Skip to content

Commit c6ebead

Browse files
authored
chore(model gallery): add steelskull_l3.3-electra-r1-70b (#4960)
Signed-off-by: Ettore Di Giacinto <[email protected]>
1 parent cff4a95 commit c6ebead

File tree

1 file changed

+18
-0
lines changed

1 file changed

+18
-0
lines changed

gallery/index.yaml

Lines changed: 18 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -817,6 +817,24 @@
817817
- filename: Llama-3.3-MagicalGirl-2.Q4_K_M.gguf
818818
sha256: 01bd7e23c764d18279da4dbd20de19e60009d6e66e8aad1c93732a33f214e6a2
819819
uri: huggingface://mradermacher/Llama-3.3-MagicalGirl-2-GGUF/Llama-3.3-MagicalGirl-2.Q4_K_M.gguf
820+
- !!merge <<: *llama33
821+
name: "steelskull_l3.3-electra-r1-70b"
822+
icon: https://cdn-uploads.huggingface.co/production/uploads/64545af5ec40bbbd01242ca6/GXLpDNkbGEvESfLmWkKpD.jpeg
823+
urls:
824+
- https://huggingface.co/Steelskull/L3.3-Electra-R1-70b
825+
- https://huggingface.co/bartowski/Steelskull_L3.3-Electra-R1-70b-GGUF
826+
description: |
827+
L3.3-Electra-R1-70b is the newest release of the Unnamed series, this is the 6th iteration based of user feedback.
828+
Built on a custom DeepSeek R1 Distill base (TheSkullery/L3.1x3.3-Hydroblated-R1-70B-v4.4), Electra-R1 integrates specialized components through the SCE merge method. The model uses float32 dtype during processing with a bfloat16 output dtype for optimized performance.
829+
Electra-R1 serves newest gold standard and baseline. User feedback consistently highlights its superior intelligence, coherence, and unique ability to provide deep character insights. Through proper prompting, the model demonstrates advanced reasoning capabilities and unprompted exploration of character inner thoughts and motivations.
830+
The model utilizes the custom Hydroblated-R1 base, created for stability and enhanced reasoning. The SCE merge method's settings are precisely tuned based on extensive community feedback (of over 10 diffrent models from Nevoria to Cu-Mai), ensuring optimal component integration while maintaining model coherence and reliability. This foundation establishes Electra-R1 as the benchmark upon which its variant models build and expand.
831+
overrides:
832+
parameters:
833+
model: Steelskull_L3.3-Electra-R1-70b-Q4_K_M.gguf
834+
files:
835+
- filename: Steelskull_L3.3-Electra-R1-70b-Q4_K_M.gguf
836+
sha256: 1f39e1d398ef659ad7074c827dc6993c2007813a303ee72c189e88c4c76f70db
837+
uri: huggingface://bartowski/Steelskull_L3.3-Electra-R1-70b-GGUF/Steelskull_L3.3-Electra-R1-70b-Q4_K_M.gguf
820838
- &rwkv
821839
url: "github:mudler/LocalAI/gallery/rwkv.yaml@master"
822840
name: "rwkv-6-world-7b"

0 commit comments

Comments
 (0)