2023-03-10 19:47:46 +00:00
# llama.cpp
2023-04-05 15:56:20 +00:00
2023-03-26 07:20:49 +00:00
2024-05-30 15:07:39 +00:00
[](https://opensource.org/licenses/MIT)
2025-06-05 07:50:55 +00:00
[](https://github.com/ggml-org/llama.cpp/releases)
2025-02-15 14:40:57 +00:00
[](https://github.com/ggml-org/llama.cpp/actions/workflows/server.yml)
2023-03-12 20:09:26 +00:00
2025-07-11 13:07:55 +00:00
[Manifesto ](https://github.com/ggml-org/llama.cpp/discussions/205 ) / [ggml ](https://github.com/ggml-org/ggml ) / [ops ](https://github.com/ggml-org/llama.cpp/blob/master/docs/ops.md )
2023-06-25 13:08:12 +00:00
2025-07-11 13:07:55 +00:00
LLM inference in C/C++
2023-03-10 19:47:46 +00:00
2024-07-05 16:08:32 +00:00
## Recent API changes
2024-03-03 10:44:03 +00:00
2025-02-15 14:40:57 +00:00
- [Changelog for `libllama` API ](https://github.com/ggml-org/llama.cpp/issues/9289 )
- [Changelog for `llama-server` REST API ](https://github.com/ggml-org/llama.cpp/issues/9291 )
2024-03-03 10:44:03 +00:00
2024-07-05 16:08:32 +00:00
## Hot topics
2023-08-27 11:44:35 +00:00
2026-03-24 09:35:07 +00:00
- **Hugging Face cache migration: models downloaded with `-hf` are now stored in the standard Hugging Face cache directory, enabling sharing with other HF tools.**
2025-11-04 15:21:31 +00:00
- **[guide : using the new WebUI of llama.cpp](https://github.com/ggml-org/llama.cpp/discussions/16938)**
- [guide : running gpt-oss with llama.cpp ](https://github.com/ggml-org/llama.cpp/discussions/15396 )
- [[FEEDBACK] Better packaging for llama.cpp to support downstream consumers 🤗](https://github.com/ggml-org/llama.cpp/discussions/15313)
2025-08-05 17:19:33 +00:00
- Support for the `gpt-oss` model with native MXFP4 format has been added | [PR ](https://github.com/ggml-org/llama.cpp/pull/15091 ) | [Collaboration with NVIDIA ](https://blogs.nvidia.com/blog/rtx-ai-garage-openai-oss ) | [Comment ](https://github.com/ggml-org/llama.cpp/discussions/15095 )
2025-07-11 13:07:55 +00:00
- Multimodal support arrived in `llama-server` : [#12898 ](https://github.com/ggml-org/llama.cpp/pull/12898 ) | [documentation ](./docs/multimodal.md )
2025-04-28 09:10:18 +00:00
- VS Code extension for FIM completions: https://github.com/ggml-org/llama.vscode
2025-01-22 17:44:26 +00:00
- Vim/Neovim plugin for FIM completions: https://github.com/ggml-org/llama.vim
2025-02-15 14:40:57 +00:00
- Hugging Face Inference Endpoints now support GGUF out of the box! https://github.com/ggml-org/llama.cpp/discussions/9669
- Hugging Face GGUF editor: [discussion ](https://github.com/ggml-org/llama.cpp/discussions/9268 ) | [tool ](https://huggingface.co/spaces/CISCai/gguf-editor )
2023-08-18 14:48:31 +00:00
----
2023-03-14 07:43:52 +00:00
2025-06-03 11:09:36 +00:00
## Quick start
Getting started with llama.cpp is straightforward. Here are several ways to install it on your machine:
- Install `llama.cpp` using [brew, nix or winget ](docs/install.md )
- Run with Docker - see our [Docker documentation ](docs/docker.md )
- Download pre-built binaries from the [releases page ](https://github.com/ggml-org/llama.cpp/releases )
- Build from source by cloning this repository - check out [our build guide ](docs/build.md )
Once installed, you'll need a model to work with. Head to the [Obtaining and quantizing models ](#obtaining-and-quantizing-models ) section to learn more.
Example command:
```sh
# Use a local model file
llama-cli -m my_model.gguf
# Or download and run a model directly from Hugging Face
llama-cli -hf ggml-org/gemma-3-1b-it-GGUF
# Launch OpenAI-compatible API server
llama-server -hf ggml-org/gemma-3-1b-it-GGUF
```
2023-03-10 19:47:46 +00:00
## Description
2024-02-05 14:55:10 +00:00
The main goal of `llama.cpp` is to enable LLM inference with minimal setup and state-of-the-art performance on a wide
2024-11-30 07:47:07 +00:00
range of hardware - locally and in the cloud.
2023-03-10 19:47:46 +00:00
2024-02-05 14:55:10 +00:00
- Plain C/C++ implementation without any dependencies
- Apple silicon is a first-class citizen - optimized via ARM NEON, Accelerate and Metal frameworks
2024-10-18 05:34:36 +00:00
- AVX, AVX2, AVX512 and AMX support for x86 architectures
2025-12-08 08:41:34 +00:00
- RVV, ZVFH, ZFH, ZICBOP and ZIHINTPAUSE support for RISC-V architectures
2024-02-19 07:39:31 +00:00
- 1.5-bit, 2-bit, 3-bit, 4-bit, 5-bit, 6-bit, and 8-bit integer quantization for faster inference and reduced memory use
2025-05-21 01:58:49 +00:00
- Custom CUDA kernels for running LLMs on NVIDIA GPUs (support for AMD GPUs via HIP and Moore Threads GPUs via MUSA)
2024-06-04 18:23:20 +00:00
- Vulkan and SYCL backend support
2024-02-05 14:55:10 +00:00
- CPU+GPU hybrid inference to partially accelerate models larger than the total VRAM capacity
2023-03-10 19:47:46 +00:00
2025-02-15 14:40:57 +00:00
The `llama.cpp` project is the main playground for developing new features for the [ggml ](https://github.com/ggml-org/ggml ) library.
2023-03-11 10:31:21 +00:00
2024-11-30 07:47:07 +00:00
< details >
< summary > Models< / summary >
2023-03-29 16:37:20 +00:00
2024-02-07 06:21:30 +00:00
Typically finetunes of the base models below are supported as well.
2024-12-01 09:25:17 +00:00
Instructions for adding support for new models: [HOWTO-add-model.md ](docs/development/HOWTO-add-model.md )
2024-11-30 07:47:07 +00:00
2024-12-01 09:25:17 +00:00
#### Text-only
2024-11-30 07:47:07 +00:00
2023-03-30 19:31:54 +00:00
- [X] LLaMA 🦙
2023-07-28 01:14:11 +00:00
- [x] LLaMA 2 🦙🦙
2024-04-25 13:52:28 +00:00
- [x] LLaMA 3 🦙🦙🦙
2024-02-07 06:21:30 +00:00
- [X] [Mistral 7B ](https://huggingface.co/mistralai/Mistral-7B-v0.1 )
2024-02-05 14:55:10 +00:00
- [x] [Mixtral MoE ](https://huggingface.co/models?search=mistral-ai/Mixtral )
2024-04-13 09:33:52 +00:00
- [x] [DBRX ](https://huggingface.co/databricks/dbrx-instruct )
2025-10-26 12:01:20 +00:00
- [x] [Jamba ](https://huggingface.co/ai21labs )
2024-04-21 12:35:40 +00:00
- [X] [Falcon ](https://huggingface.co/models?search=tiiuae/falcon )
2023-08-02 06:18:31 +00:00
- [X] [Chinese LLaMA / Alpaca ](https://github.com/ymcui/Chinese-LLaMA-Alpaca ) and [Chinese LLaMA-2 / Alpaca-2 ](https://github.com/ymcui/Chinese-LLaMA-Alpaca-2 )
2023-03-30 19:31:54 +00:00
- [X] [Vigogne (French) ](https://github.com/bofenghuang/vigogne )
2025-02-15 14:40:57 +00:00
- [X] [BERT ](https://github.com/ggml-org/llama.cpp/pull/5423 )
2023-04-10 20:41:53 +00:00
- [X] [Koala ](https://bair.berkeley.edu/blog/2023/04/03/koala/ )
2023-10-17 18:13:21 +00:00
- [X] [Baichuan 1 & 2 ](https://huggingface.co/models?search=baichuan-inc/Baichuan ) + [derivations ](https://huggingface.co/hiyouga/baichuan-7b-sft )
- [X] [Aquila 1 & 2 ](https://huggingface.co/models?search=BAAI/Aquila )
2025-02-15 14:40:57 +00:00
- [X] [Starcoder models ](https://github.com/ggml-org/llama.cpp/pull/3187 )
2023-10-06 19:13:36 +00:00
- [X] [Refact ](https://huggingface.co/smallcloudai/Refact-1_6B-fim )
2025-02-15 14:40:57 +00:00
- [X] [MPT ](https://github.com/ggml-org/llama.cpp/pull/3417 )
- [X] [Bloom ](https://github.com/ggml-org/llama.cpp/pull/3553 )
2023-12-14 07:38:49 +00:00
- [x] [Yi models ](https://huggingface.co/models?search=01-ai/Yi )
2024-02-06 14:06:48 +00:00
- [X] [StableLM models ](https://huggingface.co/stabilityai )
2023-12-14 07:38:49 +00:00
- [x] [Deepseek models ](https://huggingface.co/models?search=deepseek-ai/deepseek )
- [x] [Qwen models ](https://huggingface.co/models?search=Qwen/Qwen )
2025-02-15 14:40:57 +00:00
- [x] [PLaMo-13B ](https://github.com/ggml-org/llama.cpp/pull/3557 )
2024-02-06 14:06:48 +00:00
- [x] [Phi models ](https://huggingface.co/models?search=microsoft/phi )
2025-02-15 14:40:57 +00:00
- [x] [PhiMoE ](https://github.com/ggml-org/llama.cpp/pull/11003 )
2023-12-28 14:03:57 +00:00
- [x] [GPT-2 ](https://huggingface.co/gpt2 )
2025-02-15 14:40:57 +00:00
- [x] [Orion 14B ](https://github.com/ggml-org/llama.cpp/pull/5118 )
2024-02-06 14:06:48 +00:00
- [x] [InternLM2 ](https://huggingface.co/models?search=internlm2 )
2024-02-05 07:41:38 +00:00
- [x] [CodeShell ](https://github.com/WisdomShell/codeshell )
2024-02-21 13:08:22 +00:00
- [x] [Gemma ](https://ai.google.dev/gemma )
llama : support Mamba Selective State Space Models (#5328)
* mamba : begin working on support for Mamba SSM
* mamba : begin figuring out how to (ab)use the kv cache for Mamba
* mamba : recurrent inference almost works, but incoherent
* mamba : recurrent inference WORKS!!!
