espressovi/BODHI-qwen-3-maze-8b-rlvr
TEXT GENERATIONPricing:Input $0.468 / Output $1.82Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 4, 2026License:mitArchitecture:Transformer Open Weights Featherless Exclusive Cold
The espressovi/BODHI-qwen-3-maze-8b-rlvr is an 8 billion parameter language model, fine-tuned from Qwen/Qwen3-8B-Base. This model is specifically designed and optimized for maze-solving tasks, serving as an artifact for the BODHI project. Its primary differentiator is its specialized fine-tuning for maze environments, making it suitable for research and applications requiring navigation or pathfinding within complex mazes.
Loading preview...
BODHI Maze Model Overview
The espressovi/BODHI-qwen-3-maze-8b-rlvr is an 8 billion parameter language model, developed as a specialized artifact for the BODHI project. It is built upon the robust Qwen/Qwen3-8B-Base architecture, undergoing targeted fine-tuning to excel in a very specific domain: maze environments.
Key Capabilities
- Specialized Maze Solving: This model's core strength lies in its fine-tuned ability to process and potentially solve maze-related problems. Its training is focused on understanding the intricacies of maze structures and navigation.
- Foundation on Qwen3-8B-Base: Leveraging the capabilities of the Qwen3-8B-Base model, it inherits a strong language understanding foundation, which is then adapted for its niche application.
Good For
- Research in Maze Environments: Ideal for researchers exploring AI agents in maze navigation, pathfinding algorithms, or reinforcement learning within constrained environments.
- Specific Project Artifacts: As an artifact of the BODHI project, it is particularly relevant for those involved with or interested in the outcomes and methodologies of that specific research initiative.