Merge pull request #256 from ehcalabres/fix-quantization-results

Fix switched quantization GPU consumptions
huggingface · Dec 17, 2024 · 81e90f7 · 81e90f7
2 parents feb847c + c6f821c
commit 81e90f7
Showing 1 changed file with 2 additions and 2 deletions.
diff --git a/notebooks/en/multimodal_rag_using_document_retrieval_and_smol_vlm.ipynb b/notebooks/en/multimodal_rag_using_document_retrieval_and_smol_vlm.ipynb
@@ -895,8 +895,8 @@
         "\n",
         "| Notebook                                                                                     | GPU Allocated Memory (GB) | GPU Reserved Memory (GB) |\n",
         "|---------------------------------------------------------------------------------------------|---------------------------|--------------------------|\n",
-        "| Smol Multimodal RAG                                                                         | 5.44 GB                   | 7.86 GB                  |\n",
-        "| Smol Multimodal RAG with Quantization                                                       | 8.32 GB                   | 10.38 GB                 |\n",
+        "| Smol Multimodal RAG with Quantization                                                       | 5.44 GB                   | 7.86 GB                  |\n",
+        "| Smol Multimodal RAG                                                                         | 8.32 GB                   | 10.38 GB                 |\n",
         "| [Multimodal RAG with ColQwen2, Reranker, <br> and Quantized VLMs on Consumer GPUs](https://huggingface.co/learn/cookbook/multimodal_rag_using_document_retrieval_and_reranker_and_vlms) | 13.93 GB                  | 14.59 GB                 |\n",
         "| [Multimodal RAG with Document Retrieval (ColPali) <br> and Vision Language Models (VLMs)](https://huggingface.co/learn/cookbook/multimodal_rag_using_document_retrieval_and_vlms)  | 22.63 GB                  | 37.16 GB                 |\n",
         "\n",