* convert : optionally use d_conv and d_state from config.json for Mamba
* mamba : refactor recurrent conv, resulting in 20% perf increase
It's still slower than I'd like, but I did not really optimize `ggml_exp` yet.
I also refactored `ggml_exp` to work with tensors with more than 2 dimensions.
* ggml : parallelize ggml_exp
This results in 8% faster token generation for Mamba-130M.
* mamba : simplify the conv step with a self-overlapping view
Turns out the conv_state can be made smaller by one column.
Note that this breaks existing GGUFs of Mamba,
because the key_value_length field is tied to the conv_state size.
Convolution with a self-overlapping view is cool!
And it's much simpler than what I initially thought would be necessary
to make the convolution step work with more than 1 token at a time.
Next step is to make the SSM step work on batches of tokens too,
and thus I need to figure out a way to make a parallel selective scan
which will keep the ssm_state small and won't make it bigger
by a factor of (n_layer * batch_size).
* llama : fix Mamba KV self size wrongly displaying as f16 instead of f32
Relatedly, I also tried to see if other types than f32 worked for the states,
but they don't, because of the operators used.
It's probably better anyway to keep lots of precision there,
since the states are small anyway.
* mamba : fix self-overlapping view depth stride
* mamba : handle batches of more than 1 token
This means running Mamba no longer crashes when using the default settings!
And probably also slightly faster prompt processing.
Both batched and non-batched processing yield the same output.
Previously, the state was not cleared when starting a sequence.
Next step is to make the KV cache API work as expected for Mamba models.
* ggml: add ggml_ssm_scan to help with parallel selective scan
If the selective scan was implemented without a custom operator,
there would be waaay too many nodes in the graph. For example,
for Mamba-130M, with a batch size of 512 (the default),
a naive selective scan could add at least 24*512=12288 nodes,
which is more than LLAMA_MAX_NODES (8192),
and that's only for the smallest Mamba model.
So it's much cleaner with a custom operator.
Not sure about the name, though.
* ggml : in ggml_ssm_scan, merge multiple rows in the same vec operation
This will help with performance on CPU if ggml_vec_mul_f32
and ggml_vec_add_f32 are ever optimized with SIMD.
* mamba : very basic quantization support
Mostly works, but there is currently no difference
between the variants of a k-quant (e.g. Q4_K_S and Q4_K_M are the same).
Most of the SSM-specific weights can be kept in f32 without affecting
the size that much, since they are relatively small.
(the linear projection weights are responsible for most of Mamba's size)
Too much quantization seems to make the state degrade quite fast, and
the model begins to output gibberish.
It seems to affect bigger models to a lesser extent than small models,
but I'm not sure by how much.
Experimentation will be needed to figure out which weights are more important
for the _M (and _L?) variants of k-quants for Mamba.
* convert : fix wrong name for layer norm weight of offical Mamba models
I was using Q-bert/Mamba-* models before, which have a slighlty different
naming scheme for the weights.
(they start with "model.layers" instead of "backbone.layers")
* mamba : fuse more steps of the SSM scan in the ggml_ssm_scan operator
This increases performance on CPU by around 30% for prompt processing,
and by around 20% for text generation.
However, it also makes the ggml_exp and ggml_soft_plus operators unused.
Whether or not they should be kept will be decided later.
* convert : for Mamba, also consider the "MambaLMHeadModel" arch name
It's the name of the class of the official implementation,
though they don't use it (yet) in the "architectures" field of config.json
* mamba : fix vocab size problems with official models
The perplexity was waaaay to high for models with a non-round vocab size.
Not sure why, but it needed to be fixed in the metadata.
Note that this breaks existing GGUF-converted Mamba models,
but **only if** the vocab size was not already rounded.
* ggml : remove ggml_exp and ggml_soft_plus
They did not exist anyway outside of this branch,
and since ggml_ssm_scan fused operations together, they are unused.
It's always possible to bring them back if needed.
* mamba : remove some useless comments
No code change.
* convert : fix flake8 linter errors
* mamba : apply suggestions from code review
* mamba : remove unecessary branch for row-wise ssm_state and C multiplication
It was previously done to avoid permuting when only one token is processed
at a time (like when generating text), but permuting is cheap,
and dynamically changing the compute graph is not future-proof.
* ggml : in ggml_ssm_scan, use more appropriate asserts
* ggml : rename the destination pointer in ggml_compute_forward_ssm_scan_f32
* mamba : multiple sequences, but one at a time
This is a step towards making this Mamba implementation usable
with the server example (the way the system prompt is kept when clearing
the client slots will need to be changed before this can work, though).
The KV cache size for this kind of model is tied to the maximum number
of sequences kept at any single time.
For now, this number is obtained from n_parallel (plus one,
to have an extra sequence to dedicate to the system prompt),
but there might be a better way to do this which won't also
make the main example use 2 cells even if only 1 is really used.
(for this specific case, --parallel 0 helps)
Simultaneous sequence processing will probably require changes to
ggml_ssm_scan, and possibly a new operator for the conv step.
* mamba : support llama_kv_cache_seq_cp
This (mis)uses the logic around K shifts, because tokens in a state
can't be shifted anyway, and because inp_K_shift has the right shape and type.
Using ggml_get_rows is a nice way to do copies, but copy chains can't work.
Fortunately, copy chains don't really seem to be used in the examples.
Each KV cell is dedicated to the sequence ID corresponding to its own index.
* mamba : use a state mask
It's cleaner than the previous heuristic of
checking for the pos of the first token in the batch.
inp_KQ_mask could not be re-used for this, because it has the wrong shape
and because it seems more suited to the next step of
simultaneous sequence processing (helping with the problem of
remembering which token belongs to which sequence(s)/state(s)).
* llama : replace the usage of n_ctx with kv_self.size in many places
* mamba : use n_tokens directly instead of n_tok
* mamba : in comments, properly refer to KV cells instead of slots
* mamba : reduce memory usage of ggml_ssm_scan
From 290.37 MiB to 140.68 MiB of CPU compute buffer size
with Mamba 3B with a batch size of 512.
The result tensor of ggml_ssm_scan was previously a big part
of the CPU compute buffer size. To make it smaller,
it does not contain the intermediate ssm states anymore.
Both y and the last ssm state are combined in the result tensor,
because it seems only a single tensor can be returned by an operator
with the way the graph is built.
* mamba : simultaneous sequence processing
A batch can now contain tokens from multiple sequences.
This is necessary for at least the parallel example, the server example,
and the HellaSwag test in the perplexity example.
However, for this to be useful, uses of llama_kv_cache_seq_rm/cp
will need to be changed to work on whole sequences.
* ggml : add ggml_ssm_conv as a new operator for the conv step of Mamba
This operator makes it possible to use and update the correct states
for each token of the batch in the same way as ggml_ssm_scan.
Other solutions which use existing operators would need loops which would
add too many nodes to the graph (at least the ones I thought of).
Using this operator further reduces the size of the CPU compute buffer
from 140.68 MiB to 103.20 MiB with Mamba 3B with a batch size of 512.
And (at least on CPU), it's a bit faster than before.
Note that "ggml_ssm_conv" is probably not the most appropriate name,
and it could be changed if a better one is found.
* llama : add inp_s_seq as a new input tensor
The most convenient implementation to select the correct state (for Mamba)
for each token is to directly get the correct index from a tensor.
This is why inp_s_seq is storing int32_t and not floats.
