root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 10:54:36 +00:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

rpc : update README [no ci] (#9320)

Browse Source 
Update README with instructions how to offload model layers to both
local and remote devices
This commit is contained in:
Radoslav Gerganov 2024-09-09 11:04:39 +03:00 committed by GitHub
parent b2e89a3274
commit 54f376d0b9
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 12 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
examples/rpc/README.md
View File

@ -10,20 +10,21 @@ This can be used for distributed LLM inference with `llama.cpp` in the following
```mermaid ```mermaid
flowchart TD flowchart TD
rpcb---|TCP|srva rpcb<-->|TCP|srva
rpcb---|TCP|srvb rpcb<-->|TCP|srvb
rpcb-.-|TCP|srvn rpcb<-.->|TCP|srvn
subgraph hostn[Host N] subgraph hostn[Host N]
srvn[rpc-server]-.-backend3["Backend (CUDA,Metal,etc.)"] srvn[rpc-server]<-.->backend3["Backend (CUDA,Metal,etc.)"]
end end
subgraph hostb[Host B] subgraph hostb[Host B]
srvb[rpc-server]---backend2["Backend (CUDA,Metal,etc.)"] srvb[rpc-server]<-->backend2["Backend (CUDA,Metal,etc.)"]
end end
subgraph hosta[Host A] subgraph hosta[Host A]
srva[rpc-server]---backend["Backend (CUDA,Metal,etc.)"] srva[rpc-server]<-->backend["Backend (CUDA,Metal,etc.)"]
end end
subgraph host[Main Host] subgraph host[Main Host]
ggml[llama.cpp]---rpcb[RPC backend] local["Backend (CUDA,Metal,etc.)"]<-->ggml[llama-cli]
ggml[llama-cli]<-->rpcb[RPC backend]
end end
style hostn stroke:#66,stroke-width:2px,stroke-dasharray: 5 5 style hostn stroke:#66,stroke-width:2px,stroke-dasharray: 5 5
``` ```
@ -62,17 +63,11 @@ $ CUDA_VISIBLE_DEVICES=0 bin/rpc-server -p 50052
This way you can run multiple `rpc-server` instances on the same host, each with a different CUDA device. This way you can run multiple `rpc-server` instances on the same host, each with a different CUDA device.
On the main host build `llama.cpp` only with `-DGGML_RPC=ON`: On the main host build `llama.cpp` for the local backend and add `-DGGML_RPC=ON` to the build options.
Finally, when running `llama-cli`, use the `--rpc` option to specify the host and port of each `rpc-server`:
```bash
mkdir build-rpc
cd build-rpc
cmake .. -DGGML_RPC=ON
cmake --build . --config Release
```
Finally, use the `--rpc` option to specify the host and port of each `rpc-server`:
```bash ```bash
$ bin/llama-cli -m ../models/tinyllama-1b/ggml-model-f16.gguf -p "Hello, my name is" --repeat-penalty 1.0 -n 64 --rpc 192.168.88.10:50052,192.168.88.11:50052 -ngl 99 $ bin/llama-cli -m ../models/tinyllama-1b/ggml-model-f16.gguf -p "Hello, my name is" --repeat-penalty 1.0 -n 64 --rpc 192.168.88.10:50052,192.168.88.11:50052 -ngl 99
``` ```
This way you can offload model layers to both local and remote devices.