chore(model gallery): add q2.5-32b-slush-i1 (#4292)

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2024-11-30 11:02:41 +01:00 committed by GitHub
parent 2c8a87b1e4
commit 61358e4d35
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

View File

@ -1675,6 +1675,22 @@
- filename: QwQ-32B-Preview-Q4_K_M.gguf
sha256: c499801e682e2379528090c50e106837ca1d69dc3bf3ff3a9af830a0eb49cdf6
uri: huggingface://bartowski/QwQ-32B-Preview-GGUF/QwQ-32B-Preview-Q4_K_M.gguf
- !!merge <<: *qwen25
name: "q2.5-32b-slush-i1"
urls:
- https://huggingface.co/crestf411/Q2.5-32B-Slush
- https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF
description: |
Slush is a two-stage model trained with high LoRA dropout, where stage 1 is a pretraining continuation on the base model, aimed at boosting the model's creativity and writing capabilities. This is then merged into the instruction tune model, and stage 2 is a fine tuning step on top of this to further enhance its roleplaying capabilities and/or to repair any damage caused in the stage 1 merge.
This is still early stage. As always, feedback is welcome, and begone if you demand perfection.
The second stage, like the Sunfall series, follows the Silly Tavern preset (ChatML), so ymmv in particular if you use some other tool and/or preset.
overrides:
parameters:
model: Q2.5-32B-Slush.i1-Q4_K_M.gguf
files:
- filename: Q2.5-32B-Slush.i1-Q4_K_M.gguf
sha256: 95aecaf43077dabc72d3b556923ede2563325e1c89863800229cfa8b7f1c9659
uri: huggingface://mradermacher/Q2.5-32B-Slush-i1-GGUF/Q2.5-32B-Slush.i1-Q4_K_M.gguf
- &archfunct
license: apache-2.0
tags: