aboutsummaryrefslogtreecommitdiff
path: root/llama.cpp
AgeCommit message (Collapse)Author
2023-04-12Don't crash on ftype (formerly f16) == 4 (#917)Stephan Walter
2023-04-11Add enum llama_ftype, sync ggml_type to model files (#709)Stephan Walter
2023-04-11Windows fixes (#890)comex
Mostly for msys2 and mingw64 builds, which are different from each other and different from standard Visual Studio builds. Isn't Windows fun? - Define _GNU_SOURCE in more files (it's already used in ggml.c for Linux's sake). - Don't use PrefetchVirtualMemory if not building for Windows 8 or later (mingw64 doesn't by default). But warn the user about this situation since it's probably not intended. - Check for NOMINMAX already being defined, which it is on mingw64. - Actually use the `increment` variable (bug in my `pizza` PR). - Suppress unused variable warnings in the fake pthread_create and pthread_join implementations for Windows. - (not Windows-related) Remove mention of `asprintf` from comment; `asprintf` is no longer used. Fixes #871.
2023-04-10Print model version.comex
Also improve model type printing, and fix indentation of an unrelated switch statement.
2023-04-10Rewrite loading code to try to satisfy everyone:comex
- Support all three formats (ggml, ggmf, ggjt). (However, I didn't include the hack needed to support GPT4All files without conversion. Those can still be used after converting them with convert.py from my other PR.) - Support both mmap and read (mmap is used by default, but can be disabled with `--no-mmap`, and is automatically disabled for pre-ggjt files or on platforms where mmap is not supported). - Support multi-file models like before, but automatically determine the number of parts rather than requiring `--n_parts`. - Improve validation and error checking. - Stop using the per-file type field (f16) entirely in favor of just relying on the per-tensor type/size fields. This has no immediate benefit, but makes it easier to experiment with different formats, and should make it easier to support the new GPTQ-for-LLaMa models in the future (I have some work in progress on that front). - Support VirtualLock on Windows (using the same `--mlock` option as on Unix). - Indicate loading progress when using mmap + mlock. (Which led me to the interesting observation that on my Linux machine, with a warm file cache, mlock actually takes some time, whereas mmap without mlock starts almost instantly...) - To help implement this, move mlock support from ggml to the loading code. - madvise/PrefetchVirtualMemory support (based on #740) - Switch from ifstream to the `fopen` family of functions to avoid unnecessary copying and, when mmap is enabled, allow reusing the same file descriptor for both metadata reads and mmap (whereas the existing implementation opens the file a second time to mmap). - Quantization now produces a single-file output even with multi-file inputs (not really a feature as much as 'it was easier this way'). Implementation notes: I tried to factor the code into more discrete pieces than before. Regarding code style: I tried to follow the code style, but I'm naughty and used a few advanced C++ features repeatedly: - Destructors to make it easier to ensure everything gets cleaned up. - Exceptions. I don't even usually use exceptions when writing C++, and I can remove them if desired... but here they make the loading code much more succinct while still properly handling a variety of errors, ranging from API calls failing to integer overflow and allocation failure. The exceptions are converted to error codes at the API boundary.) Co-authored-by: Pavol Rusnak <pavol@rusnak.io> (for the bit I copied from #740)
2023-04-08Add quantize-stats command for testing quantization (#728)unbounded
Command that calculates some statistics over the errors introduced by quantization, like mean square error, max error and some percentile errors for layer weights. Should be useful for testing quantization improvements. Exposes some internal state from ggml and llama for testing
2023-04-07llama : always sort logits before nucleus sampling (#812)Ivan Stepanov
* Always sort logits before nucleus sampling * remove second normalization - fix windows build - remove normalization since std::discrete_distribution does not require it
2023-04-05ggml, llama : avoid heavy V transpose + improvements (#775)Georgi Gerganov
ggml : - added ggml_view_3d() - ggml_view_tensor() now inherits the stride too - reimplement ggml_cpy() to account for dst stride - no longer require tensor->data to be memory aligned llama : - compute RoPE on 32-bit tensors (should be more accurate) - store RoPE-ed K in the KV cache - store transposed V in the KV cache (significant speed-up) - avoid unnecessary Q copy
2023-04-05llama : define non-positive top_k; top_k range check (#779)Ivan Stepanov
* Define non-positive top_k; top_k range check * minor : brackets --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-04-03Define non-positive temperature behavior (#720)Ivan Stepanov
2023-04-02Added api for getting/setting the kv_cache (#685)Christian Falch
The api provides access methods for retrieving the current memory buffer for the kv_cache and its token number. It also contains a method for setting the kv_cache from a memory buffer. This makes it possible to load/save history - maybe support --cache-prompt paramater as well? Co-authored-by: Pavol Rusnak <pavol@rusnak.io>
2023-04-02ggml : change ne to int64_t (#626)Marian Cepok
2023-04-02llama : do not allocate KV cache for "vocab_only == true" (#682)Stephan Walter
Fixes sanitizer CI
2023-03-30Introduce GGML migration tool for new file formatJustine Tunney
If you deleted your old Meta LLaMA .pth files, then the migrate-ggml-2023-03-30-pr613.py script will allow you to convert your old ggml files into the new mmap()'able format. See #613
2023-03-30Ensure --mlock works properly with mmap() supportJustine Tunney
2023-03-30Make loading weights 10-100x fasterJustine Tunney