The other, less convenient way to select the correct state would be
to have inp_KQ_mask contain 1.0f for each state used by a token
and 0.0f otherwise. This complicates quickly fetching the first used
state of a token, and is also less efficient because a whole row
of the mask would always need to be read for each token.
Using indexes makes it easy to stop searching when there are
no more sequences for a token, and the first sequence assigned
is always very quickly available (it's the first element of each row).
* mamba : support llama_kv_cache_seq_cp copy chains
* mamba : support shifting and dividing the kv cache pos
* mamba : make the server and parallel examples work with whole sequences
A seq_id is dedicated to the system prompt in both cases.
* llama : make llama_kv_cache_seq_rm return whether it succeeded or not
* mamba : dedicate an input tensor for state copy indices
This is cleaner and makes it easier to adapt when/if token positions
(and by extension, inp_K_shift) are no longer integers.
* mamba : adapt perplexity, batched, and batched-bench examples
* perplexity : limit the max number of sequences
This adapts to what the loaded model can provide.
* llama : add llama_n_max_seq to get the upper limit for seq_ids
Used by the perplexity example.
* batched : pass n_parallel to the model's context params
This should have been there already, but it wasn't.
* batched-bench : reserve sequences to support Mamba
* batched-bench : fix tokens being put in wrong sequences
Generation quality isn't what's measured in there anyway,
but at least using the correct sequences avoids using non-consecutive
token positions.
* mamba : stop abusing attention metadata
This breaks existing converted-to-GGUF Mamba models,
but will allow supporting mixed architectures like MambaFormer
without needing to break Mamba models.
This will also allow changing the size of Mamba's states
without having to reconvert models in the future.
(e.g. using something else than d_conv - 1 columns for the conv_states
will not require breaking existing converted Mamba models again)
* gguf-py : add new KV metadata key-value pairs for Mamba
* llama : add new metadata key-value pairs for Mamba
* llama : guard against divisions by zero when n_head is 0
* mamba : rename "unlimited" KV cache property to "recurrent"
* mamba : more correctly update the "used" field of the KV cache
* ggml : in ggml_ssm_scan, use a threshold for soft_plus
This is how the official Mamba implementation does it,
and it's also what torch.nn.Softplus does.
* convert : for Mamba, fallback to internal NeoX tokenizer
The resulting models are exactly the same
as if the tokenizer.json and tokenizer_config.json of GPT-NeoX were there.
* mamba : support state saving and restoring
* ggml : implicitly pass src tensors through dst for Mamba-related ops
* mamba : clarify some comments
* server : fix cache_tokens not getting correctly resized
Otherwise, when the "we have to evaluate at least 1 token" special case
was triggered, an extra token was kept in cache_tokens even if it was
removed from the KV cache.
For Mamba, this caused useless prompt reprocessing when the previous
request triggered the above case.
* convert-hf : support new metadata keys for Mamba
For the models available at
https://huggingface.co/collections/state-spaces/transformers-compatible-mamba-65e7b40ab87e5297e45ae406
* mamba : rename metadata to be more similar to transformers library
This breaks existing converted-to-GGUF models,
but the metadata names are more "standard".
* mamba : support mamba-*-hf models
These models share their token_embd.weight with their output.weight
* mamba : add missing spaces
This is purely a formatting change.
* convert-hf : omit output.weight when identical with token_embd.weight
Only for Mamba for now, but it might be relevant for other models eventually.
Most Mamba models actually share these two tensors, albeit implicitly.
* readme : add Mamba to supported models, and add recent API changes
* mamba : move state_seq and state_mask views outside layer loop
A few tensors were also missing `struct` in front of `ggml_tensor`.
2024-03-08 22:31:00 +00:00
- [x] [Mamba ](https://github.com/state-spaces/mamba )
2024-04-25 13:52:28 +00:00
- [x] [Grok-1 ](https://huggingface.co/keyfan/grok-1-hf )
2024-03-29 13:37:03 +00:00
- [x] [Xverse ](https://huggingface.co/models?search=xverse )
2024-04-25 13:52:28 +00:00
- [x] [Command-R models ](https://huggingface.co/models?search=CohereForAI/c4ai-command-r )
2024-04-03 18:05:10 +00:00
- [x] [SEA-LION ](https://huggingface.co/models?search=sea-lion )
2024-04-07 17:33:59 +00:00
- [x] [GritLM-7B ](https://huggingface.co/GritLM/GritLM-7B ) + [GritLM-8x7B ](https://huggingface.co/GritLM/GritLM-8x7B )
2024-04-19 09:35:54 +00:00
- [x] [OLMo ](https://allenai.org/olmo )
2024-11-26 20:55:29 +00:00
- [x] [OLMo 2 ](https://allenai.org/olmo )
2024-09-16 06:47:37 +00:00
- [x] [OLMoE ](https://huggingface.co/allenai/OLMoE-1B-7B-0924 )
2024-08-05 05:54:10 +00:00
- [x] [Granite models ](https://huggingface.co/collections/ibm-granite/granite-code-models-6624c5cec322e4c148c8b330 )
2024-05-23 12:12:43 +00:00
- [x] [GPT-NeoX ](https://github.com/EleutherAI/gpt-neox ) + [Pythia ](https://github.com/EleutherAI/pythia )
2024-08-05 05:54:10 +00:00
- [x] [Snowflake-Arctic MoE ](https://huggingface.co/collections/Snowflake/arctic-66290090abe542894a5ac520 )
- [x] [Smaug ](https://huggingface.co/models?search=Smaug )
- [x] [Poro 34B ](https://huggingface.co/LumiOpen/Poro-34B )
- [x] [Bitnet b1.58 models ](https://huggingface.co/1bitLLM )
- [x] [Flan T5 ](https://huggingface.co/models?search=flan-t5 )
- [x] [Open Elm models ](https://huggingface.co/collections/apple/openelm-instruct-models-6619ad295d7ae9f868b759ca )
2025-02-02 07:48:46 +00:00
- [x] [ChatGLM3-6b ](https://huggingface.co/THUDM/chatglm3-6b ) + [ChatGLM4-9b ](https://huggingface.co/THUDM/glm-4-9b ) + [GLMEdge-1.5b ](https://huggingface.co/THUDM/glm-edge-1.5b-chat ) + [GLMEdge-4b ](https://huggingface.co/THUDM/glm-edge-4b-chat )
2025-04-11 10:10:10 +00:00
- [x] [GLM-4-0414 ](https://huggingface.co/collections/THUDM/glm-4-0414-67f3cbcb34dd9d252707cb2e )
2024-08-05 05:54:10 +00:00
- [x] [SmolLM ](https://huggingface.co/collections/HuggingFaceTB/smollm-6695016cad7167254ce15966 )
2024-08-16 06:35:18 +00:00
- [x] [EXAONE-3.0-7.8B-Instruct ](https://huggingface.co/LGAI-EXAONE/EXAONE-3.0-7.8B-Instruct )
2024-08-21 08:06:36 +00:00
- [x] [FalconMamba Models ](https://huggingface.co/collections/tiiuae/falconmamba-7b-66b9a580324dd1598b0f6d4a )
2024-09-12 00:29:53 +00:00
- [x] [Jais ](https://huggingface.co/inceptionai/jais-13b-chat )
2024-10-01 17:18:46 +00:00
