Getting Started - llama-cpp-python

Environment Variables# Linux and MacCMAKE_ARGS="-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS"\pipinstallllama-cpp-python # Windows$env:CMAKE_ARGS="-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS"pipinstallllama-cpp-pythonCLI / requirements.txt They can also be set via `pip install -C / --config-settings` command and saved to a `requirements.txt` file: pipinstall--upgradepip# ensure pip is up to datepipinstallllama-cpp-python\-Ccmake.args="-DGGML_BLAS=ON;-DGGML_BLAS_VENDOR=OpenBLAS"# requirements.txt llama-cpp-python -C cmake.args="-DGGML_BLAS=ON;-DGGML_BLAS_VENDOR=OpenBLAS" Supported Backends

Below are some common backends, their build commands and any additional environment variables required.

OpenBLAS (CPU) To install with OpenBLAS, set the `GGML_BLAS` and `GGML_BLAS_VENDOR` environment variables before installing: CMAKE_ARGS="-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS"pipinstallllama-cpp-python CUDA To install with CUDA support, set the `GGML_CUDA=on` environment variable before installing: CMAKE_ARGS="-DGGML_CUDA=on"pipinstallllama-cpp-python **Pre-built Wheel (New)** It is also possible to install a pre-built wheel with CUDA support. As long as your system meets some requirements: - CUDA Version is 11.8, 12.1, 12.2, 12.3, 12.4, 12.5, 13.0 or 13.2 - NVIDIA GPU compute capability is 6.0 through 8.9 for CUDA 11.8 wheels, 6.0 or newer for CUDA 12 wheels, or 7.5 or newer for CUDA 13 wheels - Python Version is 3.10, 3.11 or 3.12 pipinstallllama-cpp-python\--extra-index-urlhttps://abetlen.github.io/llama-cpp-python/whl/<cuda-version> Where `` is one of the following: - `cu118`: CUDA 11.8 - `cu121`: CUDA 12.1 - `cu122`: CUDA 12.2 - `cu123`: CUDA 12.3 - `cu124`: CUDA 12.4 - `cu125`: CUDA 12.5 - `cu130`: CUDA 13.0 - `cu132`: CUDA 13.2 For example, to install the CUDA 12.1 wheel: pipinstallllama-cpp-python\--extra-index-urlhttps://abetlen.github.io/llama-cpp-python/whl/cu121 Metal To install with Metal (MPS), set the `GGML_METAL=on` environment variable before installing: CMAKE_ARGS="-DGGML_METAL=on"pipinstallllama-cpp-python **Pre-built Wheel (New)** It is also possible to install a pre-built wheel with Metal support. As long as your system meets some requirements: - MacOS Version is 11.0 or later - Python Version is 3.10, 3.11 or 3.12 pipinstallllama-cpp-python\--extra-index-urlhttps://abetlen.github.io/llama-cpp-python/whl/metal HIP (ROCm) To install with HIP / ROCm support for AMD cards, set the `GGML_HIP=on` environment variable before installing: CMAKE_ARGS="-DGGML_HIP=on"pipinstallllama-cpp-python **Pre-built Wheel (New)** It is also possible to install a pre-built wheel with ROCm support for Linux: pipinstallllama-cpp-python\--extra-index-urlhttps://abetlen.github.io/llama-cpp-python/whl/rocm72 Or a pre-built wheel with HIP Radeon support for Windows: pipinstallllama-cpp-python`--extra-index-urlhttps://abetlen.github.io/llama-cpp-python/whl/hip-radeonVulkan To install with Vulkan support, set the `GGML_VULKAN=on` environment variable before installing: CMAKE_ARGS="-DGGML_VULKAN=on"pipinstallllama-cpp-python **Pre-built Wheel (New)** It is also possible to install a pre-built wheel with Vulkan support for Linux or Windows: pipinstallllama-cpp-python\--extra-index-urlhttps://abetlen.github.io/llama-cpp-python/whl/vulkan SYCL To install with SYCL support, set the `GGML_SYCL=on` environment variable before installing: source/opt/intel/oneapi/setvars.shCMAKE_ARGS="-DGGML_SYCL=on -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx"pipinstallllama-cpp-python RPC To install with RPC support, set the `GGML_RPC=on` environment variable before installing: source/opt/intel/oneapi/setvars.shCMAKE_ARGS="-DGGML_RPC=on"pipinstallllama-cpp-python Windows Notes

