From: Rinne <redacted>
Date: Wed, 19 Jul 2023 07:06:40 +0000 (+0800)
Subject: llama : extend API to get max devices at runtime (#2253)
X-Git-Tag: gguf-v0.4.0~454
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=294f424554c1599784ac9962462fc39ace92d8a5;p=pkg%2Fggml%2Fsources%2Fllama.cpp

llama : extend API to get max devices at runtime (#2253)
---

diff --git a/llama.cpp b/llama.cpp
index fa3b7c03..3319b702 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -875,6 +875,10 @@ struct llama_model_quantize_params llama_model_quantize_default_params() {
     return result;
 }
 
+int llama_max_devices() {
+    return LLAMA_MAX_DEVICES;
+}
+
 bool llama_mmap_supported() {
     return llama_mmap::SUPPORTED;
 }
diff --git a/llama.h b/llama.h
index e744584f..b676a383 100644
--- a/llama.h
+++ b/llama.h
@@ -153,6 +153,8 @@ extern "C" {
         int32_t n_eval;
     };
 
+    LLAMA_API int llama_max_devices();
+
     LLAMA_API struct llama_context_params llama_context_default_params();
     LLAMA_API struct llama_model_quantize_params llama_model_quantize_default_params();