- [x] [Bielik-11B-v2.3 ](https://huggingface.co/collections/speakleash/bielik-11b-v23-66ee813238d9b526a072408a )
2026-01-25 07:11:19 +00:00
- [x] [RWKV-7 ](https://huggingface.co/collections/shoumenchougou/rwkv7-gxx-gguf )
2024-10-22 10:33:37 +00:00
- [x] [RWKV-6 ](https://github.com/BlinkDL/RWKV-LM )
2025-01-10 01:58:08 +00:00
- [x] [QRWKV-6 ](https://huggingface.co/recursal/QRWKV6-32B-Instruct-Preview-v0.1 )
2024-12-15 17:02:46 +00:00
- [x] [GigaChat-20B-A3B ](https://huggingface.co/ai-sage/GigaChat-20B-A3B-instruct )
2025-03-30 18:38:33 +00:00
- [X] [Trillion-7B-preview ](https://huggingface.co/trillionlabs/Trillion-7B-preview )
2025-03-30 20:21:03 +00:00
- [x] [Ling models ](https://huggingface.co/collections/inclusionAI/ling-67c51c85b34a7ea0aba94c32 )
2025-07-12 17:07:08 +00:00
- [x] [LFM2 models ](https://huggingface.co/collections/LiquidAI/lfm2-686d721927015b2ad73eaa38 )
2025-09-01 07:34:59 +00:00
- [x] [Hunyuan models ](https://huggingface.co/collections/tencent/hunyuan-dense-model-6890632cda26b19119c9c5e7 )
2025-10-20 19:38:20 +00:00
- [x] [BailingMoeV2 (Ring/Ling 2.0) models ](https://huggingface.co/collections/inclusionAI/ling-v2-68bf1dd2fc34c306c1fa6f86 )
2023-12-14 07:38:49 +00:00
2024-12-01 09:25:17 +00:00
#### Multimodal
2023-12-14 07:38:49 +00:00
2024-02-28 08:39:39 +00:00
- [x] [LLaVA 1.5 models ](https://huggingface.co/collections/liuhaotian/llava-15-653aac15d994e992e2677a7e ), [LLaVA 1.6 models ](https://huggingface.co/collections/liuhaotian/llava-16-65b9e40155f60fd046a5ccf2 )
2024-02-05 14:55:10 +00:00
- [x] [BakLLaVA ](https://huggingface.co/models?search=SkunkworksAI/Bakllava )
2023-12-14 07:38:49 +00:00
- [x] [Obsidian ](https://huggingface.co/NousResearch/Obsidian-3B-V0.5 )
- [x] [ShareGPT4V ](https://huggingface.co/models?search=Lin-Chen/ShareGPT4V )
2024-01-25 20:14:32 +00:00
- [x] [MobileVLM 1.7B/3B models ](https://huggingface.co/models?search=mobileVLM )
2024-02-06 14:06:48 +00:00
- [x] [Yi-VL ](https://huggingface.co/models?search=Yi-VL )
2024-04-25 13:52:28 +00:00
- [x] [Mini CPM ](https://huggingface.co/models?search=MiniCPM )
2024-05-10 06:41:10 +00:00
- [x] [Moondream ](https://huggingface.co/vikhyatk/moondream2 )
2024-05-23 14:43:18 +00:00
- [x] [Bunny ](https://github.com/BAAI-DCAI/Bunny )
2025-02-02 07:48:46 +00:00
- [x] [GLM-EDGE ](https://huggingface.co/models?search=glm-edge )
llama : add Qwen2VL support + multimodal RoPE (#10361)
* Barebone Qwen2VL LLM convertor
* Add Qwen2VL cli entrypoint
* [WIP] add qwen2vl arch
* Verify m-rope output
* Add vl-rope/2d-rope support for qwen2vl ViT
* update qwen2vl cli tool
* update 5D tensor op workaround
* [WIP] qwen2vl vision model
* make batch and clip utils compatible with qwen2vl
* [WIP] create inference workflow, gguf convert script but fix
* correcting vision-rope behavior, add the missing last layer back to ViT
* add arg parser to qwen2vl_surgery
* replace variable size array with vector
* cuda-gdb cmake preset
* add fp32 mrope, vision rope kernel
* add fp16 support for qwen2vl and m-rope
* add `GGML_ROPE_TYPE_MROPE`, `GGML_ROPE_TYPE_VISION`
* fix rope op mode switching, out dated func args
* update `llama_hparams`
* update to keep up stream changes
* resolve linter, test errors
* add makefile entry, update speical image padding token
* add mrope unit test, fix few compiler warnings
* rename `mrope` related function, params
* minor updates on debug util, bug fixs
* add `m-rope` testcase to `test-backend-ops`
* Apply suggestions from code review
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* fix traililng whitespce
* store `llama_hparams.rope_sections` with fixed size array
* update position id tensor size check in GGML_OP_ROPE
* minor updates
* update `ggml_backend_*_supports_op` of unsupported backends
* remote old `rope_section` compare operator
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-12-14 12:43:46 +00:00
- [x] [Qwen2-VL ](https://huggingface.co/collections/Qwen/qwen2-vl-66cee7455501d7126940800d )
2025-08-22 07:29:08 +00:00
- [x] [LFM2-VL ](https://huggingface.co/collections/LiquidAI/lfm2-vl-68963bbc84a610f7638d5ffa )
2023-10-17 18:13:21 +00:00
2024-11-30 07:47:07 +00:00
< / details >
< details >
< summary > Bindings< / summary >
2023-04-05 15:56:20 +00:00
2025-06-01 08:44:30 +00:00
- Python: [ddh0/easy-llama ](https://github.com/ddh0/easy-llama )
2023-04-05 15:56:20 +00:00
- Python: [abetlen/llama-cpp-python ](https://github.com/abetlen/llama-cpp-python )
- Go: [go-skynet/go-llama.cpp ](https://github.com/go-skynet/go-llama.cpp )
2023-10-22 18:16:43 +00:00
- Node.js: [withcatai/node-llama-cpp ](https://github.com/withcatai/node-llama-cpp )
2024-01-07 20:24:11 +00:00
- JS/TS (llama.cpp server client): [lgrammel/modelfusion ](https://modelfusion.dev/integration/model-provider/llamacpp )
2024-09-23 15:58:17 +00:00
- JS/TS (Programmable Prompt Engine CLI): [offline-ai/cli ](https://github.com/offline-ai/cli )
2024-02-09 10:17:00 +00:00
- JavaScript/Wasm (works in browser): [tangledgroup/llama-cpp-wasm ](https://github.com/tangledgroup/llama-cpp-wasm )
2024-03-16 15:42:08 +00:00
- Typescript/Wasm (nicer API, available on npm): [ngxson/wllama ](https://github.com/ngxson/wllama )
2023-04-17 19:34:35 +00:00
- Ruby: [yoshoku/llama_cpp.rb ](https://github.com/yoshoku/llama_cpp.rb )
2024-04-03 17:53:37 +00:00
- Rust (more features): [edgenai/llama_cpp-rs ](https://github.com/edgenai/llama_cpp-rs )
2024-01-28 08:30:44 +00:00
- Rust (nicer API): [mdrokz/rust-llama.cpp ](https://github.com/mdrokz/rust-llama.cpp )
- Rust (more direct bindings): [utilityai/llama-cpp-rs ](https://github.com/utilityai/llama-cpp-rs )
2025-02-04 11:20:55 +00:00
- Rust (automated build from crates.io): [ShelbyJenkins/llm_client ](https://github.com/ShelbyJenkins/llm_client )
2023-05-12 05:39:40 +00:00
- C#/.NET: [SciSharp/LLamaSharp ](https://github.com/SciSharp/LLamaSharp )
2024-10-20 16:25:41 +00:00
- C#/VB.NET (more features - community license): [LM-Kit.NET ](https://docs.lm-kit.com/lm-kit-net/index.html )
2023-06-26 19:47:59 +00:00
- Scala 3: [donderom/llm4s ](https://github.com/donderom/llm4s )
2023-08-18 19:39:22 +00:00
- Clojure: [phronmophobic/llama.clj ](https://github.com/phronmophobic/llama.clj )
2023-08-29 09:30:10 +00:00
- React Native: [mybigday/llama.rn ](https://github.com/mybigday/llama.rn )
2023-09-01 13:36:14 +00:00
- Java: [kherud/java-llama.cpp ](https://github.com/kherud/java-llama.cpp )
2025-09-25 15:20:34 +00:00
- Java: [QuasarByte/llama-cpp-jna ](https://github.com/QuasarByte/llama-cpp-jna )
2023-12-22 06:49:54 +00:00
- Zig: [deins/llama.cpp.zig ](https://github.com/Deins/llama.cpp.zig )
2024-01-20 08:05:43 +00:00
- Flutter/Dart: [netdur/llama_cpp_dart ](https://github.com/netdur/llama_cpp_dart )
