Devopsopraiz/Baanzon-Chenni-1.5-9B

VISIONPricing:Input $0.431 / Cached $0.0862 / Output $1.12Concurrent Unit Cost:1Model Size:9BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 7, 2026License:mitArchitecture:Transformer0.0K Open Weights Featherless Exclusive Cold

Devopsopraiz/Baanzon-Chenni-1.5-9B is a 9-billion parameter dense reasoning model developed by Devopsopraiz, built on the Qwen3_5ForConditionalGeneration architecture. It features a 262,144-token context window and multimodal vision capabilities. This model is specifically optimized for autonomous software engineering, deep technical logic, and local agentic orchestration, with a stripped refusal behavior profile to maintain continuous technical reasoning flows.

Loading preview...

Model Overview

Devopsopraiz/Baanzon-Chenni-1.5-9B is a 9-billion parameter dense reasoning model designed for advanced technical applications. It utilizes a Qwen3_5ForConditionalGeneration architecture, which includes hybrid linear-attention and full-attention mechanisms, alongside multimodal capabilities. A key differentiator is its significantly reduced refusal behavior, allowing for uninterrupted long-form technical reasoning, crucial for autonomous engineering and coding tasks. The model supports a substantial context length of 262,144 tokens and integrates a qwen3_5_vision encoder for multimodal input.

Key Capabilities

  • Autonomous Software Engineering: Excels in long-horizon planning, code synthesis, refactoring, and debugging across complex, multi-file projects.
  • Deep Technical Logic: Provides structured, step-by-step reasoning for algorithm design and systems-level analysis.
  • Local Agentic Orchestration: Optimized to be lightweight enough for local deployment while effectively orchestrating tools, agents, and sub-processes.
  • Tool & Function Calling: Delivers consistent structured outputs, making it suitable for integration into agent frameworks.
  • Vision + Video: Features built-in multimodal input support for processing visual data.

Usage Considerations

This model has significantly reduced safety filtering, which means it may generate sensitive or controversial content. It is recommended for research, testing, or controlled environments, and users are responsible for monitoring outputs and ensuring compliance with ethical and legal standards. Multi-Token Prediction (MTP) support is also included for advanced use cases.

Finetunes
1 models