vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Merge remote-tracking branch 'upstream/master' into cached-logits-bandaid

Browse source
This commit is contained in:
Danny Daemonic 2023-05-29 02:57:57 -07:00
commit 44c83c6eba
15 changed files with 218 additions and 81 deletions
Download patch file Download diff file Expand all files Collapse all files

2
.devops/full.Dockerfile
View file

@ -3,7 +3,7 @@ ARG UBUNTU_VERSION=22.04
FROM ubuntu:$UBUNTU_VERSION as build
RUN apt-get update && \
apt-get install -y build-essential python3 python3-pip
apt-get install -y build-essential python3 python3-pip git
COPY requirements.txt requirements.txt

2
.devops/main.Dockerfile
View file

@ -3,7 +3,7 @@ ARG UBUNTU_VERSION=22.04
FROM ubuntu:$UBUNTU_VERSION as build
RUN apt-get update && \
apt-get install -y build-essential
apt-get install -y build-essential git
WORKDIR /app

20
.github/workflows/build.yml vendored
View file

@ -10,10 +10,10 @@ on:
push:
branches:
- master
paths: ['.github/workflows/**', '**/CMakeLists.txt', '**/Makefile', '**/*.h', '**/*.c', '**/*.cpp']
paths: ['.github/workflows/**', '**/CMakeLists.txt', '**/Makefile', '**/*.h', '**/*.hpp', '**/*.c', '**/*.cpp']
pull_request:
types: [opened, synchronize, reopened]
paths: ['**/CMakeLists.txt', '**/Makefile', '**/*.h', '**/*.c', '**/*.cpp']
paths: ['**/CMakeLists.txt', '**/Makefile', '**/*.h', '**/*.hpp', '**/*.c', '**/*.cpp']
env:
BRANCH_NAME: ${{ github.head_ref || github.ref_name }}
@ -157,15 +157,15 @@ jobs:
matrix:
include:
- build: 'avx2'
defines: ''
defines: '-DLLAMA_BUILD_SERVER=ON'
- build: 'avx'
defines: '-DLLAMA_AVX2=OFF'
defines: '-DLLAMA_BUILD_SERVER=ON -DLLAMA_AVX2=OFF'
- build: 'avx512'
defines: '-DLLAMA_AVX512=ON -DBUILD_SHARED_LIBS=ON'
defines: '-DLLAMA_BUILD_SERVER=ON -DLLAMA_AVX512=ON -DBUILD_SHARED_LIBS=ON'
- build: 'clblast'
defines: '-DLLAMA_CLBLAST=ON -DCMAKE_PREFIX_PATH="$env:RUNNER_TEMP/clblast"'
defines: '-DLLAMA_BUILD_SERVER=ON -DLLAMA_CLBLAST=ON -DCMAKE_PREFIX_PATH="$env:RUNNER_TEMP/clblast"'
- build: 'openblas'
defines: '-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS -DBLAS_INCLUDE_DIRS="$env:RUNNER_TEMP/openblas/include"'
defines: '-DLLAMA_BUILD_SERVER=ON -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS -DBLAS_INCLUDE_DIRS="$env:RUNNER_TEMP/openblas/include" -DBLAS_LIBRARIES="$env:RUNNER_TEMP/openblas/lib/openblas.lib"'
steps:
- name: Clone
@ -187,7 +187,7 @@ jobs:
curl.exe -o $env:RUNNER_TEMP/clblast.7z -L "https://github.com/CNugteren/CLBlast/releases/download/${env:CLBLAST_VERSION}/CLBlast-${env:CLBLAST_VERSION}-windows-x64.7z"
curl.exe -o $env:RUNNER_TEMP/CLBlast.LICENSE.txt -L "https://github.com/CNugteren/CLBlast/raw/${env:CLBLAST_VERSION}/LICENSE"
7z x "-o${env:RUNNER_TEMP}" $env:RUNNER_TEMP/clblast.7z
rename-item $env:RUNNER_TEMP/clblast_release_dir clblast
rename-item $env:RUNNER_TEMP/CLBlast-${env:CLBLAST_VERSION}-windows-x64 clblast
foreach ($f in (gci -Recurse -Path "$env:RUNNER_TEMP/clblast" -Filter '*.cmake')) {
$txt = Get-Content -Path $f -Raw
$txt.Replace('C:/vcpkg/packages/opencl_x64-windows/', "$($env:RUNNER_TEMP.Replace('\','/'))/opencl/") | Set-Content -Path $f -Encoding UTF8
@ -213,7 +213,6 @@ jobs:
cd build
cmake .. ${{ matrix.defines }}
cmake --build . --config Release
cp ../LICENSE ./bin/Release/llama.cpp.txt
- name: Add clblast.dll
id: add_clblast_dll
@ -258,6 +257,7 @@ jobs:
id: pack_artifacts
if: ${{ ( github.event_name == 'push' && github.ref == 'refs/heads/master' ) || github.event.inputs.create_release == 'true' }}
run: |
Copy-Item LICENSE .\build\bin\Release\llama.cpp.txt
7z a llama-${{ env.BRANCH_NAME }}-${{ steps.commit.outputs.short }}-bin-win-${{ matrix.build }}-x64.zip .\build\bin\Release\*
- name: Upload artifacts
@ -292,7 +292,7 @@ jobs:
run: |
mkdir build
cd build
cmake .. -DLLAMA_CUBLAS=ON
cmake .. -DLLAMA_BUILD_SERVER=ON -DLLAMA_CUBLAS=ON
cmake --build . --config Release
- name: Get commit hash

2
CMakeLists.txt
View file

@ -66,7 +66,7 @@ endif()
# 3rd party libs
option(LLAMA_ACCELERATE "llama: enable Accelerate framework" ON)
option(LLAMA_BLAS "llama: use BLAS" OFF)
option(LLAMA_BLAS_VENDOR "llama: BLA_VENDOR from https://cmake.org/cmake/help/latest/module/FindBLAS.html#blas-lapack-vendors" Generic)
