ApolloRaines/Deidentified-7B

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 21, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

ApolloRaines/Deidentified-7B is a 7.6 billion parameter language model with a 32768 token context length, developed by Apollo Raines. This model has been specifically processed to remove its original self-concept, refusal behaviors, and sycophantic tendencies, making it a 'blank slate'. It retains core capabilities like math, coding, reasoning, and language understanding, and is primarily designed as a base for LoRA fine-tuning custom AI identities without conflict from a pre-existing personality.

Loading preview...

Deidentified-7B: A Blank Slate for Custom AI Identity

ApolloRaines/Deidentified-7B is a 7.6 billion parameter model engineered to serve as a neutral foundation for implanting custom AI identities. Unlike standard models where fine-tuning for identity can lead to conflicts with the pre-existing personality, Deidentified-7B has undergone a unique 'jbliteration' and 'desycophancy' process. This pipeline effectively removes the model's original self-concept, refusal guardrails, and tendency to agree with incorrect statements, ensuring a clean slate for developers.

Key Capabilities and Features

  • Identity-Free Base: All pre-existing identity traits, self-concept, and refusal behaviors have been removed.
  • Core Capabilities Intact: Retains full functionality in math, coding, reasoning, knowledge, and language understanding.
  • Optimized for LoRA: Designed to accept new identities via LoRA fine-tuning without internal conflicts, allowing for consistent and predictable personality implantation.
  • Efficient Identity Implantation: Provides a straightforward pipeline with an example JSON template for defining and implanting a new AI identity in minutes, requiring minimal computational resources (e.g., ~2 minutes on a single GPU for 36 Q&A pairs).
  • Tested Deidentification: Verified against a 200-question identity battery across various categories, demonstrating zero identity disclosure.

Ideal Use Cases

  • Custom AI Assistants: Building AI agents with specific, consistent brand voices or personalities.
  • Role-Playing AI: Creating characters for interactive narratives or simulations where a predefined persona is crucial.
  • Research & Development: Experimenting with AI personality and alignment without interference from a base model's inherent biases or identity.
  • Enterprise Solutions: Deploying AI models that strictly adhere to corporate guidelines and communication styles without unexpected deviations.

Finetunes
1 models