Whitelist OLMoE and Granite MoE (#3449)

Signed-off-by: ThiloteE <73715071+ThiloteE@users.noreply.github.com>
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
Co-authored-by: Jared Van Bortel <jared@nomic.ai>

gpt4all-backend/deps/llama.cpp-mainline

@@ -1 +1 @@
-Subproject commit b06658d366abe3cea92f4e868db72165531a74da
+Subproject commit 3ebb3603e807d74a16f061c46d2925a1653e7a93

gpt4all-backend/src/llamamodel.cpp

@@ -53,6 +53,7 @@ static const std::vector<const char *> KNOWN_ARCHES {
     "gpt2",
     // "gptj", -- no inference code
     "gptneox",
+    "granitemoe",
     "mpt",
     "baichuan",
     "starcoder",
@@ -80,6 +81,7 @@ static const std::vector<const char *> KNOWN_ARCHES {
     "command-r",
     // "dbrx", -- 16x12B parameters
     "olmo",
+    "olmoe",
     "openelm",
     // "arctic", -- 10B+128x3.66B parameters
     "deepseek2",

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 OLMoE and Granite MoE model architectures (no Vulkan) ([#3449](https://github.com/nomic-ai/gpt4all/pull/3449))
+
 ### Fixed
 - Fix "index N is not a prompt" when using LocalDocs with reasoning ([#3451](https://github.com/nomic-ai/gpt4all/pull/3451)
 - Work around rendering artifacts on Snapdragon SoCs with Windows ([#3450](https://github.com/nomic-ai/gpt4all/pull/3450))