reaperdoesntknow/Symbiotic-1B

Hugging Face
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:May 6, 2025License:afl-3.0Architecture:Transformer0.0K Featherless Exclusive Warm

SymbioticLM-1B by Convergent Intelligence LLC is a 0.8 billion parameter hybrid symbolic-transformer model built on Qwen-1B, featuring a 32768-token context length. It integrates a symbolic processing pipeline and persistent episodic memory for enhanced reasoning. This model is specifically designed for lightweight, memory-augmented symbolic inference on CPUs and embedded systems, excelling in tasks like educational agents, graph-based explanation generation, and procedural planning.

Loading preview...

SymbioticLM-1B: A Compact Hybrid Symbolic-Transformer for Reasoning

SymbioticLM-1B, developed by Convergent Intelligence LLC, is a 0.8 billion parameter model that combines a Qwen-1B rotary transformer backbone with a unique symbolic processing pipeline and persistent episodic memory. This architecture, part of the SymbioticAI series, is optimized for symbolic reasoning in resource-constrained environments, such as CPU and embedded inference.

Key Capabilities

  • Hybrid Architecture: Fuses a transformer with a symbolic cognitive engine for enhanced reasoning.
  • Memory-Augmented: Features 2048 symbolic vectors with entropic and contextual retrieval for persistent learning.
  • Symbolic Processing: Incorporates advanced symbolic modules like ThoughtDynamicsLNN, CrystallineProcessor (DNAConv GNN), and LiquidThoughtProcessor.
  • Lightweight Design: Optimized for CPU and embedded inference, making it suitable for deployment in memory-sensitive applications.
  • "Dream Mode": Includes a symbolic simulation feature with ThoughtGenerator for complex query processing.

Good For

  • CPU-Optimized Symbolic Inference: Ideal for applications requiring logical processing on standard CPUs.
  • Educational Agents: Developing intelligent agents with memory and reasoning capabilities.
  • Graph-Based Explanation Generation: Creating structured explanations from complex data.
  • Procedural Planning & Math Modeling: Excelling in tasks that require step-by-step logical deduction and mathematical problem-solving.
  • Small-Code Generation: Assisting with the creation of concise and logical code snippets.

This model is rooted in the Discrepancy Calculus (DISC) framework, a measure-theoretic approach to understanding and controlling model behavior, developed by Convergent Intelligence LLC: Research Division. For more details on the underlying theory, refer to the Discrepancy Calculus: Foundations and Core Theory paper.

Finetunes
1 models