2024-11-13 11:17:10 +00:00
- Flutter: [xuegao-tzx/Fllama ](https://github.com/xuegao-tzx/Fllama )
2025-02-15 14:40:57 +00:00
- PHP (API bindings and features built on top of llama.cpp): [distantmagic/resonance ](https://github.com/distantmagic/resonance ) [(more info) ](https://github.com/ggml-org/llama.cpp/pull/6326 )
2024-07-07 13:21:37 +00:00
- Guile Scheme: [guile_llama_cpp ](https://savannah.nongnu.org/projects/guile-llama-cpp )
2024-10-15 08:20:34 +00:00
- Swift [srgtuszy/llama-cpp-swift ](https://github.com/srgtuszy/llama-cpp-swift )
2024-10-17 06:57:14 +00:00
- Swift [ShenghaiWang/SwiftLlama ](https://github.com/ShenghaiWang/SwiftLlama )
2025-03-06 19:15:13 +00:00
- Delphi [Embarcadero/llama-cpp-delphi ](https://github.com/Embarcadero/llama-cpp-delphi )
2025-10-20 08:20:04 +00:00
- Go (no CGo needed): [hybridgroup/yzma ](https://github.com/hybridgroup/yzma )
2025-12-17 08:14:47 +00:00
- Android: [llama.android ](/examples/llama.android )
2023-04-05 15:56:20 +00:00
2024-11-30 07:47:07 +00:00
< / details >
< details >
< summary > UIs< / summary >
*(to have a project listed here, it should clearly state that it depends on `llama.cpp` )*
2023-04-05 15:56:20 +00:00
2024-11-29 16:45:08 +00:00
- [AI Sublime Text plugin ](https://github.com/yaroslavyaroslav/OpenAI-sublime-text ) (MIT)
2026-01-11 11:46:50 +00:00
- [BonzAI App ](https://apps.apple.com/us/app/bonzai-your-local-ai-agent/id6752847988 ) (proprietary)
2024-11-29 16:45:08 +00:00
- [cztomsik/ava ](https://github.com/cztomsik/ava ) (MIT)
- [Dot ](https://github.com/alexpinel/Dot ) (GPL)
- [eva ](https://github.com/ylsdamxssjxxdd/eva ) (MIT)
- [iohub/collama ](https://github.com/iohub/coLLaMA ) (Apache-2.0)
2024-02-05 14:55:10 +00:00
- [janhq/jan ](https://github.com/janhq/jan ) (AGPL)
2025-03-10 14:13:09 +00:00
- [johnbean393/Sidekick ](https://github.com/johnbean393/Sidekick ) (MIT)
2024-11-29 16:45:08 +00:00
- [KanTV ](https://github.com/zhouwg/kantv?tab=readme-ov-file ) (Apache-2.0)
- [KodiBot ](https://github.com/firatkiral/kodibot ) (GPL)
2024-11-30 07:47:07 +00:00
- [llama.vim ](https://github.com/ggml-org/llama.vim ) (MIT)
2024-11-29 16:45:08 +00:00
- [LARS ](https://github.com/abgulati/LARS ) (AGPL)
- [Llama Assistant ](https://github.com/vietanhdev/llama-assistant ) (GPL)
2026-01-30 06:54:28 +00:00
- [LlamaLib ](https://github.com/undreamai/LlamaLib ) (Apache-2.0)
2024-11-29 16:45:08 +00:00
- [LLMFarm ](https://github.com/guinmoon/LLMFarm?tab=readme-ov-file ) (MIT)
- [LLMUnity ](https://github.com/undreamai/LLMUnity ) (MIT)
2024-02-05 14:55:10 +00:00
- [LMStudio ](https://lmstudio.ai/ ) (proprietary)
2024-02-21 14:39:10 +00:00
- [LocalAI ](https://github.com/mudler/LocalAI ) (MIT)
2024-02-05 14:55:10 +00:00
- [LostRuins/koboldcpp ](https://github.com/LostRuins/koboldcpp ) (AGPL)
2024-11-29 16:45:08 +00:00
- [MindMac ](https://mindmac.app ) (proprietary)
- [MindWorkAI/AI-Studio ](https://github.com/MindWorkAI/AI-Studio ) (FSL-1.1-MIT)
- [Mobile-Artificial-Intelligence/maid ](https://github.com/Mobile-Artificial-Intelligence/maid ) (MIT)
- [Mozilla-Ocho/llamafile ](https://github.com/Mozilla-Ocho/llamafile ) (Apache-2.0)
- [nat/openplayground ](https://github.com/nat/openplayground ) (MIT)
- [nomic-ai/gpt4all ](https://github.com/nomic-ai/gpt4all ) (MIT)
- [ollama/ollama ](https://github.com/ollama/ollama ) (MIT)
2024-02-05 14:55:10 +00:00
- [oobabooga/text-generation-webui ](https://github.com/oobabooga/text-generation-webui ) (AGPL)
2024-11-29 16:45:08 +00:00
- [PocketPal AI ](https://github.com/a-ghorbani/pocketpal-ai ) (MIT)
- [psugihara/FreeChat ](https://github.com/psugihara/FreeChat ) (MIT)
- [ptsochantaris/emeltal ](https://github.com/ptsochantaris/emeltal ) (MIT)
2024-02-05 14:55:10 +00:00
- [pythops/tenere ](https://github.com/pythops/tenere ) (AGPL)
2024-11-29 16:45:08 +00:00
- [ramalama ](https://github.com/containers/ramalama ) (MIT)
- [semperai/amica ](https://github.com/semperai/amica ) (MIT)
- [withcatai/catai ](https://github.com/withcatai/catai ) (MIT)
2025-02-06 00:55:25 +00:00
- [Autopen ](https://github.com/blackhole89/autopen ) (GPL)
2024-04-17 12:47:50 +00:00
2024-11-30 07:47:07 +00:00
< / details >
2024-03-28 20:56:03 +00:00
2024-11-30 07:47:07 +00:00
< details >
< summary > Tools< / summary >
2024-05-26 12:09:42 +00:00
2026-03-24 09:35:07 +00:00
- [akx/ggify ](https://github.com/akx/ggify ) –
2024-09-28 12:07:14 +00:00
- [akx/ollama-dl ](https://github.com/akx/ollama-dl ) –
2024-07-01 16:48:34 +00:00
- [crashr/gppm ](https://github.com/crashr/gppm ) –
2024-08-12 12:45:50 +00:00
- [gpustack/gguf-parser ](https://github.com/gpustack/gguf-parser-go/tree/main/cmd/gguf-parser ) - review/check the GGUF file and estimate the memory usage
2024-11-30 07:47:07 +00:00
- [Styled Lines ](https://marketplace.unity.com/packages/tools/generative-ai/styled-lines-llama-cpp-model-292902 ) (proprietary licensed, async wrapper of inference part for game development in Unity3d with pre-built Mobile and Web platform wrappers and a model example)
2025-11-20 19:07:36 +00:00
- [unslothai/unsloth ](https://github.com/unslothai/unsloth ) –
2024-05-26 12:09:42 +00:00
2024-11-30 07:47:07 +00:00
< / details >
< details >
< summary > Infrastructure< / summary >
2024-07-01 17:13:22 +00:00
2025-08-11 12:27:54 +00:00
- [Paddler ](https://github.com/intentee/paddler ) - Open-source LLMOps platform for hosting and scaling AI in your own infrastructure
2024-08-12 12:45:50 +00:00
- [GPUStack ](https://github.com/gpustack/gpustack ) - Manage GPU clusters for running LLMs
2024-10-20 16:01:34 +00:00
- [llama_cpp_canister ](https://github.com/onicai/llama_cpp_canister ) - llama.cpp as a smart contract on the Internet Computer, using WebAssembly
2025-01-02 07:14:54 +00:00
- [llama-swap ](https://github.com/mostlygeek/llama-swap ) - transparent proxy that adds automatic model switching with llama-server
2025-01-17 00:10:49 +00:00
- [Kalavai ](https://github.com/kalavai-net/kalavai-client ) - Crowdsource end to end LLM deployment at any scale
2025-02-26 07:49:36 +00:00
- [llmaz ](https://github.com/InftyAI/llmaz ) - ☸️ Easy, advanced inference platform for large language models on Kubernetes.
2026-03-08 10:42:28 +00:00
- [LLMKube ](https://github.com/defilantech/llmkube ) - Kubernetes operator for llama.cpp with multi-GPU and Apple Silicon Metal
support"
2024-11-30 07:47:07 +00:00
< / details >
2023-03-10 19:47:46 +00:00
2024-07-05 16:08:32 +00:00
< details >
2024-11-30 07:47:07 +00:00
< summary > Games< / summary >
2023-03-10 19:47:46 +00:00
2024-11-30 07:47:07 +00:00
- [Lucy's Labyrinth ](https://github.com/MorganRO8/Lucys_Labyrinth ) - A simple maze game where agents controlled by an AI model will try to trick you.