Error: Can't find 'nmake' or 'CMAKE_C_COMPILER' If you run into issues where it complains it can't find `'nmake'` `'?'` or CMAKE_C_COMPILER, you can extract w64devkit as [mentioned in llama.cpp repo](https://github.com/ggerganov/llama.cpp#openblas) and add those manually to CMAKE_ARGS before running `pip` install: $env:CMAKE_GENERATOR = "MinGW Makefiles" $env:CMAKE_ARGS = "-DGGML_OPENBLAS=on -DCMAKE_C_COMPILER=C:/w64devkit/bin/gcc.exe -DCMAKE_CXX_COMPILER=C:/w64devkit/bin/g++.exe" See the above instructions and set `CMAKE_ARGS` to the BLAS backend you want to use. MacOS Notes

Detailed MacOS Metal GPU install documentation is available at

docs/install/macos.md

M1 Mac Performance Issue Note: If you are using Apple Silicon (M1) Mac, make sure you have installed a version of Python that supports arm64 architecture. For example: wgethttps://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-MacOSX-arm64.sh bashMiniforge3-MacOSX-arm64.sh Otherwise, while installing it will build the llama.cpp x86 version which will be 10x slower on Apple Silicon (M1) Mac. M Series Mac Error: `(mach-o file, but is an incompatible architecture (have 'x86_64', need 'arm64'))` Try installing with CMAKE_ARGS="-DCMAKE_OSX_ARCHITECTURES=arm64 -DCMAKE_APPLE_SILICON_PROCESSOR=arm64 -DGGML_METAL=on"pipinstall--upgrade--verbose--force-reinstall--no-cache-dirllama-cpp-python Upgrading and Reinstalling

To upgrade and rebuild llama-cpp-python add --upgrade --force-reinstall --no-cache-dir flags to the pip install command to ensure the package is rebuilt from source.

High-level API

API Reference

The high-level API provides a simple managed interface through the

Llama

class.

Below is a short example demonstrating how to use the high-level API for basic text completion:

fromllama_cppimportLlamallm=Llama(model_path="./models/7B/llama-model.gguf",# n_gpu_layers=-1, # Uncomment to use GPU acceleration# seed=1337, # Uncomment to set a specific seed# n_ctx=2048, # Uncomment to increase the context window)output=llm("Q: Name the planets in the solar system? A: ",# Promptmax_tokens=32,# Generate up to 32 tokens, set to None to generate up to the end of the context windowstop=["Q:","\n"],# Stop generating just before the model would generate a new questionecho=True# Echo the prompt back in the output)# Generate a completion, can also call create_completionprint(output)By default llama-cpp-python generates completions in an OpenAI compatible format:

{"id":"cmpl-xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx","object":"text_completion","created":1679561337,"model":"./models/7B/llama-model.gguf","choices":[{"text":"Q: Name the planets in the solar system? A: Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus, Neptune and Pluto.","index":0,"logprobs":None,"finish_reason":"stop"}],"usage":{"prompt_tokens":14,"completion_tokens":28,"total_tokens":42}}Text completion is available through the

__call__

and

create_completion

methods of the

Llama

class.

Pulling models from Hugging Face Hub

You can download Llama models in gguf format directly from Hugging Face using the

from_pretrained

method. You'll need to install the huggingface-hub package to use this feature (pip install huggingface-hub).

llm=Llama.from_pretrained(repo_id="lmstudio-community/Qwen3.5-0.8B-GGUF",filename="*Q8_0.gguf",verbose=False)By default

from_pretrained

will download the model to the huggingface cache directory, you can then manage installed model files with the

tool.

Chat Completion

The high-level API also provides a simple interface for chat completion.

Chat completion requires that the model knows how to format the messages into a single prompt. The Llama class does this using pre-registered chat formats (ie. chatml, llama-2, gemma, etc) or by providing a custom chat handler object.

The model will format the messages into a single prompt using the following order of precedence: - Use the chat_handler if provided - Use the chat_format if provided - Use the tokenizer.chat_template from the gguf model's metadata (should work for most new models, older models may not have this) - else, fallback to the llama-2 chat format

Set verbose=True to see the selected chat format.

fromllama_cppimportLlamallm=Llama(model_path="path/to/llama-2/llama-model.gguf",chat_format="llama-2")llm.create_chat_completion(messages=[{"role":"system","content":"You are an assistant who perfectly describes images."},{"role":"user","content":"Describe this image in detail please."}])Chat completion is available through the

create_chat_completion

method of the

Llama

class.

For OpenAI API v1 compatibility, you use the

create_chat_completion_openai_v1

method which will return pydantic models instead of dicts.

