Mahesh111000/qwen3-8b-hanabi-rl-base-1to1-24k-step_185

TEXT GENERATIONPricing:Input $0.468 / Output $1.82Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 9, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Mahesh111000/qwen3-8b-hanabi-rl-base-1to1-24k-step_185 is an 8.2 billion parameter causal language model from the Qwen3 series, developed by Qwen. This model uniquely supports seamless switching between a 'thinking mode' for complex logical reasoning, math, and coding, and a 'non-thinking mode' for efficient general-purpose dialogue. It excels in reasoning capabilities, human preference alignment, and agentic tasks, supporting over 100 languages with a native context length of 32,768 tokens.

Loading preview...

Qwen3-8B: A Versatile Language Model with Adaptive Reasoning

Mahesh111000/qwen3-8b-hanabi-rl-base-1to1-24k-step_185 is an 8.2 billion parameter causal language model from the Qwen3 series, designed to offer advanced capabilities across various tasks. A key differentiator is its unique ability to seamlessly switch between a 'thinking mode' for complex logical reasoning, mathematics, and code generation, and a 'non-thinking mode' for efficient, general-purpose dialogue. This adaptive approach ensures optimal performance tailored to specific scenarios.

Key Capabilities

  • Adaptive Reasoning: Dynamically switches between thinking and non-thinking modes, enhancing performance in complex problem-solving and general conversation.
  • Enhanced Reasoning: Demonstrates significant improvements in mathematical, coding, and commonsense logical reasoning compared to previous Qwen models.
  • Human Preference Alignment: Excels in creative writing, role-playing, multi-turn dialogues, and instruction following, providing a more natural conversational experience.
  • Agentic Expertise: Achieves leading performance among open-source models in complex agent-based tasks, with precise integration with external tools.
  • Multilingual Support: Supports over 100 languages and dialects, offering strong multilingual instruction following and translation capabilities.
  • Extended Context: Natively handles a context length of 32,768 tokens, extendable up to 131,072 tokens using the YaRN method.

Best Practices for Usage

  • Sampling Parameters: Recommended settings vary for thinking (Temperature=0.6, TopP=0.95, TopK=20, MinP=0) and non-thinking modes (Temperature=0.7, TopP=0.8, TopK=20, MinP=0). Greedy decoding is not recommended for thinking mode.
  • Output Length: Suggests 32,768 tokens for most queries, and up to 38,912 for highly complex problems.
  • Standardized Output: Use specific prompts for math problems (e.g., "Please reason step by step, and put your final answer within \boxed{}") and multiple-choice questions (e.g., JSON structure for answers) to standardize model outputs during benchmarking.