2023-03-10 19:47:46 +00:00
2024-07-05 16:08:32 +00:00
< / details >
2025-06-03 11:09:36 +00:00
2024-11-30 07:47:07 +00:00
## Supported backends
| Backend | Target devices |
| --- | --- |
2024-12-01 09:25:17 +00:00
| [Metal ](docs/build.md#metal-build ) | Apple Silicon |
| [BLAS ](docs/build.md#blas-build ) | All |
| [BLIS ](docs/backend/BLIS.md ) | All |
| [SYCL ](docs/backend/SYCL.md ) | Intel and Nvidia GPU |
ggml : add OpenVINO backend (#15307)
* Update build doc
* Add cgraph tensor output name to OV op name
* Update openvino build instructions
* Add initial NPU support
* draft NPU support version 2: prefill + kvcache
* NPU support version 2: prefill + kvcache
* Change due to ggml cgraph changes, not correct yet
* Change due to ggml cgraph changes, llama-3.2 CPU work
* Add AMD64 to CMakeLists
* Change due to ggml cgraph changes, all device work
* Refactor: clean, fix warning
* Update clang-format
* Statful transformation for CPU GPU
* Add SwiGLU
* Fuse to SDPA
* Replace Concat with Broadcast in MulMat for GQA
* Pull out indices creation for kv cache update
* Refactor: remove past_token_len from extra_inputs
* Fix Phi3 SwiGLU and SoftMax
* Pull out sin cos from rope
* Reduce memory: free ov weights node after graph conversion
* Fix CPY due to cgraph change
* Added OpenVINO CI/CD. Updated docs
* Fix llama-cli
* Fix Phi3 ROPE; Add test-backend-ops
* Fix NPU
* Fix llama-bench; Clang-format
* Fix llama-perplexity
* temp. changes for mark decomp
* matmul in fp32
* mulmat input conversion fix
* mulmat type conversion update
* add mark decomp pass
* Revert changes in fuse_to_sdpa
* Update build.md
* Fix test-backend-ops
* Skip test-thread-safety; Run ctest only in ci/run.sh
* Use CiD for NPU
* Optimize tensor conversion, improve TTFT
* Support op SET_ROWS
* Fix NPU
* Remove CPY
* Fix test-backend-ops
* Minor updates for raising PR
* Perf: RMS fused to OV internal RMS op
* Fix after rebasing
- Layout of cache k and cache v are unified: [seq, n_head, head_size]
- Add CPY and FLASH_ATTN_EXT, flash attn is not used yet
- Skip test-backend-ops due to flash attn test crash
- Add mutex around graph conversion to avoid test-thread-safety fali in the future
- Update NPU config
- Update GPU config to disable SDPA opt to make phi-3 run
* Change openvino device_type to GPU; Enable flash_attn
* Update supports_buft and supports_op for quantized models
* Add quant weight conversion functions from genai gguf reader
* Quant models run with accuracy issue
* Fix accuracy: disable cpu_repack
* Fix CI; Disable test-backend-ops
* Fix Q4_1
* Fix test-backend-ops: Treat quantized tensors as weights
* Add NPU Q4_0 support
* NPU perf: eliminate zp
* Dequantize q4_1 q4_k q6_k for NPU
* Add custom quant type: q8_1_c, q4_0_128
* Set m_is_static=false as default in decoder
* Simpilfy translation of get_rows
* Fix after rebasing
* Improve debug util; Eliminate nop ReshapeReshape
* STYLE: make get_types_to_requant a function
* Support BF16 model
* Fix NPU compile
* WA for npu 1st token acc issue
* Apply EliminateZP only for npu
* Add GeGLU
* Fix Hunyuan
* Support iSWA
* Fix NPU accuracy
* Fix ROPE accuracy when freq_scale != 1
* Minor: not add attention_size_swa for non-swa model
* Minor refactor
* Add Q5_K to support phi-3-q4_k_m
* Requantize Q6_K (gs16) to gs32 on GPU
* Fix after rebasing
* Always apply Eliminate_ZP to fix GPU compile issue on some platforms
* kvcachefusion support
* env variable GGML_OPENVINO_DISABLE_SDPA_OPTIMIZATION added
* Fix for Phi3
* Fix llama-cli (need to run with --no-warmup)
* Fix add_sliced_mask; Revert mulmat, softmax; Remove input attention_size, iSWA model not working
* fix after rebasing
* Fix llama-3-8b and phi3-mini q4_0 NPU
* Update to OV-2025.3 and CMakeLists.txt
* Add OV CI cache
* Apply CISC review and update CI to OV2025.3
* Update CI to run OV dep install before build
* Update OV dockerfile to use OV2025.3 and update build docs
* Style: use switch in supports_ops
* Style: middle ptr and ref align, omit optional struct keyword
* NPU Unify PD (#14)
* Stateless. Fix llama-cli llama-server
* Simplify broadcast op in attention
* Replace get_output_tensor+memcpy with set_output_tensor
* NPU unify PD. Unify dynamic and static dims
* Clean placeholders in ggml-openvino.cpp
* NPU unify PD (handled internally)
* change graph to 4d, support multi sequences
* Fix llama-bench
* Fix NPU
* Update ggml-decoder.cpp
Hitting error while compiling on windows:
error C3861: 'unsetenv': identifier not found
Reason: unsetenv() is a POSIX function; it doesn’t exist on Windows. Visual Studio (MSVC) won’t recognize it.
Proposed fix: Use _putenv_s() (Windows equivalent)
This is supported by MSVC and achieves the same effect: it removes the environment variable from the process environment.
This keeps cross-platform compatibility.
* Update ggml-decoder.cpp
* Update ggml-decoder.cpp
* Update ggml-decoder.cpp
* Update ggml-decoder.cpp
* Update ggml-decoder.cpp
* Remove the second decoder for node. Moving the function into the model decoder
* Fix error for naive
* NPU prefill chunking
* NPU fix llama-bench
* fallback naive run with accuracy issue
* NPU support llma-perplexity -b 512 --no-warmup
* Refactor: split ov_graph_compute for dynamic and static
* remove unused API GgmlOvDecoder::get_output_stride(const std::string & name)
* minor update due to ov 2025.4
* remove unused API GgmlOvDecoder::get_output_names()
* remove unused API get_output_shape(const std::string & name)
* Modified API GgmlOvDecoder::get_output_type(const std::string & name)
* Removed API GgmlOvDecoder::get_output_op_params(const std::string & name)
* Removed API get_output_ggml_tensor(const std::string & name)
* Removed API m_outputs
* Removed m_output_names
* Removed API GgmlOvDecoder::get_input_names()
* Removed API GgmlOvDecoder::get_input_stride(const std::string& name)
* Removed API get_input_type
* Removed API get_input_type
* Removed API GgmlOvDecoder::get_input_shape(const std::string & name)
* Removed API GgmlOvDecoder::get_input_op_params(const std::string & name)
* Fix error for decoder cache
* Reuse cached decoder
* GPU remove Q6_K requantization
* NPU fix wrong model output shape
* NPU fix q4 perf regression
* Remove unused variable nodes
* Fix decoder can_reuse for llama-bench
* Update build.md for Windows
* backend buffer: allocate on host
* Use shared_buffer for GPU NPU; Refactor
* Add ov_backend_host_buffer; Use cached remote context
* Put kvcache on GPU
* Use ggml_aligned_malloc
* only use remote tensor for kvcache
* only use remote tensor for kvcache for GPU
* FIX: use remote tensor from singleton
* Update build.md to include OpenCL
* NPU always requant to q4_0_128
* Optimize symmetric quant weight extraction: use single zp
* Use Q8_0_C in token embd, lm_head, and for 5 and 6 bits quant
* Update build.md
* Support -ctk f32
* Initial stateful graph support
* Update ggml/src/ggml-openvino/ggml-decoder.cpp
Co-authored-by: Yamini Nimmagadda <yamini.nimmagadda@intel.com>
* code cleanup
* npu perf fix
* requant to f16 for Q6 embed on NPU
* Update ggml/src/ggml-openvino/ggml-decoder.cpp
* Update ggml/src/ggml-openvino/ggml-openvino-extra.cpp
* Create OPENVINO.md in llama.cpp backend docs
* Update OPENVINO.md
* Update OPENVINO.md
* Update OPENVINO.md
* Update build.md
* Update OPENVINO.md
* Update OPENVINO.md
* Update OPENVINO.md
* kq_mask naming fix
* Syntax correction for workflows build file
* Change ov backend buffer is_host to false
* Fix llama-bench -p -n where p<=256
* Fix --direct-io 0
* Don't put kvcache on GPU in stateful mode
* Remove hardcode names
* Fix stateful shapes
* Simplification for stateful and update output shape processing
* Remove hardcode names
* Avoid re-compilation in llama-bench
* Extract zp directly instead of bias
* Refactor weight tensor processing
* create_weight_node accept non-ov backend buffer
* remove changes in llama-graph.cpp
* stateful masking fix (#38)
Fix for stateful accuracy issues and cl_out_of_resources error in stateful GPU with larger context sizes.
* Fix test-backend-ops crash glu, get_rows, scale, rms_norm, add
* hardcoded name handling for rope_freqs.weight
* Suppress logging and add error handling to allow test-backend-ops to complete
* Fix MUL_MAT with broadcast; Add unsupported MUL_MAT FLASH_ATTN cases
* Use bias instead of zp in test-backend-ops
* Update OV in CI, Add OV CI Tests in GH Actions
* Temp fix for multithreading bug
* Update OV CI, fix review suggestions.
