qwen3-8b-reasoning-agent-GGUF

qwen3-8b-reasoning-agent-GGUF : GGUF

This model was finetuned and converted to GGUF format using Unsloth.

Example usage:

• For text only LLMs: llama-cli -hf yukiii02/qwen3-8b-reasoning-agent-GGUF --jinja
• For multimodal models: llama-mtmd-cli -hf yukiii02/qwen3-8b-reasoning-agent-GGUF --jinja

Available Model files:

• qwen3-8b.Q4_K_M.gguf

Datasets

1. lambda/hermes-agent-reasoning-traces (kimi)

• Samples: 7,646
• Type: Multi-turn tool-calling trajectories with chain-of-thought reasoning
• Format: <think> → <tool_call> → <tool_response> → response
• Source: Generated by Moonshot AI Kimi-K2.5 via Hermes Agent harness with real tool execution
• License: Apache 2.0

2. jasonrqh/Math-CoT-20k

• Samples: 20,480 (sampled 7,646 for balance)
• Type: Math reasoning chain-of-thought
• Format: Question → step-by-step solution
• License: MIT

Both datasets were mixed with equal sampling (7,646 each, shuffled).

Ollama

An Ollama Modelfile is included for easy deployment. This was trained 2x faster with Unsloth <img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>