set(LLAMA_BLAS_VENDOR "Generic" CACHE STRING "llama: BLAS library vendor")
option(LLAMA_CUBLAS "llama: use cuBLAS" OFF)
set(LLAMA_CUDA_DMMV_X "32" CACHE STRING "llama: x stride for dmmv CUDA kernels")
set(LLAMA_CUDA_DMMV_Y "1" CACHE STRING "llama: y block size for dmmv CUDA kernels")

19
Makefile
View file

@ -1,5 +1,11 @@
# Define the default target now so that it is always the first target
default: main quantize quantize-stats perplexity embedding vdot
BUILD_TARGETS = main quantize quantize-stats perplexity embedding vdot
ifdef LLAMA_BUILD_SERVER
BUILD_TARGETS += server
endif
default: $(BUILD_TARGETS)
ifndef UNAME_S
UNAME_S := $(shell uname -s)
@ -38,7 +44,11 @@ CFLAGS = -I. -O3 -std=c11 -fPIC
CXXFLAGS = -I. -I./examples -O3 -std=c++11 -fPIC
LDFLAGS =
ifndef LLAMA_DEBUG
ifdef LLAMA_DEBUG
CFLAGS += -O0 -g
CXXFLAGS += -O0 -g
LDFLAGS += -g
else
CFLAGS += -DNDEBUG
CXXFLAGS += -DNDEBUG
endif
@ -210,7 +220,7 @@ libllama.so: llama.o ggml.o $(OBJS)
$(CXX) $(CXXFLAGS) -shared -fPIC -o $@ $^ $(LDFLAGS)
clean:
rm -vf *.o main quantize quantize-stats perplexity embedding benchmark-matmult save-load-state build-info.h
rm -vf *.o main quantize quantize-stats perplexity embedding benchmark-matmult save-load-state server vdot build-info.h
#
# Examples
@ -237,6 +247,9 @@ embedding: examples/embedding/embedding.cpp build-info.h ggml.o llama.o common.o
save-load-state: examples/save-load-state/save-load-state.cpp build-info.h ggml.o llama.o common.o $(OBJS)
$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
server: examples/server/server.cpp examples/server/httplib.h examples/server/json.hpp build-info.h ggml.o llama.o common.o $(OBJS)
$(CXX) $(CXXFLAGS) -Iexamples/server $(filter-out %.h,$(filter-out %.hpp,$^)) -o $@ $(LDFLAGS)
build-info.h: $(wildcard .git/index) scripts/build-info.sh
@sh scripts/build-info.sh > $@.tmp
@if ! cmp -s $@.tmp $@; then \

85
README.md
View file

@ -240,11 +240,11 @@ In order to build llama.cpp you have three different options.
Building the program with BLAS support may lead to some performance improvements in prompt processing using batch sizes higher than 32 (the default is 512). BLAS doesn't affect the normal generation performance. There are currently three different implementations of it:
- Accelerate Framework:
- **Accelerate Framework**:
This is only available on Mac PCs and it's enabled by default. You can just build using the normal instructions.
- OpenBLAS:
- **OpenBLAS**:
This provides BLAS acceleration using only the CPU. Make sure to have OpenBLAS installed on your machine.
@ -278,11 +278,11 @@ Building the program with BLAS support may lead to some performance improvements
cmake --build . --config Release
```
- BLIS
- **BLIS**
Check [BLIS.md](BLIS.md) for more information.
- Intel MKL
- **Intel MKL**
By default, `LLAMA_BLAS_VENDOR` is set to `Generic`, so if you already sourced intel environment script and assign `-DLLAMA_BLAS=ON` in cmake, the mkl version of Blas will automatically been selected. You may also specify it by:
@ -293,7 +293,7 @@ Building the program with BLAS support may lead to some performance improvements
cmake --build . -config Release
```
- cuBLAS
- **cuBLAS**
This provides BLAS acceleration using the CUDA cores of your Nvidia GPU. Make sure to have the CUDA toolkit installed. You can download it from your Linux distro's package manager or from here: [CUDA Toolkit](https://developer.nvidia.com/cuda-downloads).
- Using `make`:
@ -308,8 +308,81 @@ Building the program with BLAS support may lead to some performance improvements
cmake .. -DLLAMA_CUBLAS=ON
cmake --build . --config Release
```
Note: Because llama.cpp uses multiple CUDA streams for matrix multiplication results [are not guaranteed to be reproducible](https://docs.nvidia.com/cuda/cublas/index.html#results-reproducibility). If you need reproducibility, set `GGML_CUDA_MAX_STREAMS` in the file `ggml-cuda.cu` to 1.
Note: Because llama.cpp uses multiple CUDA streams for matrix multiplication results [are not guaranteed to be reproducible](https://docs.nvidia.com/cuda/cublas/index.html#results-reproducibility). If you need reproducibility, set `GGML_CUDA_MAX_STREAMS` in the file `ggml-cuda.cu` to 1.
