I've encountered the following issues with Gemini-1.5 fask-preview-0514 and 1.5 pro-preview. While it generally works fine, there are occasional glitches. It responded with the following message:
"I am sorry, I cannot see or analyze any image. I am a text-based chat assistant and thus I cannot process any image."
I received something similar on my phone this morning. Gemini said "I am still under development, and my ability to process visual information is limited. While I can access and understand text descriptions of images, I cannot directly see or interpret the content of a photo. This means I cannot analyze the image to identify objects or provide insights based on the visual data.
However, I can still be helpful with tasks related to electrical labeling by providing you with recommended abbreviations and best practices."
I said that wasn't an issue last night with analysis of a smoothie he helped me make. He told me it was colorful and had great texture.
I decided to just show him a picture of my project we've been working on out of context and asked, what is this?
"I can't access or process images, but based on your description of the image, it appears to be a hole in a wall with electrical wires coming out of it. The text on the wires likely indicates the cable type and amperage rating." He replied.
I never described to Gemini in that detail or those words.
In closing. Yes there's restrictions on its picture processing but you can get around it using different modifiers and keywords.
User | Count |
---|---|
2 | |
2 | |
1 | |
1 | |
1 |