GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
dejavoor
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
dejavoor
Model Tree
Input Modalities
Output Modalities
Supported Functionality
LlamaForCausalLMThis model is useful for:
The model is small and trained for one epoch. It often produces plausible period style but may drift semantically, repeat phrases, or invent incoherent details. Best generations are usually short atmospheric fragments with strong prompts.
python
from transformers import AutoModelForCausalLM, AutoTokenizerimport torch model_id = "YOUR_USERNAME/timecapsulellm-nyc-100m" tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=True)model = AutoModelForCausalLM.from_pretrained(model_id) prompt = "THE NIGHT FERRY.\nIt was past eleven o'clock when I came down to the East River."inputs = tokenizer(prompt, return_tensors="pt", add_special_tokens=False)inputs.pop("token_type_ids", None) with torch.no_grad(): output = model.generate( **inputs, max_new_tokens=80, do_sample=True, temperature=0.82, top_p=0.92, top_k=80, repetition_penalty=1.28, no_repeat_ngram_size=5, pad_token_id=tokenizer.pad_token_id, eos_token_id=tokenizer.eos_token_id, ) print(tokenizer.decode(output[0], skip_special_tokens=True))from transformers import AutoModelForCausalLM, AutoTokenizerimport torch model_id = "YOUR_USERNAME/timecapsulellm-nyc-100m" tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=True)model = AutoModelForCausalLM.from_pretrained(model_id) prompt = "THE NIGHT FERRY.\nIt was past eleven o'clock when I came down to the East River."inputs = tokenizer(prompt, return_tensors="pt", add_special_tokens=False)inputs.pop("token_type_ids", None) with torch.no_grad(): output = model.generate( **inputs, max_new_tokens=80, do_sample=True, temperature=0.82, top_p=0.92, top_k=80, repetition_penalty=1.28, no_repeat_ngram_size=5, pad_token_id=tokenizer.pad_token_id, eos_token_id=tokenizer.eos_token_id, ) print(tokenizer.decode(output[0], skip_special_tokens=True))