General discussion.
Quants being prepared. You can link already if you want.
Did some reorganization of the ReadMe, let me know if anything is unclear or can be improved.
Looks good from my house! (albeit i usually am a lazy bastard with the description's)
When I'm not feeling lazy, I should also add some VRAM => QUANT recommendations, but at this point I expect people to just figure out what is best for them.
I tried:
A few general recommendations for quant options:
Assuming a context size of 8192 for simplicity and 1GB of Operating System VRAM overhead with some safety margin to avoid overflowing buffers.
For 11-12GB VRAM: A GPU with 12GB of VRAM capacity can comfortably use the Q6_K-imat quant option and run it at good speeds.
This is the same with or without using #vision capabilities.
For 8GB VRAM: If not using #vision, for GPUs with 8GB of VRAM capacity the Q5_K_M-imat quant option will fit comfortably and should run at good speeds.
If you are also using #vision from this model opt for the Q4_K_M-imat quant option to avoid filling the buffers and potential slowdow.
For 6GB VRAM: If not using #vision, for GPUs with 6GB of VRAM capacity the IQ3_M-imat quant option should fit comfortably to run at good speeds.
If you are also using #vision from this model opt for the IQ3_XXS-imat quant option.