GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
AlexThunder0
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
AlexThunder0
Model Tree
Qwen/Qwen3.6-27B
Input Modalities
Output Modalities
Supported Functionality
The adapter keys are canonical PEFT format, so it loads with standard peft — no patching needed:
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelimport torch base = AutoModelForCausalLM.from_pretrained( "Qwen/Qwen3.6-27B", torch_dtype=torch.bfloat16, device_map="auto")model = PeftModel.from_pretrained(base, "AlexThunder0/qwen-cobol-27b")tok = AutoTokenizer.from_pretrained("AlexThunder0/qwen-cobol-27b") prompt = '''Complete the following COBOL program by implementing the PROCEDURE DIVISION.The code MUST compile with the open-source GnuCOBOL compiler. Declare every variablein WORKING-STORAGE, do not use IBM-only intrinsics, end with exactly one END PROGRAM.Provide the complete, compilable COBOL program inside a single ```cobol code block. ```cobol<the COBOLEval-style skeleton with a LINKAGE SECTION and a RESULT field>```''' msg = tok.apply_chat_template([{"role": "user", "content": prompt}], tokenize=False, add_generation_prompt=True, enable_thinking=False)out = model.generate(**tok(msg, return_tensors="pt").to(model.device), max_new_tokens=2048, do_sample=False)print(tok.decode(out[0], skip_special_tokens=True))The model expects the COBOLEval sub-program format: a skeleton with a LINKAGE SECTION and a RESULT field, which it completes with the WORKING-STORAGE SECTION and PROCEDURE DIVISION, ending in GOBACK / END PROGRAM. Greedy decoding is what the reported numbers use.
Full write-up: https://github.com/AlexThunder01/qwenCobol
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelimport torch base = AutoModelForCausalLM.from_pretrained( "Qwen/Qwen3.6-27B", torch_dtype=torch.bfloat16, device_map="auto")model = PeftModel.from_pretrained(base, "AlexThunder0/qwen-cobol-27b")tok = AutoTokenizer.from_pretrained("AlexThunder0/qwen-cobol-27b") prompt = '''Complete the following COBOL program by implementing the PROCEDURE DIVISION.The code MUST compile with the open-source GnuCOBOL compiler. Declare every variablein WORKING-STORAGE, do not use IBM-only intrinsics, end with exactly one END PROGRAM.Provide the complete, compilable COBOL program inside a single ```cobol code block. ```cobol<the COBOLEval-style skeleton with a LINKAGE SECTION and a RESULT field>```''' msg = tok.apply_chat_template([{"role": "user", "content": prompt}], tokenize=False, add_generation_prompt=True, enable_thinking=False)out = model.generate(**tok(msg, return_tensors="pt").to(model.device), max_new_tokens=2048, do_sample=False)print(tok.decode(out[0], skip_special_tokens=True))