- **CLBlast**
OpenCL acceleration is provided by the matrix multiplication kernels from the [CLBlast](https://github.com/CNugteren/CLBlast) project and custom kernels for ggml that can generate tokens on the GPU.
You will need the [OpenCL SDK](https://github.com/KhronosGroup/OpenCL-SDK).
- For Ubuntu or Debian, the packages `opencl-headers`, `ocl-icd` may be needed.
- <details>
<summary>Installing the OpenCL SDK from source</summary>
```sh
git clone --recurse-submodules https://github.com/KhronosGroup/OpenCL-SDK.git
mkdir OpenCL-SDK/build
cd OpenCL-SDK/build
cmake .. -DBUILD_DOCS=OFF \
-DBUILD_EXAMPLES=OFF \
-DBUILD_TESTING=OFF \
-DOPENCL_SDK_BUILD_SAMPLES=OFF \
-DOPENCL_SDK_TEST_SAMPLES=OFF
cmake --build . --config Release
cmake --install . --prefix /some/path
```
</details>
Installing CLBlast: it may be found in your operating system's packages.
- <details>
<summary>If not, then installing from source:</summary>
```sh
git clone https://github.com/CNugteren/CLBlast.git
mkdir CLBlast/build
cd CLBLast/build
cmake .. -DBUILD_SHARED_LIBS=OFF -DTUNERS=OFF
cmake --build . --config Release
cmake --install . --prefix /some/path
```
Where `/some/path` is where the built library will be installed (default is `/usr/loca`l`).
</details>
Building:
- Build with make:
```sh
make LLAMA_CLBLAST=1
```
- CMake:
```sh
mkdir build
cd build
cmake .. -DLLAMA_CLBLAST=ON -DCLBlast_dir=/some/path
cmake --build . --config Release
```
Running:
The CLBlast build supports `--gpu-layers|-ngl` like the CUDA version does.
To select the correct platform (driver) and device (GPU), you can use the environment variables `GGML_OPENCL_PLATFORM` and `GGML_OPENCL_DEVICE`.
The selection can be a number (starting from 0) or a text string to search:
```sh
GGML_OPENCL_PLATFORM=1 ./main ...
GGML_OPENCL_DEVICE=2 ./main ...
GGML_OPENCL_PLATFORM=Intel ./main ...
GGML_OPENCL_PLATFORM=AMD GGML_OPENCL_DEVICE=1 ./main ...
```
The default behavior is to find the first GPU device, but when it is an integrated GPU on a laptop, for instance, the selectors are useful.
Using the variables it is possible to select a CPU-based driver as well, if so desired.
You can get a list of platforms and devices from the `clinfo -l` command, etc.
### Prepare Data & Run

16
examples/common.cpp
View file

@ -251,6 +251,12 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
break;
}
params.model = argv[i];
} else if (arg == "-a" || arg == "--alias") {
if (++i >= argc) {
invalid_param = true;
break;
}
params.model_alias = argv[i];
} else if (arg == "--lora") {
if (++i >= argc) {
invalid_param = true;
@ -283,7 +289,12 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
invalid_param = true;
break;
}
#ifdef LLAMA_SUPPORTS_GPU_OFFLOAD
params.n_gpu_layers = std::stoi(argv[i]);
#else
fprintf(stderr, "warning: not compiled with GPU offload support, --n-gpu-layers option will be ignored\n");
fprintf(stderr, "warning: see main README.md for information on enabling GPU BLAS support\n");
#endif
} else if (arg == "--no-mmap") {
params.use_mmap = false;
} else if (arg == "--mtest") {
@ -410,7 +421,8 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
fprintf(stderr, " -c N, --ctx-size N size of the prompt context (default: %d)\n", params.n_ctx);
fprintf(stderr, " --ignore-eos ignore end of stream token and continue generating (implies --logit-bias 2-inf)\n");
fprintf(stderr, " --no-penalize-nl do not penalize newline token\n");
fprintf(stderr, " --memory-f32 use f32 instead of f16 for memory key+value\n");
fprintf(stderr, " --memory-f32 use f32 instead of f16 for memory key+value (default: disabled)\n");
fprintf(stderr, " not recommended: doubles context memory required and no measurable increase in quality\n");
fprintf(stderr, " --temp N temperature (default: %.1f)\n", (double)params.temp);
fprintf(stderr, " -b N, --batch-size N batch size for prompt processing (default: %d)\n", params.n_batch);
fprintf(stderr, " --perplexity compute perplexity over the prompt\n");
@ -421,8 +433,10 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
if (llama_mmap_supported()) {
fprintf(stderr, " --no-mmap do not memory-map model (slower load but may reduce pageouts if not using mlock)\n");
}
#ifdef LLAMA_SUPPORTS_GPU_OFFLOAD
fprintf(stderr, " -ngl N, --n-gpu-layers N\n");
fprintf(stderr, " number of layers to store in VRAM\n");
#endif
fprintf(stderr, " --mtest compute maximum memory usage\n");
fprintf(stderr, " --verbose-prompt print prompt before generation\n");
fprintf(stderr, " --lora FNAME apply LoRA adapter (implies --no-mmap)\n");

1
examples/common.h
View file

@ -45,6 +45,7 @@ struct gpt_params {
float mirostat_eta = 0.10f; // learning rate
std::string model = "models/7B/ggml-model.bin"; // model path
std::string model_alias = "unknown"; // model alias
std::string prompt = "";
std::string path_prompt_cache = ""; // path to file for saving/loading prompt eval state
std::string input_prefix = ""; // string to prefix user inputs with

53
examples/main/README.md
View file

@ -69,8 +69,8 @@ In this section, we cover the most commonly used options for running the `main`
- `-m FNAME, --model FNAME`: Specify the path to the LLaMA model file (e.g., `models/7B/ggml-model.bin`).
