espressovi/BODHI-qwen-3.5-9b-distil

VISIONPricing:Input $0.431 / Cached $0.0862 / Output $1.12Concurrent Unit Cost:1Model Size:9BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Apr 8, 2026License:mitArchitecture:Transformer Open Weights Featherless Exclusive Cold

BODHI-qwen-3.5-9b-distil is a math-focused language model developed by espressovi, distilled from Qwen3.5-9B-Base. This model is specifically optimized for mathematical reasoning tasks, achieving a 65.64% score on AIME25. It is designed for applications requiring strong performance in complex mathematical problem-solving.

Loading preview...

Overview

The espressovi/BODHI-qwen-3.5-9b-distil model is a specialized language model developed by espressovi, created through Long-CoT distillation from the Qwen3.5-9B-Base architecture. Its primary focus is on mathematical reasoning and problem-solving.

Key Capabilities

  • Mathematical Proficiency: The model is specifically trained and optimized for math-related tasks.
  • Distilled Architecture: Benefits from the distillation process, likely offering efficiency while retaining strong performance in its niche.

Performance

  • Achieves 65.64% on the AIME25 benchmark, evaluated at T=1.0 with 16K tokens and pass@8.

When to Use This Model

  • Mathematical Applications: Ideal for use cases requiring robust mathematical problem-solving abilities.
  • Research in Math AI: Suitable for researchers exploring advanced mathematical reasoning in LLMs.
  • Educational Tools: Potentially useful for developing tools that assist with complex math problems.