JSON and JSON Schema Mode

To constrain chat responses to only valid JSON or a specific JSON Schema use the response_format argument in

create_chat_completion

JSON Mode

The following example will constrain the response to valid JSON strings only.

fromllama_cppimportLlamallm=Llama(model_path="path/to/model.gguf",chat_format="chatml")llm.create_chat_completion(messages=[{"role":"system","content":"You are a helpful assistant that outputs in JSON.",},{"role":"user","content":"Who won the world series in 2020"},],response_format={"type":"json_object",},temperature=0.7,)JSON Schema Mode

To constrain the response further to a specific JSON Schema add the schema to the schema property of the response_format argument.

The high-level API supports OpenAI compatible function and tool calling. This is possible through the functionary pre-trained models chat format or through the generic chatml-function-calling chat format.

fromllama_cppimportLlamallm=Llama(model_path="path/to/chatml/llama-model.gguf",chat_format="chatml-function-calling")llm.create_chat_completion(messages=[{"role":"system","content":"A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. The assistant calls functions with appropriate input when necessary"},{"role":"user","content":"Extract Jason is 25 years old"}],tools=[{"type":"function","function":{"name":"UserDetail","parameters":{"type":"object","title":"UserDetail","properties":{"name":{"title":"Name","type":"string"},"age":{"title":"Age","type":"integer"}},"required":["name","age"]}}}],tool_choice={"type":"function","function":{"name":"UserDetail"}})Functionary v2 The various gguf-converted files for this set of models can be found [here](https://huggingface.co/meetkai). Functionary is able to intelligently call functions and also analyze any provided function outputs to generate coherent responses. All v2 models of functionary supports **parallel function calling**. You can provide either `functionary-v1` or `functionary-v2` for the `chat_format` when initializing the Llama class. Due to discrepancies between llama.cpp and HuggingFace's tokenizers, it is required to provide HF Tokenizer for functionary. The `LlamaHFTokenizer` class can be initialized and passed into the Llama class. This will override the default llama.cpp tokenizer used in Llama class. The tokenizer files are already included in the respective HF repositories hosting the gguf files. fromllama_cppimportLlamafromllama_cpp.llama_tokenizerimportLlamaHFTokenizerllm=Llama.from_pretrained(repo_id="meetkai/functionary-small-v2.2-GGUF",filename="functionary-small-v2.2.q4_0.gguf",chat_format="functionary-v2",tokenizer=LlamaHFTokenizer.from_pretrained("meetkai/functionary-small-v2.2-GGUF")) **NOTE**: There is no need to provide the default system messages used in Functionary as they are added automatically in the Functionary chat handler. Thus, the messages should contain just the chat messages and/or system messages that provide additional context for the model (e.g.: datetime, etc.). Multi-modal Models

llama-cpp-python supports such as llava1.5 which allow the language model to read information from both text and images.

Below are the supported multi-modal models and their respective chat handlers (Python API) and chat formats (Server API).

ModelLlamaChatHandlerchat_format

llava-v1.5-7b

Llava15ChatHandlerllava-1-5

llava-v1.5-13b

Llava15ChatHandlerllava-1-5

llava-v1.6-34b

Llava16ChatHandlerllava-1-6

moondream2

MoondreamChatHandlermoondream2

nanollava

NanoLlavaChatHandlernanollava

llama-3-vision-alpha

Llama3VisionAlphaChatHandlerllama-3-vision-alpha

minicpm-v-2.6

MiniCPMv26ChatHandlerminicpm-v-2.6

qwen2.5-vl

Qwen25VLChatHandlerqwen2.5-vl

gemma-4

Gemma4ChatHandlergemma4GGUF models with an mtmd projector and embedded chat templateMTMDChatHandlermtmdTry Gemma 4 12B in Google Colab ->

Try Gemma 4 12B QAT in Google Colab ->

Then you'll need to use a custom chat handler to load the clip model and process the chat messages and images.

fromllama_cppimportLlamafromllama_cpp.llama_chat_formatimportLlava15ChatHandlerchat_handler=Llava15ChatHandler(clip_model_path="path/to/llava/mmproj.bin")llm=Llama(model_path="./path/to/llava/llama-model.gguf",chat_handler=chat_handler,n_ctx=2048,# n_ctx should be increased to accommodate the image embedding)llm.create_chat_completion(messages=[{"role":"system","content":"You are an assistant who perfectly describes images."},{"role":"user","content":[{"type":"text","text":"What's in this image?"},{"type":"image_url","image_url":{"url":"https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg"}}]}])You can also pull the model from the Hugging Face Hub using the from_pretrained method.

