Merge pull request #411 from jmorganca/mxyng/34b

patch llama.cpp for 34B

Merge pull request #411 from jmorganca/mxyng/34b
patch llama.cpp for 34B
dad63f08 · Michael Yang · GitHub · de0bdd7f · 7a378f8b · dad63f08
Unverified Commit dad63f08 authored Aug 25, 2023 by Michael Yang Committed by GitHub Aug 25, 2023
Hide whitespace changes
Inline Side-by-side

Showing with 3 additions and 0 deletions

llm/llama.cpp llm/llama.cpp +3 -0

No files found.
--- a/llm/llama.cpp
+++ b/llm/llama.cpp
@@ -105,6 +105,7 @@ enum e_model {
    MODEL_7B,
    MODEL_13B,
    MODEL_30B,
+    MODEL_34B,
    MODEL_65B,
    MODEL_70B,
 };
@@ -1053,6 +1054,7 @@ static const char *llama_model_type_name(e_model type) {
        case MODEL_7B: return "7B";
        case MODEL_13B: return "13B";
        case MODEL_30B: return "30B";
+        case MODEL_34B: return "34B";
        case MODEL_65B: return "65B";
        case MODEL_70B: return "70B";
        default: LLAMA_ASSERT(false);
@@ -1100,6 +1102,7 @@ static void llama_model_load_internal(
            case 26: model.type = e_model::MODEL_3B; break;
            case 32: model.type = e_model::MODEL_7B; break;
            case 40: model.type = e_model::MODEL_13B; break;
+            case 48: model.type = e_model::MODEL_34B; break;
            case 60: model.type = e_model::MODEL_30B; break;
            case 80: model.type = e_model::MODEL_65B; break;
            default: