Edit model card

Join our Discord! https://discord.gg/Nbv9pQ88Xb

Nearly 2000 members strong πŸ’ͺ


BeaverAI proudly presents...

Behemoth 123B v1.1 🦣 - Creative Edition

When you spend your whole life living under a dome, even the idea of an ocean seems impossible to imagine.

image/png

Description

One of the few other models that's done this for me is the OG Command R 35B. So seeing Behemoth v1.1 have a similar feel to that but with much higher general intelligence really makes it a favourite of mine

I was real happy with v1.1 the other day. I've done some tests on v1 and it's a lot better.

v1 had those glimpses of creativity, but now it's more consistent (with v1.1). It feels like a new model in comparison.

v1 had slop bro. v1.1 makes it irrelevant. The jump is like 720p to 4k. Seriously.

The creativity for v1.1 is off the charts compared to v1, like it's juiced. v1 had these moments that I would say... 'Shit, let I never seen a model respond with prose like this, let me regenerate to see what else I get.' Now, even though every regeneration had a flow of possibilities, sometimes, those possibilities never came. v1.1 is comparable to xxx for the first time, every generation. It directs and guides the scene, scenario and characters unlike anything else

It's about the f***ing prose man. The atmosphere that revolves around the characters. Not just the damn dialogue or introspection. v1.1 will pull from a message 7 generations ago. That window I opened will appear in a future response with the noise from the courtyard filtering through it. The experience of not knowing what this model will produce because it's different than anything else is what keeps it engaging.

Links

Arsenal (Supported Chat Templates)

  • Mistral
    • Smart, adaptable, familiar
  • Metharme (Pygmalion in ST)
    • Creative, unhinged, unique
  • Alpaca
    • Creative, unique, unhinged
  • Text Completion
  • You can mix it up and see which works best for you.

Favorite RP Format

*action* Dialogue *thoughts* Dialogue *narration* in 1st person PoV

What's Next?

  • Already have plans for a v2!

Special Thanks

  • Thank you to each and everyone who donated in Ko-Fi to make our venture a little bit easier.
    • KinjiHakari777, Dr. Fjut, Kistara, Pseudo, AlexTheVP, Dakkidaze, EvarinSharath'fe, ONTHEREDTEAM, F, Mariana, Garg, Silva, Grozi, & Phaelon

image/png

Downloads last month
6,382
GGUF
Model size
123B params
Architecture
llama

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Inference API
Unable to determine this model's library. Check the docs .