reaperdoesntknow/DistilQwen3-1.7B-uncensored

Hugging Face
TEXT GENERATIONPricing:Input $0.32 / Cached $0.064 / Output $1.6Concurrent Unit Cost:1Model Size:2BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Mar 25, 2026Architecture:Transformer Featherless Exclusive Warm

reaperdoesntknow/DistilQwen3-1.7B-uncensored is a 1.7 billion parameter uncensored language model from the DistilQwen3 series by Convergent Intelligence LLC: Research Division. This model is part of a proof-weighted distillation chain built on Discrepancy Calculus, designed to amplify reasoning-critical tokens. It excels in instruction following, structured output, and legal reasoning, making it suitable for tasks requiring robust logical inference.

Loading preview...

Model Overview

reaperdoesntknow/DistilQwen3-1.7B-uncensored is a 1.7 billion parameter model developed by Convergent Intelligence LLC: Research Division, forming part of their DistilQwen3 series. This model is a product of a proof-weighted knowledge distillation process, utilizing a 30B-parameter teacher model and trained on H100 hardware at BF16 precision. The distillation methodology incorporates 55% cross-entropy with decaying proof weights and 45% KL divergence, specifically designed to enhance the student model's structural understanding over surface-level pattern matching by amplifying loss on reasoning-critical tokens.

Key Capabilities

  • Instruction Following: Demonstrates strong ability to adhere to given instructions.
  • Structured Output: Capable of generating responses in a structured format.
  • Legal Reasoning: Optimized for tasks requiring legal reasoning and logical inference.
  • Discrepancy Calculus Foundation: Built upon a measure-theoretic framework that quantifies local structural mismatch, aiming for deeper understanding.

Good For

  • Applications requiring precise instruction adherence.
  • Generating structured data or text.
  • Tasks involving legal analysis or complex logical deductions.
  • Use cases where robust reasoning is prioritized over simple pattern recognition.

Popular Sampler Settings

Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.

temperature
top_p
top_k
frequency_penalty
presence_penalty
repetition_penalty
min_p