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

Merge branch 'master' into mpi

Browse source
This commit is contained in:
Evan Miller 2023-07-06 19:04:50 -04:00
commit 32deabfdc8
33 changed files with 5794 additions and 2048 deletions
Download patch file Download diff file Expand all files Collapse all files

6
README.md
View file

@ -11,6 +11,7 @@ Inference of [LLaMA](https://arxiv.org/abs/2302.13971) model in pure C/C++
**Hot topics:**
- Simple web chat example: https://github.com/ggerganov/llama.cpp/pull/1998
- k-quants now support super-block size of 64: https://github.com/ggerganov/llama.cpp/pull/2001
- New roadmap: https://github.com/users/ggerganov/projects/7
- Azure CI brainstorming: https://github.com/ggerganov/llama.cpp/discussions/1985
@ -85,7 +86,7 @@ as the main playground for developing new features for the [ggml](https://github
- [X] [OpenBuddy 🐶 (Multilingual)](https://github.com/OpenBuddy/OpenBuddy)
- [X] [Pygmalion 7B / Metharme 7B](#using-pygmalion-7b--metharme-7b)
- [X] [WizardLM](https://github.com/nlpxucan/WizardLM)
- [X] [Baichuan-7B](https://huggingface.co/baichuan-inc/baichuan-7B)
- [X] [Baichuan-7B](https://huggingface.co/baichuan-inc/baichuan-7B) and its derivations (such as [baichuan-7b-sft](https://huggingface.co/hiyouga/baichuan-7b-sft))
**Bindings:**
@ -373,8 +374,9 @@ Building the program with BLAS support may lead to some performance improvements
| Option | Legal values | Default | Description |
|-------------------------|------------------------|---------|-------------|
| LLAMA_CUDA_FORCE_DMMV | Boolean | false | Force the use of dequantization + matrix vector multiplication kernels instead of using kernels that do matrix vector multiplication on quantized data. By default the decision is made based on compute capability (MMVQ for 7.0/Turing/RTX 2000 or higher). Does not affect k-quants. |
| LLAMA_CUDA_DMMV_X | Positive integer >= 32 | 32 | Number of values in x direction processed by the CUDA dequantization + matrix vector multiplication kernel per iteration. Increasing this value can improve performance on fast GPUs. Power of 2 heavily recommended. Does not affect k-quants. |
| LLAMA_CUDA_DMMV_Y | Positive integer | 1 | Block size in y direction for the CUDA dequantization + mul mat vec kernels. Increasing this value can improve performance on fast GPUs. Power of 2 recommended. Does not affect k-quants. |
| LLAMA_CUDA_MMV_Y | Positive integer | 1 | Block size in y direction for the CUDA mul mat vec kernels. Increasing this value can improve performance on fast GPUs. Power of 2 recommended. Does not affect k-quants. |
| LLAMA_CUDA_DMMV_F16 | Boolean | false | If enabled, use half-precision floating point arithmetic for the CUDA dequantization + mul mat vec kernels. Can improve performance on relatively recent GPUs. |
| LLAMA_CUDA_KQUANTS_ITER | 1 or 2 | 2 | Number of values processed per iteration and per CUDA thread for Q2_K and Q6_K quantization formats. Setting this value to 1 can improve performance for slow GPUs. |