GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
KittyLM
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
KittyLM
Model Tree
ibm-granite/granite-4.2-3b
Input Modalities
Output Modalities
Supported Functionality
KittyLM/kittylm-data-reasoning), system prompt baked inscripts/train.py in the KittyLM project), RTX 3060 12GBadapter_*.safetensors) live side by side.
AutoModelForCausalLM loads the merged model; PeftModel picks up the adapter.eval/).python
from transformers import AutoModelForCausalLM, AutoTokenizertok = AutoTokenizer.from_pretrained("KittyLM/kittylm-granite-4.2-3b")model = AutoModelForCausalLM.from_pretrained("KittyLM/kittylm-granite-4.2-3b", device_map="auto", dtype="bfloat16")msgs = [{"role": "user", "content": "What's the capital of Japan?"}]x = tok.apply_chat_template(msgs, tokenize=False, add_generation_prompt=True)print(tok.decode(model.generate(**tok(x, return_tensors="pt").to(model.device), max_new_tokens=120)[0]))# mrrp... Tokyo. big city. lots of cats. nya~bash
hf download KittyLM/kittylm-granite-4.2-3b-gguf --include "*.gguf" --local-dir ./ggufollama create kittylm-granite-4.2-3b -f Modelfile # see Modelfile template in projectollama run kittylm-granite-4.2-3b "Good night!"hf download KittyLM/kittylm-granite-4.2-3b-gguf --include "*.gguf" --local-dir ./ggufollama create kittylm-granite-4.2-3b -f Modelfile # see Modelfile template in projectollama run kittylm-granite-4.2-3b "Good night!"