Modelfile

FROM hf.co/boonk/boonk:Q8_0

TEMPLATE """Below are some instructions that describe some tasks. Write responses that appropriately complete each request.{{ if .Prompt }}

### Instruction:
Extract every US street address found in the text below. Return ONLY a JSON array named 'addresses', where each element contains street, city, state, and zip. Infer missing pieces from context; do not add conversational text.

### Input:
{{ .Prompt }}{{ end }}

### Response:
{{ .Response }}<|end_of_text|>"""

PARAMETER stop "<|eot_id|>"
PARAMETER stop "<|start_header_id|>"
PARAMETER stop "<|end_header_id|>"
PARAMETER stop "<|end_of_text|>"
PARAMETER stop "<|endoftext|>"
PARAMETER stop "<|reserved_special_token_"
PARAMETER temperature 1.5
PARAMETER min_p 0.1

Uploaded model

  • Developed by: boonk
  • License: apache-2.0
  • Finetuned from model : unsloth/Qwen3-14B-Base-unsloth-bnb-4bit

This qwen3 model was trained 2x faster with Unsloth and Huggingface's TRL library.

Downloads last month
83
GGUF
Model size
15B params
Architecture
qwen3
Hardware compatibility
Log In to add your hardware

4-bit

5-bit

8-bit

16-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for boonk/boonk-qwen

Quantized
(2)
this model