Odia NLP · Low-resource LLMs · Model abliteration · LoRA fine-tuning · Local inference · Quantization · Data curation
-
qwen3-8b-heretic
Qwen3-8B fine-tuned for uncensored creative writing and roleplay. Made by RACER IS OP.
tools3,234 Pulls 4 Tags Updated 1 month ago
-
mistral-7b-instruct-v0.3-heretic
Mistral-7B v0.3 fine-tuned for uncensored creative writing and roleplay. Made by RACER IS OP.
tools1,631 Pulls 4 Tags Updated 1 month ago
-
qwen2.5-coder-7b-instruct-heretic
Qwen2.5-Coder-7B fine-tuned for uncensored code generation. Made by RACER IS OP.
tools1,575 Pulls 4 Tags Updated 1 month ago
-
qwen2.5-3b-heretic
Decensored Qwen2.5-3B-Instruct with 2/100 refusals via Heretic abliteration. General-purpose 3B model for local use.
tools1,531 Pulls 1 Tag Updated 1 month ago
-
gemma-3-12b-it-heretic
Gemma-3-12B-IT fine-tuned for uncensored creative writing and roleplay. Made by RACER IS OP.
1,267 Pulls 2 Tags Updated 1 month ago
-
qwen2.5-coder-3b-heretic
Decensored Qwen2.5-Coder-3B-Instruct with 3/100 refusals and near-zero KL divergence via Heretic abliteration.
tools1,244 Pulls 1 Tag Updated 1 month ago
-
qwen2.5-7b-instruct-heretic
Qwen2.5-7B fine-tuned for uncensored creative writing and roleplay. Made by RACER IS OP.
tools1,027 Pulls 4 Tags Updated 1 month ago
-
qwen2.5-14b-instruct-heretic
Qwen2.5-14B fine-tuned for uncensored creative writing and roleplay. Made by RACER IS OP.
tools835 Pulls 4 Tags Updated 1 month ago
-
gemma-3-4b-it-heretic
Gemma-3-4B-IT fine-tuned for uncensored creative writing and roleplay. Made by RACER IS OP.
725 Pulls 2 Tags Updated 1 month ago
-
minicpm5-1b-fable5-heretic
A decensored bilingual 1B mergeset model for roleplay and thinking tasks. Made by RACER IS OP.
tools621 Pulls 4 Tags Updated 1 month ago
-
minicpm5-1b-heretic
A decensored bilingual (EN/ZH) 1B model built on MiniCPM5. Made by RACER IS OP.
tools299 Pulls 4 Tags Updated 1 month ago
-
qwen2.5-1.5b-heretic
A decensored 1.5B model with strong instruction-following and suppressed refusal behavior. Made by RACER IS OP.
tools260 Pulls 4 Tags Updated 1 month ago
-
qwen2.5-0.5b-heretic
A decensored general-purpose 0.5B model with suppressed refusal behavior, ideal for edge deployment. Made by RACER IS OP.
tools163 Pulls 4 Tags Updated 1 month ago
-
llama3.2-1b-heretic
A decensored multilingual 1B Llama 3.2 model with suppressed refusal behavior. Made by RACER IS OP.
tools135 Pulls 4 Tags Updated 1 month ago
-
qwen3-0.6b-heretic
A decensored ultra-lightweight 0.6B Qwen3 model for resource-constrained environments. Made by RACER IS OP.
tools130 Pulls 4 Tags Updated 1 month ago
-
qwen2.5-coder-0.5b-heretic
A decensored code-oriented 0.5B model with suppressed refusal behavior, ideal for local coding agents. Made by RACER IS OP.
tools85 Pulls 4 Tags Updated 1 month ago
-
ghosty-7b
A Hinglish-speaking paranormal investigator LLM. Trained on real Indian ghost encounter transcripts. Made by RACER IS OP.
16 Pulls 1 Tag Updated 1 month ago