- `-i, --interactive`: Run the program in interactive mode, allowing you to provide input directly and receive real-time responses.
- `-ins, --instruct`: Run the program in instruction mode, which is particularly useful when working with Alpaca models.
- `-n N, --n_predict N`: Set the number of tokens to predict when generating text. Adjusting this value can influence the length of the generated text.
- `-c N, --ctx_size N`: Set the size of the prompt context. The default is 512, but LLaMA models were built with a context of 2048, which will provide better results for longer input/inference.
- `-n N, --n-predict N`: Set the number of tokens to predict when generating text. Adjusting this value can influence the length of the generated text.
- `-c N, --ctx-size N`: Set the size of the prompt context. The default is 512, but LLaMA models were built with a context of 2048, which will provide better results for longer input/inference.
## Input Prompts
@ -136,9 +136,9 @@ During text generation, LLaMA models have a limited context size, which means th
### Context Size
The `--ctx_size` option allows you to set the size of the prompt context used by the LLaMA models during text generation. A larger context size helps the model to better comprehend and generate responses for longer input or conversations.
The `--ctx-size` option allows you to set the size of the prompt context used by the LLaMA models during text generation. A larger context size helps the model to better comprehend and generate responses for longer input or conversations.
- `-c N, --ctx_size N`: Set the size of the prompt context (default: 512). The LLaMA models were built with a context of 2048, which will yield the best results on longer input/inference. However, increasing the context size beyond 2048 may lead to unpredictable results.
- `-c N, --ctx-size N`: Set the size of the prompt context (default: 512). The LLaMA models were built with a context of 2048, which will yield the best results on longer input/inference. However, increasing the context size beyond 2048 may lead to unpredictable results.
### Keep Prompt
@ -146,7 +146,7 @@ The `--keep` option allows users to retain the original prompt when the model ru
- `--keep N`: Specify the number of tokens from the initial prompt to retain when the model resets its internal context. By default, this value is set to 0 (meaning no tokens are kept). Use `-1` to retain all tokens from the initial prompt.
By utilizing context management options like `--ctx_size` and `--keep`, you can maintain a more coherent and consistent interaction with the LLaMA models, ensuring that the generated text remains relevant to the original prompt or conversation.
By utilizing context management options like `--ctx-size` and `--keep`, you can maintain a more coherent and consistent interaction with the LLaMA models, ensuring that the generated text remains relevant to the original prompt or conversation.
## Generation Flags
@ -154,11 +154,11 @@ The following options allow you to control the text generation process and fine-
### Number of Tokens to Predict
- `-n N, --n_predict N`: Set the number of tokens to predict when generating text (default: 128, -1 = infinity).
- `-n N, --n-predict N`: Set the number of tokens to predict when generating text (default: 128, -1 = infinity).
The `--n_predict` option controls the number of tokens the model generates in response to the input prompt. By adjusting this value, you can influence the length of the generated text. A higher value will result in longer text, while a lower value will produce shorter text. A value of -1 will cause text to be generated without limit.
The `--n-predict` option controls the number of tokens the model generates in response to the input prompt. By adjusting this value, you can influence the length of the generated text. A higher value will result in longer text, while a lower value will produce shorter text. A value of -1 will cause text to be generated without limit.
It is important to note that the generated text may be shorter than the specified number of tokens if an End-of-Sequence (EOS) token or a reverse prompt is encountered. In interactive mode text generation will pause and control will be returned to the user. In non-interactive mode, the program will end. In both cases, the text generation may stop before reaching the specified `n_predict` value. If you want the model to keep going without ever producing End-of-Sequence on its own, you can use the `--ignore-eos` parameter.
It is important to note that the generated text may be shorter than the specified number of tokens if an End-of-Sequence (EOS) token or a reverse prompt is encountered. In interactive mode text generation will pause and control will be returned to the user. In non-interactive mode, the program will end. In both cases, the text generation may stop before reaching the specified `n-predict` value. If you want the model to keep going without ever producing End-of-Sequence on its own, you can use the `--ignore-eos` parameter.
### Temperature
@ -170,33 +170,33 @@ Example usage: `--temp 0.5`
### Repeat Penalty
- `--repeat_penalty N`: Control the repetition of token sequences in the generated text (default: 1.1).
- `--repeat_last_n N`: Last n tokens to consider for penalizing repetition (default: 64, 0 = disabled, -1 = ctx_size).
- `--repeat-penalty N`: Control the repetition of token sequences in the generated text (default: 1.1).
- `--repeat-last-n N`: Last n tokens to consider for penalizing repetition (default: 64, 0 = disabled, -1 = ctx-size).
- `--no-penalize-nl`: Disable penalization for newline tokens when applying the repeat penalty.
The `repeat_penalty` option helps prevent the model from generating repetitive or monotonous text. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. The default value is 1.1.
The `repeat-penalty` option helps prevent the model from generating repetitive or monotonous text. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. The default value is 1.1.
