gguf-org/fastvlm-gguf

fastvlm-gguf

• —run it with gguf-connector; simply execute the command below in console/terminal

ggc f6

GGUF file(s) available. Select which one to use: 1. fastvlm-0.5b-iq4nl.gguf 3. fastvlm-0.5b-q40.gguf 4. fastvlm-0.5b-q80.gguf Enter your choice (1 to 3):

• —opt a gguf file in your current directory to interact with; nothing else

for the latest update, you should be able to customize how many tokens for output (see picture)

connector f5 (alternative 1)

ggc f5

• —note: f5 is different from f6 (above) and f9 (below); you don't need gguf files with f5

connector f7 (alternative 2)*

ggc f7

*real time screen describer; live captioning

connector f9 - advanced mode (alternative 3)**

ggc f9

**for advanced mode, it's available to specify your text prompt with picture input (see above)

reference

• —base model from apple
• —gguf-connector (pypi)