github/gpt4all
1
0
Fork 0
mirror of https://github.com/nomic-ai/gpt4all.git synced 2025-09-04 01:54:49 +00:00
Code Issues Packages Projects Releases Wiki Activity

embllm: fix use of llama ctx before loading (#2465)

Browse Source 
This fixes a regression in PR #2396.

Signed-off-by: Jared Van Bortel <jared@nomic.ai>
This commit is contained in:
Jared Van Bortel
2024-06-25 11:04:01 -04:00
committed by GitHub
parent 9273b49b62
commit 1a00882276
1 changed files with 5 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
gpt4all-chat/embllm.cpp
View File

@@ -84,10 +84,6 @@ bool EmbeddingLLMWorker::loadModel()
return false;
}
// FIXME(jared): the user may want this to take effect without having to restart
int n_threads = MySettings::globalInstance()->threadCount();
m_model->setThreadCount(n_threads);
// NOTE: explicitly loads model on CPU to avoid GPU OOM
// TODO(cebtenzzre): support GPU-accelerated embeddings
bool success = m_model->loadModel(filePath.toStdString(), 2048, 0);
@@ -104,6 +100,11 @@ bool EmbeddingLLMWorker::loadModel()
m_model = nullptr;
return false;
}
// FIXME(jared): the user may want this to take effect without having to restart
int n_threads = MySettings::globalInstance()->threadCount();
m_model->setThreadCount(n_threads);
return true;
}