GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
sivamani-31
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
sivamani-31
Model Tree
Qwen/Qwen2.5-1.5B-Instruct
Input Modalities
Output Modalities
Supported Functionality
sivamani-31/sika-gpten), Tamil (ta), Malayalam (ml), Telugu (te), Sanskrit (sa), Chinese (zh)transformers, accelerate, vLLM, and Ollama.transformers)python
import torchfrom transformers import AutoTokenizer, AutoModelForCausalLM, pipeline model_id = "sivamani-31/sika-gpt" # Load tokenizer and modeltokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32, device_map="auto") # Text generation pipelinegenerator = pipeline("text-generation", model=model, tokenizer=tokenizer) messages = [ {"role": "system", "content": "You are Sika-GPT, an ultra-efficient open AI assistant for code and reasoning."}, {"role": "user", "content": "Write a clean Python function to find prime numbers using the Sieve of Eratosthenes:"}] formatted_prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) outputs = generator(formatted_prompt, max_new_tokens=512, temperature=0.7, do_sample=True)print(outputs[0]["generated_text"])python
import torchfrom transformers import AutoModelForCausalLM, BitsAndBytesConfig bnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_compute_dtype=torch.float16, bnb_4bit_quant_type="nf4") model = AutoModelForCausalLM.from_pretrained( "sivamani-31/sika-gpt", quantization_config=bnb_config, device_map="auto")q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_projbibtex
@misc{sika_gpt_2026, author = {sivamani-31}, title = {Sika-GPT: An Open, High-Efficiency AI Language Model}, year = {2026}, publisher = {Hugging Face}, howpublished = {\url{https://huggingface.co/sivamani-31/sika-gpt}}}import torchfrom transformers import AutoTokenizer, AutoModelForCausalLM, pipeline model_id = "sivamani-31/sika-gpt" # Load tokenizer and modeltokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32, device_map="auto") # Text generation pipelinegenerator = pipeline("text-generation", model=model, tokenizer=tokenizer) messages = [ {"role": "system", "content": "You are Sika-GPT, an ultra-efficient open AI assistant for code and reasoning."}, {"role": "user", "content": "Write a clean Python function to find prime numbers using the Sieve of Eratosthenes:"}] formatted_prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) outputs = generator(formatted_prompt, max_new_tokens=512, temperature=0.7, do_sample=True)print(outputs[0]["generated_text"])import torchfrom transformers import AutoModelForCausalLM, BitsAndBytesConfig bnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_compute_dtype=torch.float16, bnb_4bit_quant_type="nf4") model = AutoModelForCausalLM.from_pretrained( "sivamani-31/sika-gpt", quantization_config=bnb_config, device_map="auto")@misc{sika_gpt_2026, author = {sivamani-31}, title = {Sika-GPT: An Open, High-Efficiency AI Language Model}, year = {2026}, publisher = {Hugging Face}, howpublished = {\url{https://huggingface.co/sivamani-31/sika-gpt}}}