GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
0xAbhi
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
0xAbhi
Model Tree
unsloth/Qwen3-0.6B-unsloth-bnb-4bit
Input Modalities
Output Modalities
Supported Functionality
| Tool | Args | Meaning |
|---|---|---|
Forward | {"duration": n} | drive forward n seconds (1–10) |
Backward | {"duration": n} | drive backward n seconds (1–10) |
Turn_Left | {} | turn left 90° |
Turn_Right | {} | turn right 90° |
Stop | {} | stop the motors |
Rules the model learned: default duration is 2 when unspecified, word-numbers are converted to digits, durations are clamped to 1–10, and the output always ends with exactly one Stop.
python
from transformers import AutoModelForCausalLM, AutoTokenizermodel_name = "0xAbhi/qwen3-0.6b-rc-car"tokenizer = AutoTokenizer.from_pretrained(model_name)model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto")system_prompt = ("You control an RC car. Convert the user's command into a JSON array of tool calls. ""Available tools: Forward(duration), Backward(duration), Turn_Left(), Turn_Right(), Stop(). ""Always end with exactly one Stop().")messages = [{"role": "system", "content": system_prompt},{"role": "user", "content": "do a square"},]inputs = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True,enable_thinking=False, return_tensors="pt").to(model.device)out = model.generate(inputs, max_new_tokens=200, temperature=0.1)print(tokenizer.decode(out[0][inputs.shape[1]:], skip_special_tokens=True))
Important: always pass enable_thinking=False in the chat template. Qwen3 supports a reasoning mode, but this model was fine-tuned to answer directly with JSON, not with a <think> block.
SFTTrainerApache 2.0, inherited from the base Qwen3-0.6B model.