GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
kvn12
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
kvn12
Model Tree
Qwen/Qwen3-1.7B
Input Modalities
Output Modalities
Supported Functionality
Held-out 1,100-item test split the model never trained on, served at Q4_K_M via Ollama, judged by Claude Opus 5 (structured rubric). Base is re-baselined through the identical serving pipeline, so only the weights differ.
| metric | base | tuned | Δ |
|---|---|---|---|
| Issue understanding (1–5) | 4.11 | 4.86 | +0.76 |
| Correctness (1–5) | 3.63 | 4.57 | +0.94 |
| Resolution quality (1–5) | 3.26 | 4.16 | +0.90 |
| Support tone (1–5) | 3.87 | 4.60 | +0.73 |
| Hallucination rate | 39.7% | 11.7% | −28 pts |
Improves on every metric across all 11 categories and 27 intents; hallucination drops ~70% relative.
Qwen3 chat format with thinking disabled (enable_thinking=False, i.e. an empty
<think></think> block in the assistant turn). Fixed system prompt:
You are a customer support assistant for an online business. A customer has sent you a message. Respond directly to the customer in a helpful, professional, and empathetic tone. Understand what they need, give accurate information, and clearly explain any steps required to resolve their request. Keep the response focused and concise.
Decoding: greedy (temperature=0), seed=42, max_new_tokens=512.
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelSYSTEM = ("You are a customer support assistant for an online business. A customer has ""sent you a message. Respond directly to the customer in a helpful, professional, ""and empathetic tone. Understand what they need, give accurate information, and ""clearly explain any steps required to resolve their request. Keep the response ""focused and concise.")tok = AutoTokenizer.from_pretrained("Qwen/Qwen3-1.7B")base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-1.7B", torch_dtype="float16", device_map="auto")model = PeftModel.from_pretrained(base, "kvn12/qwen3-1.7b-cs-support-lora")msgs = [{"role": "system", "content": SYSTEM},{"role": "user", "content": "I was charged twice for order 8842, please help"}]ids = tok.apply_chat_template(msgs, add_generation_prompt=True,enable_thinking=False, return_tensors="pt").to(model.device)out = model.generate(ids, max_new_tokens=512, do_sample=False)print(tok.decode(out[0][ids.shape[1]:], skip_special_tokens=True))
LoRA r=16, α=32, dropout=0.05 on q,k,v,o,gate,up,down; 3 epochs (early stopping,
patience 2); lr 2e-4 cosine, 3% warmup; effective batch 32; max seq len 1024;
completion-only loss (prompt masked). ~497 min on one Kaggle T4 (fp16).
Checkpoint chosen by task metrics (judge), not just val loss; val loss 0.60→0.57→0.566.
{{placeholder}} style (e.g. {{Order Number}}) — de-template
or post-process for production.