From d018c7b01dd08518b59ebcedc603111243a39391 Mon Sep 17 00:00:00 2001
From: Billy Cao <aliencaocao@gmail.com>
Date: Sat, 12 Aug 2023 18:41:47 +0800
Subject: [PATCH] Add doc string for n_gpu_layers argument

---
 llama_cpp/llama.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/llama_cpp/llama.py b/llama_cpp/llama.py
index a996d5c..20a5e0c 100644
--- a/llama_cpp/llama.py
+++ b/llama_cpp/llama.py
@@ -239,6 +239,7 @@ class Llama:
             n_ctx: Maximum context size.
             n_parts: Number of parts to split the model into. If -1, the number of parts is automatically determined.
             seed: Random seed. -1 for random.
+            n_gpu_layers: Number of layers to offload to GPU (-ngl). If -1, all layers are offloaded.
             f16_kv: Use half-precision for key/value cache.
             logits_all: Return logits for all tokens, not just the last token.
             vocab_only: Only load the vocabulary no weights.