* fix editorconfig-checker, update docs
* Fix tabs to spaces for editorconfig-checker
* fix editorconfig-checker
* Update docs
* updated model link to be GGUF model links
* Remove GGML_CPU_REPACK=OFF
* Skip permuted ADD and MUL
* Removed static variables from utils.cpp
* Removed initializing non-existing variable
* Remove unused structs
* Fix test-backend-ops for OV GPU
* unify api calling
* Update utils.cpp
* When the dim is dynamic, throw an error, need to is stastic forst
* Add interface compute_model_outputs(), which get the model output through computing the node use count & status in the cgraph to avoid the flag using
* No need to return
* Fix test-backend-ops for OV GPU LNL
* Fix test-thread-safety
* use the shape from infer request of output tensor create to avoid issue
* fix dynamic output shape issue
* fix issue for the unused node in tests
* Remove unused lock
* Add comment
* Update openvino docs
* update to OV release version 2026.0
* add ci ov-gpu self hosted runner
* fix editorconfig
* Fix perplexity
* Rewrite the model inputs finding mechanism (#54)
* Rewrite the model inputs finding logistic
* Put stateful shape handle in get input shape
* Put the iteration logistic in func
* Added ggml-ci-intel-openvino-gpu and doc update
* .hpp files converted to .h
* fix ggml-ci-x64-intel-openvino-gpu
* Fix for stateful execution bug in llama-bench
* Minor updates after stateful llama-bench fix
* Update ggml/src/ggml-openvino/utils.cpp
Co-authored-by: Yamini Nimmagadda <yamini.nimmagadda@intel.com>
* Remove multiple get_shape calls
* Bring back mutex into compute
* Fix VIEW op, which slice the input node
* Added token_len_per_seq existence check before slicing masks and moved node retrieval inside guarded block to prevent missing-key access
* Temp. fix for test requant errors
* Update to OV ggml-ci to low-perf
* ci : temporary disable "test-llama-archs"
* ci : cache v4 -> v5, checkout v4 -> v6, fix runner tag
* docs : update url
* Fix OV link in docker and Update docs
---------
Co-authored-by: Ravi Panchumarthy <ravi.panchumarthy@intel.com>
Co-authored-by: Cavus Mustafa <mustafa.cavus@intel.com>
Co-authored-by: Arshath <arshath.ramzan@intel.com>
Co-authored-by: XuejunZhai <Xuejun.Zhai@intel.com>
Co-authored-by: Yamini Nimmagadda <yamini.nimmagadda@intel.com>
Co-authored-by: Xuejun Zhai <Xuejun.Zhai@intel>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-03-14 05:56:55 +00:00
| [OpenVINO [In Progress]](docs/backend/OPENVINO.md) | Intel CPUs, GPUs, and NPUs |
2025-05-21 01:58:49 +00:00
| [MUSA ](docs/build.md#musa ) | Moore Threads GPU |
2024-12-01 09:25:17 +00:00
| [CUDA ](docs/build.md#cuda ) | Nvidia GPU |
2024-12-18 08:35:00 +00:00
| [HIP ](docs/build.md#hip ) | AMD GPU |
2025-12-06 16:13:33 +00:00
| [ZenDNN ](docs/build.md#zendnn ) | AMD CPU |
2024-12-01 09:25:17 +00:00
| [Vulkan ](docs/build.md#vulkan ) | GPU |
| [CANN ](docs/build.md#cann ) | Ascend NPU |
2025-02-11 22:04:13 +00:00
| [OpenCL ](docs/backend/OPENCL.md ) | Adreno GPU |
2025-09-23 06:53:05 +00:00
| [IBM zDNN ](docs/backend/zDNN.md ) | IBM Z & LinuxONE |
2025-07-16 15:18:51 +00:00
| [WebGPU [In Progress]](docs/build.md#webgpu) | All |
2025-05-02 18:27:13 +00:00
| [RPC ](https://github.com/ggml-org/llama.cpp/tree/master/tools/rpc ) | All |
2026-03-06 02:29:13 +00:00
| [Hexagon [In Progress]](docs/backend/snapdragon/README.md) | Snapdragon |
2026-02-09 12:15:42 +00:00
| [VirtGPU ](docs/backend/VirtGPU.md ) | VirtGPU APIR |
2024-12-01 09:25:17 +00:00
## Obtaining and quantizing models
2024-11-30 07:47:07 +00:00
The [Hugging Face ](https://huggingface.co ) platform hosts a [number of LLMs ](https://huggingface.co/models?library=gguf&sort=trending ) compatible with `llama.cpp` :
- [Trending ](https://huggingface.co/models?library=gguf&sort=trending )
- [LLaMA ](https://huggingface.co/models?sort=trending&search=llama+gguf )
2024-07-05 16:08:32 +00:00
2026-03-24 09:35:07 +00:00
You can either manually download the GGUF file or directly use any `llama.cpp` -compatible models from [Hugging Face ](https://huggingface.co/ ) or other model hosting sites, by using this CLI argument: `-hf <user>/<model>[:quant]` . For example:
2025-06-03 11:09:36 +00:00
```sh
llama-cli -hf ggml-org/gemma-3-1b-it-GGUF
```
2025-04-11 12:01:56 +00:00
2026-03-24 09:35:07 +00:00
By default, the CLI would download from Hugging Face, you can switch to other options with the environment variable `MODEL_ENDPOINT` . The `MODEL_ENDPOINT` must point to a Hugging Face compatible API endpoint.
2025-01-13 19:18:12 +00:00
2024-11-30 07:47:07 +00:00
After downloading a model, use the CLI tools to run it locally - see below.
2023-03-10 19:47:46 +00:00
2025-02-15 14:40:57 +00:00
`llama.cpp` requires the model to be stored in the [GGUF ](https://github.com/ggml-org/ggml/blob/master/docs/gguf.md ) file format. Models in other data formats can be converted to GGUF using the `convert_*.py` Python scripts in this repo.
2023-04-13 13:43:22 +00:00
2024-11-30 07:47:07 +00:00
The Hugging Face platform provides a variety of online tools for converting, quantizing and hosting models with `llama.cpp` :
2023-04-26 20:03:03 +00:00
2024-11-30 07:47:07 +00:00
- Use the [GGUF-my-repo space ](https://huggingface.co/spaces/ggml-org/gguf-my-repo ) to convert to GGUF format and quantize model weights to smaller sizes
2025-02-15 14:40:57 +00:00
- Use the [GGUF-my-LoRA space ](https://huggingface.co/spaces/ggml-org/gguf-my-lora ) to convert LoRA adapters to GGUF format (more info: https://github.com/ggml-org/llama.cpp/discussions/10123)
- Use the [GGUF-editor space ](https://huggingface.co/spaces/CISCai/gguf-editor ) to edit GGUF meta data in the browser (more info: https://github.com/ggml-org/llama.cpp/discussions/9268)
- Use the [Inference Endpoints ](https://ui.endpoints.huggingface.co/ ) to directly host `llama.cpp` in the cloud (more info: https://github.com/ggml-org/llama.cpp/discussions/9669)
2023-04-26 20:03:03 +00:00
2025-05-02 18:27:13 +00:00
To learn more about model quantization, [read this documentation ](tools/quantize/README.md )
2024-11-30 07:47:07 +00:00
2025-12-15 10:47:04 +00:00
## [`llama-cli`](tools/cli)
2024-11-30 07:47:07 +00:00
2024-12-01 09:25:17 +00:00
#### A CLI tool for accessing and experimenting with most of `llama.cpp`'s functionality.
2023-03-10 19:47:46 +00:00
2024-12-01 09:25:17 +00:00
- < details open >
< summary > Run in conversation mode< / summary >
2024-06-17 15:37:55 +00:00
2025-01-13 19:18:12 +00:00
Models with a built-in chat template will automatically activate conversation mode. If this doesn't occur, you can manually enable it by adding `-cnv` and specifying a suitable chat template with `--chat-template NAME`
2024-12-01 09:25:17 +00:00
```bash
2025-01-13 19:18:12 +00:00
llama-cli -m model.gguf
2024-06-17 15:37:55 +00:00
2024-12-01 09:25:17 +00:00
# > hi, who are you?
# Hi there! I'm your helpful assistant! I'm an AI-powered chatbot designed to assist and provide information to users like you. I'm here to help answer your questions, provide guidance, and offer support on a wide range of topics. I'm a friendly and knowledgeable AI, and I'm always happy to help with anything you need. What's on your mind, and how can I assist you today?
#
# > what is 1+1?
# Easy peasy! The answer to 1+1 is... 2!
```
2023-05-20 14:58:31 +00:00
2024-12-01 09:25:17 +00:00
< / details >
2024-05-05 05:21:46 +00:00
2024-12-01 09:25:17 +00:00
- < details >
2025-01-13 19:18:12 +00:00
< summary > Run in conversation mode with custom chat template< / summary >
2023-09-21 19:00:24 +00:00
2024-12-01 09:25:17 +00:00
```bash
2025-01-13 19:18:12 +00:00
# use the "chatml" template (use -h to see the list of supported templates)
llama-cli -m model.gguf -cnv --chat-template chatml
2023-09-21 19:00:24 +00:00
2024-12-01 09:25:17 +00:00
# use a custom template
2025-01-13 19:18:12 +00:00
llama-cli -m model.gguf -cnv --in-prefix 'User: ' --reverse-prompt 'User:'
2024-12-01 09:25:17 +00:00
```
2023-05-24 06:24:01 +00:00
2025-01-13 19:18:12 +00:00
< / details >
2024-12-01 09:25:17 +00:00
- < details >
< summary > Constrain the output with a custom grammar< / summary >
2023-08-23 01:01:57 +00:00
2024-12-01 09:25:17 +00:00
```bash
llama-cli -m model.gguf -n 256 --grammar-file grammars/json.gbnf -p 'Request: schedule a call at 8pm; Command:'
2023-08-23 01:01:57 +00:00
2024-12-01 09:25:17 +00:00
# {"appointmentTime": "8pm", "appointmentDetails": "schedule a a call"}
```
2023-07-28 01:14:11 +00:00
2024-12-01 09:25:17 +00:00
The [grammars/ ](grammars/ ) folder contains a handful of sample grammars. To write your own, check out the [GBNF Guide ](grammars/README.md ).
