TheDrummer (creator of Artemis) has not released the full precision FP16 weights for the Artemis v1h model, so I had to get creative with a Q8_0 GGUF file to produce this merge. There is some precision loss in doing that, but Q8_0 is already pretty close to FP16 in practice, and I think the noise effectively washes out during the merge process.
Glimmer has more diversity in its prose and style compared to Mero-Artemis thanks to Ortenzya's influence. However, Mero-Artemis is overall more stable.
Known Issues
This model requires conservative sampler settings to maintain coherence as the conversation grows in context length. It will produce erroneous outputs if you run it with "aggressive" sampler settings for creativity, so dial it back if you encounter misspellings, grammatical errors, or illogical outputs.
I'd say this model is right on the cusp of instability, but that puts it in an interesting position.
IF YOU GET WEIRD OUTPUTS, TURN DOWN THE HEAT
Sampler Tips
You can use the master import JSON in this repo (Glimmer_SillyTavern_Master_Import.json) to deploy the conservative sampler settings below, which are likely to be compatible with more backend/frontend combos. I recommend using these values as a starting point for your own experiments. It's not like the model falls apart if you deviate from these settings, but they should be a reliable starting point for most creative tasks.
Conservative Settings
Run these settings as a starting point. Raise temperature and lower Min-P if you want more creativity.
Temp0.7
Min-P0.2
Adaptive-P Target0.6
Adaptive-P Decay0.5
DRY Mult.0.8
DRY Base1.8
Prompting Tips
You can download the Glistening-Gem_SillyTavern_Master_Import.json file from this repo and import it directly into SillyTavern to get system prompt, chat template, and sampler settings all in one go.
