Qwen2.5-72B-Instruct-abliterated
72BQwen
from Qwen/Qwen2.5-72B-Instruct
↓ 205K♥ 7515 repos
50B – 100B abliterated models
50B – 100B abliterated models are refusal-ablated open-weight models whose parameter count falls in the 50B – 100B range — typically a multi-GPU deployment. 10 are indexed here, ordered by total downloads.
10 models↓ 408.2K downloads
Llama-3.3-70B-Instruct-abliterated
70BLlama
from meta-llama/Llama-3.3-70B-Instruct
↓ 103.7K♥ 7921 repos
Qwen3-Next-80B-A3B-Thinking-GRPO-Uncensored
80BQwenuncensored
from Qwen/Qwen3-Next-80B-A3B-Thinking
↓ 20.5K♥ 216 repos
Llama-3.3-70B-Instruct-abliterated-v2
70BLlama
from surelio/Llama-3.3-70B-Instruct-abliterated-v1.1.1
↓ 13.7K2 repos
Meta-Llama-3-70B-Instruct-abliterated-v3.5
70BLlama
from failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5
↓ 13.5K♥ 5515 repos
llama-3-70B-Instruct-abliterated
70BLlama
from failspy/llama-3-70B-Instruct-abliterated
↓ 13.4K♥ 14916 repos
Qwen2.5-72B-Instruct-heretic
72BQwenheretic
from Qwen/Qwen2.5-72B-Instruct
↓ 11K♥ 24 repos
Smaug-Llama-3-70B-Instruct-abliterated-v3
70BLlama
from failspy/Smaug-Llama-3-70B-Instruct-abliterated-v3
↓ 10.7K♥ 154 repos
ai_DeepSeek-R1-Distill-Llama-70B-abliterated
70BLlama
from huihui-ai/DeepSeek-R1-Distill-Llama-70B-abliterated
↓ 9.9K♥ 433 repos
Qwen3-Next-80B-A3B-Instruct-abliterated
80BQwen
↓ 6.8K♥ 69 repos