This model was finetuned and converted to GGUF format on unsloth, fine tuned on 1million tokens (Or 2 epochs) of Hermes-Function-Calling, this in theory make this model accel at Hermes harness based tool calls, and other tasks.

Example usage:

For text only LLMs: llama-cli -hf koshuro/Qwen2.5-Hermes-Instruct-1.5B --jinja

Available Model files:

Qwen2.5-Hermes-Instruct-1.5B.Q3_K_M.gguf
Qwen2.5-Hermes-Instruct-1.5B.F16.gguf
Qwen2.5-Hermes-Instruct-1.5B.Q5_K_M.gguf
Qwen2.5-Hermes-Instruct-1.5B.Q6_K.gguf
Qwen2.5-Hermes-Instruct-1.5B.Q8_0.gguf
Qwen2.5-Hermes-Instruct-1.5B.Q4_K_M.gguf

koshuro/Qwen2.5-Hermes-Instruct-1.5B

Qwen2.5-Hermes-Instruct-1.5B : GGUF

Available Model files:

koshuro/
Qwen2.5-Hermes-Instruct-1.5B