Add Granite arch to model whitelist (#3487)

ThiloteE · cebtenzzre · web-flow · commit 02e12089d36b · 2025-02-12T14:17:49.000-05:00
Signed-off-by: ThiloteE &lt;73715071+ThiloteE@users.noreply.github.com&gt;
Signed-off-by: Jared Van Bortel &lt;jared@nomic.ai&gt;
Co-authored-by: Jared Van Bortel &lt;jared@nomic.ai&gt;
diff --git a/gpt4all-backend/deps/llama.cpp-mainline b/gpt4all-backend/deps/llama.cpp-mainline
@@ -1 +1 @@
-Subproject commit 3ebb3603e807d74a16f061c46d2925a1653e7a93
+Subproject commit 11f734c3b0334dbae4823b4a7467764e447fc6d6
diff --git a/gpt4all-backend/src/llamamodel.cpp b/gpt4all-backend/src/llamamodel.cpp
@@ -53,6 +53,7 @@ static const std::vector<const char *> KNOWN_ARCHES {
     "gpt2",
     // "gptj", -- no inference code
     "gptneox",
+    "granite",
     "granitemoe",
     "mpt",
     "baichuan",
diff --git a/gpt4all-chat/CHANGELOG.md b/gpt4all-chat/CHANGELOG.md
@@ -6,6 +6,9 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
 ## [Unreleased]
 
+### Added
+- Whitelist Granite (non-MoE) model architecture (by [@ThiloteE](https://github.com/ThiloteE) in [#3487](https://github.com/nomic-ai/gpt4all/pull/3487))
+
 ### Fixed
 - Fix several potential crashes ([#3465](https://github.com/nomic-ai/gpt4all/pull/3465))
 - Fix visual spacing issues with deepseek models ([#3470](https://github.com/nomic-ai/gpt4all/pull/3470))
