fix n_seq_id * sampling : fix malloc ggml-ci * swift : fix build ggml-ci * swift : try to fix build ggml-ci * prompts : add assistant.txt * common : add llama_batch_add() and llama_batch_clear() helpers * speculative : minor refactor ggml-ci * minor : comments + rename ggml-ci * speculative : fix off-by-one for n_drafted * speculative : fix the n_drafted fix + p constants
2023-10-17fix embeddings when using CUDA (#3657)slaren
2023-10-17llama : avoid fprintf in favor of LLAMA_LOG (#3538)Georgi Gerganov
2023-10-17tokenizer : special token handling (#3538)staviq
* Rewrite special token handling from #1931 * shorten param name, add st verification by type * use offsets instead of copy by substr * formatting, remove copying iterator on delete * llama : normalize code-style * swift fix * print pfx/sfx if verb, main: split pfx input sfx * dont add space when using special tokens * minor : comment + spacing --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-10-15MPT : support GQA for replit-code-v1.5 (#3627)cebtenzzre
2023-10-13llama : remove n_threads from llama_decode_internal (#3614)Daniel Bevenius
This commit removes `n_threads` from the `llama_decode_internal` functions doc comment as it does not exist anymore. It looks like this parameter was removed in Commit 16bc66d9479edd5ee12ec734973554d4493c5dfa ("llama.cpp : split llama_context_params into model and context params"). Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
generated by cgit v1.2.3 (git 2.25.1) at 2025-10-04 23:31:31 +0000