mmarch/text-generation-webui
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Add tensor split support for llama.cpp (#3171)

Browse source
This commit is contained in:
Shouyi 2023-07-26 07:59:26 +10:00 • committed by GitHub
parent f653546484
commit 031fe7225e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
7 changed files with 20 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

1
README.md
View file

@ -218,6 +218,7 @@ Optionally, you can use the following command-line flags:
| `--mlock` | Force the system to keep the model in RAM. |
| `--cache-capacity CACHE_CAPACITY` | Maximum cache capacity. Examples: 2000MiB, 2GiB. When provided without units, bytes will be assumed. |
| `--n-gpu-layers N_GPU_LAYERS` | Number of layers to offload to the GPU. Only works if llama-cpp-python was compiled with BLAS. Set this to 1000000000 to offload all layers to the GPU. |
| `--tensor_split TENSOR_SPLIT` | Split the model across multiple GPUs, comma-separated list of proportions, e.g. 18,17 |
| `--n_ctx N_CTX` | Size of the prompt context. |
| `--llama_cpp_seed SEED` | Seed for llama-cpp models. Default 0 (random). |
| `--n_gqa N_GQA` | grouped-query attention. Must be 8 for llama2 70b. |