CoolFace
Modelpublic

Pimmetjeoss/tribe-crm-qwen-2b-merged

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face
0likes7downloads
Model Card

Tribe CRM Agent — Qwen3.5-2B (merged)

Merged version of `Pimmetjeoss/tribe-crm-qwen-2b` (LoRA adapter) combined with its base model `unsloth/Qwen3.5-2B`.

Use this if you want a single `from_pretrained()` load without PEFT dependencies. Use the adapter-only version if you want to swap/combine adapters.

Performance

MetricSynthetic heldoutRealistic heldout
Combined accuracy91.2% (52/57)98.2% (56/57)
Tool accuracy98.2% (56/57)98.2% (56/57)
Args accuracy91.2% (52/57)98.2% (56/57)

Usage

python
from transformers import AutoModelForCausalLM, AutoTokenizer

MODEL = "Pimmetjeoss/tribe-crm-qwen-2b-merged"
tokenizer = AutoTokenizer.from_pretrained(MODEL)
model = AutoModelForCausalLM.from_pretrained(MODEL, dtype="bfloat16").to("cuda").eval()

messages = [
    {"role": "system", "content": "You are a model that can do function calling with the following functions"},
    {"role": "user", "content": "Zoek Bakkerij De Wit op"},
]
inputs = tokenizer.apply_chat_template(messages, tools=YOUR_TOOLS_SCHEMA,
                                       add_generation_prompt=True,
                                       return_dict=True, return_tensors="pt").to("cuda")
out = model.generate(**inputs, max_new_tokens=128, do_sample=False)
print(tokenizer.decode(out[0][inputs["input_ids"].shape[1]:]))

Related