This is a breaking change that's going to give you three benefits: 1. Your inference commands should load 100x faster 2. You may be able to safely load models 2x larger 3. You can run many concurrent inference processes This was accomplished by changing the file format so we can mmap() weights directly into memory without having to read() or copy them thereby ensuring the kernel can make its file cache pages directly accessible to our inference processes; and secondly, that the file cache pages are much less likely to get evicted (which would force loads to hit disk) because they're no longer competing with memory pages that were needlessly created by gigabytes of standard i/o. The new file format supports single-file models like LLaMA 7b, and it also supports multi-file models like LLaMA 13B. Our Python tool now merges the foo.1, foo.2, etc. files back into a single file so that the C++ code which maps it doesn't need to reshape data every time. That's made llama.cpp so much simpler. Much of its load code has now been deleted. Furthermore, this change ensures that tensors are aligned properly on a 32-byte boundary. That opens the door to seeing if we can get additional performance gains on some microprocessors, by using ops that require memory alignment. Lastly note that both POSIX and the Windows platform are supported Fixes #91
2023-03-30Initial windows support (untested)Slaren
2023-03-30Always initialize mm_addr and mm_length in llama_modelSlaren
2023-03-30Unmap the file in llama_freeSlaren
2023-03-30Make mmap_file staticSlaren
2023-03-30Fix ggml_init_params in quantizeSlaren
2023-03-30Add mmap support for model filesSlaren
2023-03-29llama : fix compile warnings when reading the vocabGeorgi Gerganov
2023-03-29llama : use the same threshold for OpenBLAS and ggml thread limiting (#577)Maël Kerbiriou
2023-03-28py : add temporary script to convert old ggml files to newer version (#539)thement
Co-authored-by: Jakub Horak <jakub.horak@ibawizard.net>
2023-03-28all : be more strict about converting float to double (#458)Stephan Walter
* Be more strict about converting float to double * Test equivalence of round, SILU implementations Test module is commented out in CMakeLists.txt because the tests may take a long time, depending on how much the compiler optimizes. * Fix softmax in perplexity.cpp * all : prefer float over double where appropriate * perplexity : add <cmath> --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-03-28ggml : introduce structs for the q4 data blocks (#356)Stephan Walter
* Introduce structs for the q4 data blocks * ggml : rename quant struct variables + fix ARM_NEON --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-03-25Cleanup STL headers + fix embedding examples + minor stuffGeorgi Gerganov
2023-03-25Don't interefe with BLAS for large prompts by running only 1 threadGeorgi Gerganov
2023-03-25Add timings for the prompt evaluation (#478)slaren
2023-03-25Fix nasty bug in ggml_compute_forward_mul_mat_f32() and reenable BLASGeorgi Gerganov
2023-03-25Add support for file load progress reporting callbacks (#434)Jed Fox
* File load progress reporting * Move llama_progress_handler into llama_context_params * Renames * Use seekg to find file size instead * More correct load progress * Call progress callback more frequently * Fix typo
2023-03-25Fix crash for 65B model with pre-allocated memory (#485)Chris Kuehl
2023-03-24Reduce memory usage and allocate enough memory for largest context (#473)Georgi Gerganov
* Reduce memory usage and allocate enough memory for large contexts * Simpler scratch buffer usage * Reenable BLAS for quantized mul_mat * Fix number of layers in 30B and 65B * Fix KV cache size for F32
2023-03-24Temporary bump the memory buffer size - hopefully fix issues from 483bab2eGeorgi Gerganov
2023-03-24Properly free llama_context on failureGeorgi Gerganov
2023-03-24Support calling mlock() on loaded model data on Linux and macOS (#453)comex
* Support calling mlock() on loaded model data on Linux and macOS This is enabled by a new --mlock command line option. Using mlock() disables swapping and memory compression for the model data. Doing so can be useful on systems where the model takes up a large fraction of system RAM. In my experience, macOS is quite eager to start compressing llama.cpp's memory, which then makes it halt for a few seconds while it decompresses, even with a model that uses "only" 25GB out of 32GB. Of course, this comes at the cost of forcing the system to swap or compress other processes' memory instead, so it needs to be used with care and shouldn't be enabled by default. In theory it should be possible to support this on Windows as well using VirtualLock(), but I'm not much of a Windows user. * Update llama.cpp --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-03-24Add embedding mode with arg flag. Currently working (#282)Luciano
* working but ugly * add arg flag, not working on embedding mode * typo * Working! Thanks to @nullhook * make params argument instead of hardcoded boolean. remove useless time check * start doing the instructions but not finished. This probably doesnt compile * Embeddings extraction support --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-03-24Revert "Fix memory allocation issues and seg faults"Georgi Gerganov
This reverts commit 4870e455b3653f7d7769fa5772b2c90ffad088df. Will provide the correct fix later
2023-03-24Fix memory allocation issues and seg faultsGeorgi Gerganov
2023-03-23Avoid the transposed X branch in the Z = X * Y matrix multiplication (#439)Georgi Gerganov
Should make results reproducible for different number of threads and batch sizes
2023-03-22Add missing header for memcpy (#386)Yusuf Kağan Hanoğlu
fixed: memcpy is not defined
2023-03-22Init llama_context_params properly from CLI (#370)Georgi Gerganov
2023-03-22Introduce C-style API (#370)Georgi Gerganov
* Major refactoring - introduce C-style API * Clean up * Add <cassert> * Add <iterator> * Add <algorithm> .... * Fix timing reporting and accumulation * Measure eval time only for single-token calls * Change llama_tokenize return meaning