fromllama_cppimportLlamafromllama_cpp.llama_chat_formatimportMoondreamChatHandlerchat_handler=MoondreamChatHandler.from_pretrained(repo_id="vikhyatk/moondream2",filename="*mmproj*",)llm=Llama.from_pretrained(repo_id="vikhyatk/moondream2",filename="*text-model*",chat_handler=chat_handler,n_ctx=2048,# n_ctx should be increased to accommodate the image embedding)response=llm.create_chat_completion(messages=[{"role":"user","content":[{"type":"text","text":"What's in this image?"},{"type":"image_url","image_url":{"url":"https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg"}}]}])print(response["choices"][0]["text"])Note: Multi-modal models also support tool calling and JSON mode.

Loading a Local Image Images can be passed as base64 encoded data URIs. The following example demonstrates how to do this. importbase64defimage_to_base64_data_uri(file_path):withopen(file_path,"rb")asimg_file:base64_data=base64.b64encode(img_file.read()).decode('utf-8')returnf"data:image/png;base64,{base64_data}"# Replace 'file_path.png' with the actual path to your PNG filefile_path='file_path.png'data_uri=image_to_base64_data_uri(file_path)messages=[{"role":"system","content":"You are an assistant who perfectly describes images."},{"role":"user","content":[{"type":"image_url","image_url":{"url":data_uri}},{"type":"text","text":"Describe this image in detail please."}]}]Speculative Decoding

llama-cpp-python supports speculative decoding which allows the model to generate completions based on a draft model.

The fastest way to use speculative decoding is through the LlamaPromptLookupDecoding class.

Just pass this as a draft model to the Llama class during initialization.

fromllama_cppimportLlamafromllama_cpp.llama_speculativeimportLlamaPromptLookupDecodingllama=Llama(model_path="path/to/model.gguf",draft_model=LlamaPromptLookupDecoding(num_pred_tokens=10)# num_pred_tokens is the number of tokens to predict 10 is the default and generally good for gpu, 2 performs better for cpu-only machines.)Embeddings

To generate text embeddings use

create_embedding

embed

. Note that you must pass embedding=True to the constructor upon model creation for these to work properly.

importllama_cppllm=llama_cpp.Llama(model_path="path/to/model.gguf",embedding=True)embeddings=llm.create_embedding("Hello, world!")# or create multiple embeddings at onceembeddings=llm.create_embedding(["Hello, world!","Goodbye, world!"])There are two primary notions of embeddings in a Transformer-style model: token level and sequence level. Sequence level embeddings are produced by "pooling" token level embeddings together, usually by averaging them or using the first token.

Models that are explicitly geared towards embeddings will usually return sequence level embeddings by default, one for each input string. Non-embedding models such as those designed for text generation will typically return only token level embeddings, one for each token in each sequence. Thus the dimensionality of the return type will be one higher for token level embeddings.

It is possible to control pooling behavior in some cases using the pooling_type flag on model creation. You can ensure token level embeddings from any model using LLAMA_POOLING_TYPE_NONE. The reverse, getting a generation oriented model to yield sequence level embeddings is currently not possible, but you can always do the pooling manually.

Adjusting the Context Window

The context window of the Llama models determines the maximum number of tokens that can be processed at once. By default, this is set to 512 tokens, but can be adjusted based on your requirements.

For instance, if you want to work with larger contexts, you can expand the context window by setting the n_ctx parameter when initializing the Llama object:

llm=Llama(model_path="./models/7B/llama-model.gguf",n_ctx=2048)OpenAI Compatible Web Server

llama-cpp-python offers a web server which aims to act as a drop-in replacement for the OpenAI API. This allows you to use llama.cpp compatible models with any OpenAI compatible client (language libraries, services, etc).

To install the server package and get started:

pipinstall'llama-cpp-python[server]'python3-mllama_cpp.server--modelmodels/7B/llama-model.gguf Similar to Hardware Acceleration section above, you can also install with GPU (cuBLAS) support like this:

CMAKE_ARGS="-DGGML_CUDA=on"FORCE_CMAKE=1pipinstall'llama-cpp-python[server]'python3-mllama_cpp.server--modelmodels/7B/llama-model.gguf--n_gpu_layers35Navigate to

http://localhost:8000/docs

to see the OpenAPI documentation.

To bind to 0.0.0.0 to enable remote connections, use python3 -m llama_cpp.server --host 0.0.0.0. Similarly, to change the port (default is 8000), use --port.

