GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
antontuzovAI
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
antontuzovAI
Model Tree
openbmb/MiniCPM5-2B
Input Modalities
Output Modalities
Supported Functionality
<think> reasoning tags.openbmb/MiniCPM5-2BSalesforce/xlam-function-calling-60k (Full 60,000 samples)q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_projYou need the base model and the peft library to load this adapter.
python
import torchimport jsonfrom transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfigfrom peft import PeftModel model_id = "openbmb/MiniCPM5-2B"adapter_id = "antontuzovAI/MiniCPM5-2B-xLAM-60k-Production-Agent" # 1. Setup 4-bit Quantizationbnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.bfloat16, bnb_4bit_use_double_quant=True) # 2. Load Base Model & Tokenizertokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)tokenizer.pad_token = tokenizer.eos_token base_model = AutoModelForCausalLM.from_pretrained( model_id, quantization_config=bnb_config, device_map="auto", trust_remote_code=True, torch_dtype=torch.bfloat16) # 3. Load the Agent LoRAmodel = PeftModel.from_pretrained(base_model, adapter_id)model.eval() # 4. Define Toolstools = [ { "name": "get_weather", "description": "Get current weather for a city.", "parameters": { "city": {"type": "string", "description": "City name", "required": True}, "units": {"type": "string", "enum": ["celsius", "fahrenheit"]} } }] # 5. Build Prompttools_str = json.dumps(tools, indent=2)system_msg = ( "You are a strict tool-calling agent.\n" "Output ONLY a valid JSON array of tool calls.\n" 'Format: [{"name": "tool_name", "arguments": {"arg": "val"}}]\n' "If no tool is needed, output: []\n\n" f"Available tools:\n{tools_str}") messages = [ {"role": "system", "content": system_msg}, {"role": "user", "content": "What is the weather in Tokyo in celsius?"}] prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) # 6. Generateinputs = tokenizer(prompt, return_tensors="pt").to(model.device)with torch.no_grad(): outputs = model.generate(**inputs, max_new_tokens=128, do_sample=False, pad_token_id=tokenizer.eos_token_id) response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)print("Agent Output:\n", response)Because this LoRA is heavily optimized for strict JSON routing, it is recommended to use a "Two-Brain" architecture in production:
To switch back to the base model for synthesis, simply use model.disable_adapter_layers() before generating the final answer.
import torchimport jsonfrom transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfigfrom peft import PeftModel model_id = "openbmb/MiniCPM5-2B"adapter_id = "antontuzovAI/MiniCPM5-2B-xLAM-60k-Production-Agent" # 1. Setup 4-bit Quantizationbnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.bfloat16, bnb_4bit_use_double_quant=True) # 2. Load Base Model & Tokenizertokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)tokenizer.pad_token = tokenizer.eos_token base_model = AutoModelForCausalLM.from_pretrained( model_id, quantization_config=bnb_config, device_map="auto", trust_remote_code=True, torch_dtype=torch.bfloat16) # 3. Load the Agent LoRAmodel = PeftModel.from_pretrained(base_model, adapter_id)model.eval() # 4. Define Toolstools = [ { "name": "get_weather", "description": "Get current weather for a city.", "parameters": { "city": {"type": "string", "description": "City name", "required": True}, "units": {"type": "string", "enum": ["celsius", "fahrenheit"]} } }] # 5. Build Prompttools_str = json.dumps(tools, indent=2)system_msg = ( "You are a strict tool-calling agent.\n" "Output ONLY a valid JSON array of tool calls.\n" 'Format: [{"name": "tool_name", "arguments": {"arg": "val"}}]\n' "If no tool is needed, output: []\n\n" f"Available tools:\n{tools_str}") messages = [ {"role": "system", "content": system_msg}, {"role": "user", "content": "What is the weather in Tokyo in celsius?"}] prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) # 6. Generateinputs = tokenizer(prompt, return_tensors="pt").to(model.device)with torch.no_grad(): outputs = model.generate(**inputs, max_new_tokens=128, do_sample=False, pad_token_id=tokenizer.eos_token_id) response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)print("Agent Output:\n", response)