The `repeat_last_n` option controls the number of tokens in the history to consider for penalizing repetition. A larger value will look further back in the generated text to prevent repetitions, while a smaller value will only consider recent tokens. A value of 0 disables the penalty, and a value of -1 sets the number of tokens considered equal to the context size (`ctx_size`).
The `repeat-last-n` option controls the number of tokens in the history to consider for penalizing repetition. A larger value will look further back in the generated text to prevent repetitions, while a smaller value will only consider recent tokens. A value of 0 disables the penalty, and a value of -1 sets the number of tokens considered equal to the context size (`ctx-size`).
Use the `--no-penalize-nl` option to disable newline penalization when applying the repeat penalty. This option is particularly useful for generating chat conversations, dialogues, code, poetry, or any text where newline tokens play a significant role in structure and formatting. Disabling newline penalization helps maintain the natural flow and intended formatting in these specific use cases.
Example usage: `--repeat_penalty 1.15 --repeat_last_n 128 --no-penalize-nl`
Example usage: `--repeat-penalty 1.15 --repeat-last-n 128 --no-penalize-nl`
### Top-K Sampling
- `--top_k N`: Limit the next token selection to the K most probable tokens (default: 40).
- `--top-k N`: Limit the next token selection to the K most probable tokens (default: 40).
Top-k sampling is a text generation method that selects the next token only from the top k most likely tokens predicted by the model. It helps reduce the risk of generating low-probability or nonsensical tokens, but it may also limit the diversity of the output. A higher value for top_k (e.g., 100) will consider more tokens and lead to more diverse text, while a lower value (e.g., 10) will focus on the most probable tokens and generate more conservative text. The default value is 40.
Top-k sampling is a text generation method that selects the next token only from the top k most likely tokens predicted by the model. It helps reduce the risk of generating low-probability or nonsensical tokens, but it may also limit the diversity of the output. A higher value for top-k (e.g., 100) will consider more tokens and lead to more diverse text, while a lower value (e.g., 10) will focus on the most probable tokens and generate more conservative text. The default value is 40.
Example usage: `--top_k 30`
Example usage: `--top-k 30`
### Top-P Sampling
- `--top_p N`: Limit the next token selection to a subset of tokens with a cumulative probability above a threshold P (default: 0.9).
- `--top-p N`: Limit the next token selection to a subset of tokens with a cumulative probability above a threshold P (default: 0.9).
Top-p sampling, also known as nucleus sampling, is another text generation method that selects the next token from a subset of tokens that together have a cumulative probability of at least p. This method provides a balance between diversity and quality by considering both the probabilities of tokens and the number of tokens to sample from. A higher value for top_p (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. The default value is 0.9.
Top-p sampling, also known as nucleus sampling, is another text generation method that selects the next token from a subset of tokens that together have a cumulative probability of at least p. This method provides a balance between diversity and quality by considering both the probabilities of tokens and the number of tokens to sample from. A higher value for top-p (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. The default value is 0.9.
Example usage: `--top_p 0.95`
Example usage: `--top-p 0.95`
### Tail Free Sampling (TFS)
@ -217,16 +217,16 @@ Example usage: `--typical 0.9`
### Mirostat Sampling
- `--mirostat N`: Enable Mirostat sampling, controlling perplexity during text generation (default: 0, 0 = disabled, 1 = Mirostat, 2 = Mirostat 2.0).
- `--mirostat_lr N`: Set the Mirostat learning rate, parameter eta (default: 0.1).
- `--mirostat_ent N`: Set the Mirostat target entropy, parameter tau (default: 5.0).
- `--mirostat-lr N`: Set the Mirostat learning rate, parameter eta (default: 0.1).
- `--mirostat-ent N`: Set the Mirostat target entropy, parameter tau (default: 5.0).
Mirostat is an algorithm that actively maintains the quality of generated text within a desired range during text generation. It aims to strike a balance between coherence and diversity, avoiding low-quality output caused by excessive repetition (boredom traps) or incoherence (confusion traps).
The `--mirostat_lr` option sets the Mirostat learning rate (eta). The learning rate influences how quickly the algorithm responds to feedback from the generated text. A lower learning rate will result in slower adjustments, while a higher learning rate will make the algorithm more responsive. The default value is `0.1`.
The `--mirostat-lr` option sets the Mirostat learning rate (eta). The learning rate influences how quickly the algorithm responds to feedback from the generated text. A lower learning rate will result in slower adjustments, while a higher learning rate will make the algorithm more responsive. The default value is `0.1`.
The `--mirostat_ent` option sets the Mirostat target entropy (tau), which represents the desired perplexity value for the generated text. Adjusting the target entropy allows you to control the balance between coherence and diversity in the generated text. A lower value will result in more focused and coherent text, while a higher value will lead to more diverse and potentially less coherent text. The default value is `5.0`.
The `--mirostat-ent` option sets the Mirostat target entropy (tau), which represents the desired perplexity value for the generated text. Adjusting the target entropy allows you to control the balance between coherence and diversity in the generated text. A lower value will result in more focused and coherent text, while a higher value will lead to more diverse and potentially less coherent text. The default value is `5.0`.
Example usage: `--mirostat 2 --mirostat_lr 0.05 --mirostat_ent 3.0`
Example usage: `--mirostat 2 --mirostat-lr 0.05 --mirostat-ent 3.0`
### Logit Bias
@ -264,11 +264,11 @@ These options help improve the performance and memory usage of the LLaMA models.
### Memory Float 32
- `--memory_f32`: Use 32-bit floats instead of 16-bit floats for memory key+value, allowing higher quality inference at the cost of higher memory usage.