You probably also want to set the prompt format. For chatml, use

python3-mllama_cpp.server--modelmodels/7B/llama-model.gguf--chat_formatchatml That will format the prompt according to how model expects it. You can find the prompt format in the model card. For possible options, see

llama_cpp/llama_chat_format.py

and look for lines starting with "

@register_chat_format

If you have huggingface-hub installed, you can also use the --hf_model_repo_id flag to load a model from the Hugging Face Hub.

python3-mllama_cpp.server--hf_model_repo_idlmstudio-community/Qwen3.5-0.8B-GGUF--model'*Q8_0.gguf'Web Server Features

Local Copilot replacement

Function Calling support

Vision API support

Multiple Models

Docker image

A Docker image is available on

GHCR

. To run the server:

dockerrun--rm-it-p8000:8000-v/path/to/models:/models-eMODEL=/models/llama-model.ggufghcr.io/abetlen/llama-cpp-python:latest

Docker on termux (requires root)

is currently the only known way to run this on phones, see

termux support issue

Low-level API

API Reference

The low-level API is a direct

ctypes

binding to the C API provided by llama.cpp. The entire low-level API can be found in

llama_cpp/llama_cpp.py

and directly mirrors the C API in

llama.h

Below is a short example demonstrating how to use the low-level API to tokenize a prompt:

importllama_cppimportctypesllama_cpp.llama_backend_init()# Must be called once at the start of each programmodel_params=llama_cpp.llama_model_default_params()ctx_params=llama_cpp.llama_context_default_params()prompt=b"Q: Name the planets in the solar system? A: "# use bytes for char * paramsmodel=llama_cpp.llama_model_load_from_file(b"./models/7b/llama-model.gguf",model_params)ctx=llama_cpp.llama_init_from_model(model,ctx_params)vocab=llama_cpp.llama_model_get_vocab(model)max_tokens=ctx_params.n_ctx# use ctypes arrays for array paramstokens=(llama_cpp.llama_token*int(max_tokens))()n_tokens=llama_cpp.llama_tokenize(vocab,prompt,len(prompt),tokens,max_tokens,True,False)llama_cpp.llama_free(ctx)llama_cpp.llama_model_free(model)Check out the

examples folder

for more examples of using the low-level API.

Documentation

Documentation is available via

https://llama-cpp-python.readthedocs.io/

. If you find any issues with the documentation, please open an issue or submit a PR.

Development

This package is under active development and I welcome any contributions. See

CONTRIBUTING.md

for contribution workflow, PR title, changelog, testing, and style guidelines.

To get started, clone the repository and install the package in editable / development mode:

gitclone--recurse-submoduleshttps://github.com/abetlen/llama-cpp-python.git cdllama-cpp-python # Upgrade pip (required for editable mode)pipinstall--upgradepip # Install with pippipinstall-e. # install development tooling (tests, docs, ruff)pipinstall-e'.[dev]'# if you want to use the fastapi / openapi serverpipinstall-e'.[server]'# to install all optional dependenciespipinstall-e'.[all]'# to clear the local build cachemakeclean Now try running the tests

pytest And check formatting / linting before opening a PR:

python-mruffcheckllama_cpptests python-mruffformat--checkllama_cpptests # or use the Makefile targetsmakelint makeformat There's a Makefile available with useful targets. A typical workflow would look like this:

makebuild maketestYou can also test out specific commits of llama.cpp by checking out the desired commit in the vendor/llama.cpp submodule and then running make clean and pip install -e . again. Any changes in the llama.h API will require changes to the llama_cpp/llama_cpp.py file to match the new API (additional changes may be required elsewhere).

FAQ

Are there pre-built binaries / binary wheels available?

The recommended installation method is to install from source as described above. The reason for this is that llama.cpp is built with compiler optimizations that are specific to your system. Using pre-built binaries would require disabling these optimizations or supporting a large number of pre-built binaries for each platform.

That being said there are some pre-built binaries available through the Releases as well as some community provided wheels.

In the future, I would like to provide pre-built binaries and wheels for common platforms and I'm happy to accept any useful contributions in this area. This is currently being tracked in

#741

How does this compare to other Python bindings of llama.cpp?

I originally wrote this package for my own use with two goals in mind:

Provide a simple process to install llama.cpp and access the full C API in llama.h from Python

Provide a high-level Python API that can be used as a drop-in replacement for the OpenAI API so existing apps can be easily ported to use llama.cpp

Any contributions and changes to this package will be made with these goals in mind.

License

This project is licensed under the terms of the MIT license.