Update README.md
Browse files
README.md
CHANGED
|
@@ -9,4 +9,7 @@ If your prompt is just 576 + a few tokens, you are using llava-1.5 code (or proj
|
|
| 9 |
The mmproj files are the embedded ViT's that came with llava-1.6, I've not compared them but given the previous releases from the team I'd be surprised if the ViT has not been fine tuned this time.
|
| 10 |
If that's the case, using another ViT can cause issues.
|
| 11 |
|
| 12 |
-
Original models: https://github.com/haotian-liu/LLaVA
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
The mmproj files are the embedded ViT's that came with llava-1.6, I've not compared them but given the previous releases from the team I'd be surprised if the ViT has not been fine tuned this time.
|
| 10 |
If that's the case, using another ViT can cause issues.
|
| 11 |
|
| 12 |
+
Original models: https://github.com/haotian-liu/LLaVA
|
| 13 |
+
|
| 14 |
+
For more llava-34B quantization variants check here: https://huggingface.co/cjpais/llava-v1.6-34B-gguf
|
| 15 |
+
Warning: Do NOT use the mmproj model from that page, use the 34B mmproj from here and combine it with any of the llm-GGUFs
|