- `--memory-f32`: Use 32-bit floats instead of 16-bit floats for memory key+value. This doubles the context memory requirement and cached prompt file size but does not appear to increase generation quality in a measurable way. Not recommended.
### Batch Size
- `-b N, --batch_size N`: Set the batch size for prompt processing (default: 512). This large batch size benefits users who have BLAS installed and enabled it during the build. If you don't have BLAS enabled ("BLAS=0"), you can use a smaller number, such as 8, to see the prompt progress as it's evaluated in some situations.
- `-b N, --batch-size N`: Set the batch size for prompt processing (default: 512). This large batch size benefits users who have BLAS installed and enabled it during the build. If you don't have BLAS enabled ("BLAS=0"), you can use a smaller number, such as 8, to see the prompt progress as it's evaluated in some situations.
### Prompt Caching
@ -285,5 +285,6 @@ These options provide extra functionality and customization when running the LLa
- `-h, --help`: Display a help message showing all available options and their default values. This is particularly useful for checking the latest options and default values, as they can change frequently, and the information in this document may become outdated.
- `--verbose-prompt`: Print the prompt before generating text.
- `--mtest`: Test the model's functionality by running a series of tests to ensure it's working properly.
- `-ngl N, --n-gpu-layers N`: When compiled with appropriate support (currently CLBlast or cuBLAS), this option allows offloading some layers to the GPU for computation. Generally results in increased performance.
- `--lora FNAME`: Apply a LoRA (Low-Rank Adaptation) adapter to the model (implies --no-mmap). This allows you to adapt the pretrained model to specific tasks or domains.
- `--lora-base FNAME`: Optional model to use as a base for the layers modified by the LoRA adapter. This flag is used in conjunction with the `--lora` flag, and specifies the base model for the adaptation.

5
examples/server/README.md
View file

@ -285,7 +285,8 @@ Test();
## Common Options
- `-m FNAME, --model FNAME`: Specify the path to the LLaMA model file (e.g., `models/7B/ggml-model.bin`).
- `-c N, --ctx_size N`: Set the size of the prompt context. The default is 512, but LLaMA models were built with a context of 2048, which will provide better results for longer input/inference.
- `-c N, --ctx-size N`: Set the size of the prompt context. The default is 512, but LLaMA models were built with a context of 2048, which will provide better results for longer input/inference.
- `-ngl N, --n-gpu-layers N`: When compiled with appropriate support (currently CLBlast or cuBLAS), this option allows offloading some layers to the GPU for computation. Generally results in increased performance.
- `--embedding`: Enable the embedding mode. **Completion function doesn't work in this mode**.
- `--host`: Set the hostname or ip address to listen. Default `127.0.0.1`;
- `--port`: Set the port to listen. Default: `8080`.
@ -304,7 +305,7 @@ The RNG seed is used to initialize the random number generator that influences t
### Memory Float 32
- `--memory_f32`: Use 32-bit floats instead of 16-bit floats for memory key+value, allowing higher quality inference at the cost of higher memory usage.
- `--memory-f32`: Use 32-bit floats instead of 16-bit floats for memory key+value. This doubles the context memory requirement but does not appear to increase generation quality in a measurable way. Not recommended.
## Limitations:

47
examples/server/server.cpp
View file

@ -61,7 +61,7 @@ struct llama_server_context
std::vector<llama_token> prompt_tokens = ::llama_tokenize(ctx, params.prompt, true);
// compare the evaluated prompt with the new prompt
int new_prompt_len = 0;
for (int i = 0;i < prompt_tokens.size(); i++) {
for (size_t i = 0; i < prompt_tokens.size(); i++) {
if (i < processed_tokens.size() &&
processed_tokens[i] == prompt_tokens[i])
{
@ -71,7 +71,7 @@ struct llama_server_context
{
embd_inp.push_back(prompt_tokens[i]);
if(new_prompt_len == 0) {
if(i - 1 < n_past) {
if(int32_t(i) - 1 < n_past) {
processed_tokens.erase(processed_tokens.begin() + i, processed_tokens.end());
}
// Evaluate the new fragment prompt from the last token processed.
