GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
EphAsad
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
EphAsad
Model Tree
openai/gpt-oss-20b
Input Modalities
Output Modalities
Supported Functionality
python
from unsloth import FastLanguageModelfrom peft import PeftModelimport torch base, tokenizer = FastLanguageModel.from_pretrained( model_name = "unsloth/gpt-oss-20b", max_seq_length = 8192, dtype = torch.bfloat16, load_in_4bit = True, trust_remote_code = True,)model = PeftModel.from_pretrained(base, "EphAsad/Sibyl-20B-LoRA")FastLanguageModel.for_inference(model)python
messages = [{"role": "user", "content": "Your question here"}] inputs = tokenizer.apply_chat_template( messages, tokenize=True, add_generation_prompt=True, return_tensors="pt", return_dict=True, reasoning_effort="high",).to(model.device) import torchwith torch.no_grad(): out = model.generate( **inputs, max_new_tokens=4096, temperature=0.6, top_p=0.95, do_sample=True, ) response = tokenizer.decode( out[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True,)print(response)unsloth/gpt-oss-20b as the base — the adapter was trained against Unsloth's internal format for this modelreasoning_effort="high" in apply_chat_template — the model was trained exclusively at high reasoning efforttrust_remote_code=True required for the gpt-oss architectureApache 2.0, consistent with the base model.
messages = [{"role": "user", "content": "Your question here"}] inputs = tokenizer.apply_chat_template( messages, tokenize=True, add_generation_prompt=True, return_tensors="pt", return_dict=True, reasoning_effort="high",).to(model.device) import torchwith torch.no_grad(): out = model.generate( **inputs, max_new_tokens=4096, temperature=0.6, top_p=0.95, do_sample=True, ) response = tokenizer.decode( out[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True,)print(response)