2023-07-28 01:14:11 +00:00
2024-12-01 09:25:17 +00:00
For authoring more complex JSON grammars, check out https://grammar.intrinsiclabs.ai/
2023-03-20 20:14:06 +00:00
2024-12-01 09:25:17 +00:00
< / details >
2023-07-07 18:25:25 +00:00
2025-05-02 18:27:13 +00:00
## [`llama-server`](tools/server)
2023-07-07 18:25:25 +00:00
2024-12-01 09:25:17 +00:00
#### A lightweight, [OpenAI API](https://github.com/openai/openai-openapi) compatible, HTTP server for serving LLMs.
2023-07-07 18:25:25 +00:00
2024-12-01 09:25:17 +00:00
- < details open >
< summary > Start a local HTTP server with default configuration on port 8080< / summary >
```bash
llama-server -m model.gguf --port 8080
# Basic web UI can be accessed via browser: http://localhost:8080
# Chat completion endpoint: http://localhost:8080/v1/chat/completions
```
< / details >
- < details >
< summary > Support multiple-users and parallel decoding< / summary >
```bash
# up to 4 concurrent requests, each with 4096 max context
llama-server -m model.gguf -c 16384 -np 4
```
< / details >
- < details >
< summary > Enable speculative decoding< / summary >
```bash
# the draft.gguf model should be a small variant of the target model.gguf
llama-server -m model.gguf -md draft.gguf
```
< / details >
- < details >
< summary > Serve an embedding model< / summary >
```bash
# use the /embedding endpoint
llama-server -m model.gguf --embedding --pooling cls -ub 8192
```
< / details >
- < details >
< summary > Serve a reranking model< / summary >
```bash
# use the /reranking endpoint
llama-server -m model.gguf --reranking
```
< / details >
- < details >
< summary > Constrain all outputs with a grammar< / summary >
```bash
# custom grammar
llama-server -m model.gguf --grammar-file grammar.gbnf
# JSON
llama-server -m model.gguf --grammar-file grammars/json.gbnf
```
< / details >
2025-05-02 18:27:13 +00:00
## [`llama-perplexity`](tools/perplexity)
2024-12-01 09:25:17 +00:00
2025-07-20 18:13:47 +00:00
#### A tool for measuring the [perplexity](tools/perplexity/README.md) [^1] (and other quality metrics) of a model over a given text.
2024-12-01 09:25:17 +00:00
- < details open >
< summary > Measure the perplexity over a text file< / summary >
```bash
llama-perplexity -m model.gguf -f file.txt
# [1]15.2701,[2]5.4007,[3]5.3073,[4]6.2965,[5]5.8940,[6]5.6096,[7]5.7942,[8]4.9297, ...
# Final estimate: PPL = 5.4007 +/- 0.67339
```
< / details >
- < details >
< summary > Measure KL divergence< / summary >
```bash
# TODO
```
< / details >
2025-07-20 18:13:47 +00:00
[^1]: [https://huggingface.co/docs/transformers/perplexity ](https://huggingface.co/docs/transformers/perplexity )
2024-12-01 09:25:17 +00:00
2025-05-02 18:27:13 +00:00
## [`llama-bench`](tools/llama-bench)
2024-12-01 09:25:17 +00:00
#### Benchmark the performance of the inference for various parameters.
- < details open >
< summary > Run default benchmark< / summary >
```bash
llama-bench -m model.gguf
# Output:
# | model | size | params | backend | threads | test | t/s |
# | ------------------- | ---------: | ---------: | ---------- | ------: | ------------: | -------------------: |
# | qwen2 1.5B Q4_0 | 885.97 MiB | 1.54 B | Metal,BLAS | 16 | pp512 | 5765.41 ± 20.55 |
# | qwen2 1.5B Q4_0 | 885.97 MiB | 1.54 B | Metal,BLAS | 16 | tg128 | 197.71 ± 0.81 |
#
# build: 3e0ba0e60 (4229)
```
< / details >
## [`llama-simple`](examples/simple)
#### A minimal example for implementing apps with `llama.cpp`. Useful for developers.
- < details >
< summary > Basic text completion< / summary >
```bash
llama-simple -m model.gguf
# Hello my name is Kaitlyn and I am a 16 year old girl. I am a junior in high school and I am currently taking a class called "The Art of
```
< / details >
2023-07-07 18:25:25 +00:00
2024-07-05 16:08:32 +00:00
## Contributing
2023-03-13 07:42:26 +00:00
2023-03-13 17:21:51 +00:00
- Contributors can open PRs
2023-03-13 07:42:26 +00:00
- Collaborators will be invited based on contributions
2025-09-22 07:58:02 +00:00
- Maintainers can push to branches in the `llama.cpp` repo and merge PRs into the `master` branch
2024-09-29 11:38:18 +00:00
- Any help with managing issues, PRs and projects is very appreciated!
2025-02-15 14:40:57 +00:00
- See [good first issues ](https://github.com/ggml-org/llama.cpp/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22 ) for tasks suitable for first contributions
2024-07-05 06:09:47 +00:00
- Read the [CONTRIBUTING.md ](CONTRIBUTING.md ) for more information
2025-02-15 14:40:57 +00:00
- Make sure to read this: [Inference at the edge ](https://github.com/ggml-org/llama.cpp/discussions/205 )
2023-03-23 08:46:58 +00:00
- A bit of backstory for those who are interested: [Changelog podcast ](https://changelog.com/podcast/532 )
2023-03-13 07:42:26 +00:00
2024-11-17 08:06:34 +00:00
## Other documentation
2023-04-05 15:56:20 +00:00
2025-12-15 10:47:04 +00:00
- [cli ](tools/cli/README.md )
- [completion ](tools/completion/README.md )
2025-05-02 18:27:13 +00:00
- [server ](tools/server/README.md )
2024-12-01 09:25:17 +00:00
- [GBNF grammars ](grammars/README.md )
2024-07-05 16:08:32 +00:00
2024-12-01 09:25:17 +00:00
#### Development documentation
2024-07-05 16:08:32 +00:00
2024-12-01 09:25:17 +00:00
- [How to build ](docs/build.md )
- [Running on Docker ](docs/docker.md )
- [Build on Android ](docs/android.md )
- [Performance troubleshooting ](docs/development/token_generation_performance_tips.md )
2025-02-15 14:40:57 +00:00
- [GGML tips & tricks ](https://github.com/ggml-org/llama.cpp/wiki/GGML-Tips-&-Tricks )
2024-07-06 17:01:23 +00:00
2024-12-01 09:25:17 +00:00
#### Seminal papers and background on the models
2024-07-06 17:01:23 +00:00
If your issue is with model generation quality, then please at least scan the following links and papers to understand the limitations of LLaMA models. This is especially important when choosing an appropriate model size and appreciating both the significant and subtle differences between LLaMA models and ChatGPT:
- LLaMA:
- [Introducing LLaMA: A foundational, 65-billion-parameter large language model ](https://ai.facebook.com/blog/large-language-model-llama-meta-ai/ )
- [LLaMA: Open and Efficient Foundation Language Models ](https://arxiv.org/abs/2302.13971 )
- GPT-3
- [Language Models are Few-Shot Learners ](https://arxiv.org/abs/2005.14165 )
- GPT-3.5 / InstructGPT / ChatGPT:
- [Aligning language models to follow instructions ](https://openai.com/research/instruction-following )
- [Training language models to follow instructions with human feedback ](https://arxiv.org/abs/2203.02155 )
2024-12-01 09:25:17 +00:00
2025-04-04 08:24:12 +00:00
## XCFramework
The XCFramework is a precompiled version of the library for iOS, visionOS, tvOS,
and macOS. It can be used in Swift projects without the need to compile the
library from source. For example:
```swift
// swift-tools-version: 5.10
// The swift-tools-version declares the minimum version of Swift required to build this package.
import PackageDescription
let package = Package(
name: "MyLlamaPackage",
targets: [
.executableTarget(
name: "MyLlamaPackage",
dependencies: [
"LlamaFramework"
]),
.binaryTarget(
name: "LlamaFramework",
url: "https://github.com/ggml-org/llama.cpp/releases/download/b5046/llama-b5046-xcframework.zip",
checksum: "c19be78b5f00d8d29a25da41042cb7afa094cbf6280a225abe614b03b20029ab"
)
]
)
```
The above example is using an intermediate build `b5046` of the library. This can be modified
to use a different version by changing the URL and checksum.
2025-02-13 22:16:56 +00:00
## Completions
2025-02-13 13:46:59 +00:00
Command-line completion is available for some environments.
#### Bash Completion
```bash
$ build/bin/llama-cli --completion-bash > ~/.llama-completion.bash
$ source ~/.llama-completion.bash
```
Optionally this can be added to your `.bashrc` or `.bash_profile` to load it
automatically. For example:
```console
$ echo "source ~/.llama-completion.bash" >> ~/.bashrc
```
2025-02-13 22:16:56 +00:00
2025-05-16 18:04:18 +00:00
## Dependencies
- [yhirose/cpp-httplib ](https://github.com/yhirose/cpp-httplib ) - Single-header HTTP server, used by `llama-server` - MIT license
- [stb-image ](https://github.com/nothings/stb ) - Single-header image format decoder, used by multimodal subsystem - Public domain
- [nlohmann/json ](https://github.com/nlohmann/json ) - Single-header JSON library, used by various tools/examples - MIT License
2025-05-22 18:42:48 +00:00
- [miniaudio.h ](https://github.com/mackron/miniaudio ) - Single-header audio format decoder, used by multimodal subsystem - Public domain
2025-12-01 18:41:04 +00:00
- [subprocess.h ](https://github.com/sheredom/subprocess.h ) - Single-header process launching solution for C and C++ - Public domain