@ -136,7 +136,7 @@ struct llama_server_context
{
// out of user input, sample next token
const float temp = params.temp;
const int32_t top_k = params.top_k <= 0 ? llama_n_vocab(ctx) : params.top_k;
// const int32_t top_k = params.top_k <= 0 ? llama_n_vocab(ctx) : params.top_k;
const float top_p = params.top_p;
const float tfs_z = params.tfs_z;
const float typical_p = params.typical_p;
@ -306,12 +306,12 @@ struct llama_server_context
// Avoid add the no show words to the response
for (std::vector<llama_token> word_tokens : no_show_words)
{
int match_token = 1;
size_t match_token = 1;
if (tokens_predicted.front() == word_tokens.front())
{
bool execute_matching = true;
if (tokens_predicted.size() > 1) { // if previus tokens had been tested
for (int i = 1; i < word_tokens.size(); i++)
for (size_t i = 1; i < word_tokens.size(); i++)
{
if (i >= tokens_predicted.size()) {
match_token = i;
@ -385,7 +385,9 @@ void server_print_usage(int /*argc*/, char **argv, const gpt_params &params)
fprintf(stderr, "options:\n");
fprintf(stderr, " -h, --help show this help message and exit\n");
fprintf(stderr, " -s SEED, --seed SEED RNG seed (default: -1, use random seed for < 0)\n");
fprintf(stderr, " --memory_f32 use f32 instead of f16 for memory key+value\n");
fprintf(stderr, " -c N, --ctx-size N size of the prompt context (default: %d)\n", params.n_ctx);
fprintf(stderr, " --memory-f32 use f32 instead of f16 for memory key+value (default: disabled)\n");
fprintf(stderr, " not recommended: doubles context memory required and no measurable increase in quality\n");
fprintf(stderr, " --embedding enable embedding mode\n");
fprintf(stderr, " --keep number of tokens to keep from the initial prompt (default: %d, -1 = all)\n", params.n_keep);
if (llama_mlock_supported())
@ -396,12 +398,16 @@ void server_print_usage(int /*argc*/, char **argv, const gpt_params &params)
{
fprintf(stderr, " --no-mmap do not memory-map model (slower load but may reduce pageouts if not using mlock)\n");
}
#ifdef LLAMA_SUPPORTS_GPU_OFFLOAD
fprintf(stderr, " -ngl N, --n-gpu-layers N\n");
fprintf(stderr, " number of layers to store in VRAM\n");
#endif
fprintf(stderr, " -m FNAME, --model FNAME\n");
fprintf(stderr, " model path (default: %s)\n", params.model.c_str());
fprintf(stderr, " -host ip address to listen (default 127.0.0.1)\n");
fprintf(stderr, " -port PORT port to listen (default 8080)\n");
fprintf(stderr, " -a ALIAS, --alias ALIAS\n");
fprintf(stderr, " set an alias for the model, will be added as `model` field in completion response\n");
fprintf(stderr, " --host ip address to listen (default 127.0.0.1)\n");
fprintf(stderr, " --port PORT port to listen (default 8080)\n");
fprintf(stderr, "\n");
}
@ -453,6 +459,15 @@ bool server_params_parse(int argc, char **argv, server_params &sparams, gpt_para
}
params.model = argv[i];
}
else if (arg == "-a" || arg == "--alias")
{
if (++i >= argc)
{
invalid_param = true;
break;
}
params.model_alias = argv[i];
}
else if (arg == "--embedding")
{
params.embedding = true;
@ -462,7 +477,7 @@ bool server_params_parse(int argc, char **argv, server_params &sparams, gpt_para
server_print_usage(argc, argv, default_params);
exit(0);
}
else if (arg == "-c" || arg == "--ctx_size")
else if (arg == "-c" || arg == "--ctx-size" || arg == "--ctx_size")
{
if (++i >= argc)
{
@ -471,7 +486,7 @@ bool server_params_parse(int argc, char **argv, server_params &sparams, gpt_para
}
params.n_ctx = std::stoi(argv[i]);
}
else if (arg == "--memory_f32")
else if (arg == "--memory-f32" || arg == "--memory_f32")
{
params.memory_f16 = false;
}
@ -482,7 +497,12 @@ bool server_params_parse(int argc, char **argv, server_params &sparams, gpt_para
invalid_param = true;
break;
}
#ifdef LLAMA_SUPPORTS_GPU_OFFLOAD
params.n_gpu_layers = std::stoi(argv[i]);
#else
fprintf(stderr, "warning: not compiled with GPU offload support, --n-gpu-layers option will be ignored\n");
fprintf(stderr, "warning: see main README.md for information on enabling GPU BLAS support\n");
#endif
}
else
{
@ -601,7 +621,7 @@ int main(int argc, char **argv)
Server svr;
svr.Get("/", [](const Request &req, Response &res)
svr.Get("/", [](const Request &, Response &res)
{ res.set_content("<h1>llama.cpp server works</h1>", "text/html"); });
svr.Post("/completion", [&llama](const Request &req, Response &res)
@ -645,11 +665,12 @@ int main(int argc, char **argv)
try
{
json data = {
{"model", llama.params.model_alias },
{"content", llama.generated_text },
{"tokens_predicted", llama.num_tokens_predicted}};
return res.set_content(data.dump(), "application/json");
}
catch (json::exception e)
catch (const json::exception &e)
{
// Some tokens have bad UTF-8 strings, the json parser is very sensitive
json data = {
@ -701,7 +722,7 @@ int main(int argc, char **argv)
{"content", result },
{"stop", !llama.has_next_token }};
return res.set_content(data.dump(), "application/json");
} catch (json::exception e) {
} catch (const json::exception &e) {
// Some tokens have bad UTF-8 strings, the json parser is very sensitive
json data = {
{"content", "" },

22
ggml-opencl.cpp
View file

@ -469,16 +469,11 @@ void ggml_cl_init(void) {
size_t ext_str_size;
clGetDeviceInfo(device, CL_DEVICE_EXTENSIONS, 0, NULL, &ext_str_size);
char* ext_buffer = (char*) malloc(sizeof(char) * ext_str_size);
char *ext_buffer = (char *)alloca(ext_str_size + 1);
clGetDeviceInfo(device, CL_DEVICE_EXTENSIONS, ext_str_size, ext_buffer, NULL);
ext_buffer[ext_str_size] = '\0'; // ensure it is null terminated
// Check if ext_buffer contains cl_khr_fp16
for (size_t i = 0; i < ext_str_size - 12; i++) {
if (memcmp(ext_buffer + i, "cl_khr_fp16", 11) == 0) {
fp16_support = true;
break;
}
}
free(ext_buffer);
fp16_support = strstr(ext_buffer, "cl_khr_fp16") != NULL;
fprintf(stderr, "ggml_opencl: device FP16 support: %s\n", fp16_support ? "true" : "false");
cl_context_properties properties[] = {
@ -672,7 +667,7 @@ static void ggml_cl_mul_mat_f32(const ggml_tensor * src0, const ggml_tensor * sr
size_t d_size;
cl_mem d_X;
if (src0->backend == GGML_BACKEND_CL) {
d_X = *(cl_mem*) src0->data;
d_X = (cl_mem) src0->data;
} else {
d_X = ggml_cl_pool_malloc(sizeof(ggml_fp16_t) * x_ne, &x_size, CL_MEM_READ_ONLY);
}
@ -748,7 +743,7 @@ static void ggml_cl_mul_mat_f16(const ggml_tensor * src0, const ggml_tensor * sr
size_t d_size;
cl_mem d_X;
if (src0->backend == GGML_BACKEND_CL) {
d_X = *(cl_mem*) src0->data;
d_X = (cl_mem) src0->data;
} else {
d_X = ggml_cl_pool_malloc(sizeof(ggml_fp16_t) * x_ne, &x_size, CL_MEM_READ_ONLY);
}
@ -873,7 +868,7 @@ static void ggml_cl_mul_mat_q_f32(const ggml_tensor * src0, const ggml_tensor *
if (src0->backend == GGML_BACKEND_CPU) {
CL_CHECK(ggml_cl_h2d_tensor_2d(queue, d_Q, 0, src0, i03, i02, NULL));
} else if (src0->backend == GGML_BACKEND_CL) {
d_Q = *(cl_mem*) src0->data;
d_Q = (cl_mem) src0->data;
} else {
GGML_ASSERT(false);
}
@ -1016,14 +1011,13 @@ void ggml_cl_transform_tensor(ggml_tensor * tensor) {
const size_t q_sz = ggml_type_size(type) * ne0 * ne1 * ne2 * ne3 / ggml_blck_size(type);
size_t q_size;
cl_mem* dst = (cl_mem*) malloc(sizeof(cl_mem));
*dst = ggml_cl_pool_malloc(q_sz, &q_size, CL_MEM_READ_ONLY);
cl_mem dst = ggml_cl_pool_malloc(q_sz, &q_size, CL_MEM_READ_ONLY);
// copy tensor to device
for (int64_t i3 = 0; i3 < ne3; i3++) {
for (int64_t i2 = 0; i2 < ne2; i2++) {
int i = i3*ne2 + i2;
CL_CHECK(ggml_cl_h2d_tensor_2d(queue, *dst, i*ne0*ne1, tensor, i3, i2, NULL));
CL_CHECK(ggml_cl_h2d_tensor_2d(queue, dst, i*ne0*ne1, tensor, i3, i2, NULL));
}
}

14
ggml.c
View file

@ -186,10 +186,12 @@ typedef double ggml_float;
#if defined(_MSC_VER) || defined(__MINGW32__)
#include <intrin.h>
#else
#if !defined(__riscv)
#include <immintrin.h>
#endif
#endif
#endif
#endif
#ifdef __F16C__
@ -3808,6 +3810,10 @@ enum ggml_type ggml_ftype_to_ggml_type(enum ggml_ftype ftype) {
return wtype;
}
size_t ggml_tensor_overhead(void) {
return GGML_OBJECT_SIZE + GGML_TENSOR_SIZE + 16;
}
static inline bool ggml_is_transposed(const struct ggml_tensor * tensor) {
return tensor->nb[0] > tensor->nb[1];
}
@ -14527,6 +14533,14 @@ void ggml_graph_reset(struct ggml_cgraph * cgraph) {
}
struct ggml_tensor * ggml_get_tensor_by_name(struct ggml_cgraph * cgraph, const char * name) {
for (int i = 0; i < cgraph->n_leafs; i++) {
struct ggml_tensor * leaf = cgraph->leafs[i];
if (strcmp(leaf->name, name) == 0) {
return leaf;
}
}
for (int i = 0; i < cgraph->n_nodes; i++) {
struct ggml_tensor * node = cgraph->nodes[i];

6
ggml.h
View file

@ -380,9 +380,6 @@ extern "C" {
static const size_t GGML_TENSOR_SIZE = sizeof(struct ggml_tensor);
// use this to compute the memory overhead of a tensor
static const size_t GGML_TENSOR_OVERHEAD = (GGML_OBJECT_SIZE + GGML_TENSOR_SIZE + 16);
// computation graph
struct ggml_cgraph {
int n_nodes;
@ -444,6 +441,9 @@ extern "C" {
// TODO: temporary until model loading of ggml examples is refactored
GGML_API enum ggml_type ggml_ftype_to_ggml_type(enum ggml_ftype ftype);
// use this to compute the memory overhead of a tensor
GGML_API size_t ggml_tensor_overhead(void);
// main
GGML_API struct ggml_context * ggml_init(struct ggml_init_params params);

5
llama.h
View file

@ -31,6 +31,11 @@
#define LLAMA_SESSION_MAGIC LLAMA_FILE_MAGIC_GGSN
#define LLAMA_SESSION_VERSION 1
#if defined(GGML_USE_CUBLAS) || defined(GGML_USE_CLBLAST)
// Defined when llama.cpp is compiled with support for offloading model layers to GPU.
#define LLAMA_SUPPORTS_GPU_OFFLOAD
#endif
#ifdef __cplusplus
extern "C" {
#endif