GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
nabin2004
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
nabin2004
Model Tree
Qwen/Qwen3-8B
Input Modalities
Output Modalities
Supported Functionality
You can run this merged model directly in Ollama:
bash
ollama run hf.co/nabin2004/qwen-Manimator-1-mergedOr run the dedicated GGUF repository:
bash
ollama run hf.co/nabin2004/qwen-Manimator-1-ggufVoiceoverScene.<bookmark mark='NAME'/> tags + self.wait_until_bookmark("NAME") for audio-visual sync.<Plan> reasoning block followed by the Python implementation.python
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model_id = "nabin2004/qwen-Manimator-1-merged"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.bfloat16, device_map="auto") messages = [ {"role": "system", "content": "You are an expert Python programmer and mathematics educator specializing in ManimCE and Manim Voiceover..."}, {"role": "user", "content": "Create a Manim animation explaining Fourier series with audio narration."},]ids = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to("cuda")out = model.generate(ids, max_new_tokens=2048, temperature=0.2, top_p=0.9)print(tokenizer.decode(out[0][ids.shape[-1]:], skip_special_tokens=True))| Parameter | Value |
|---|---|
| Base Model | Qwen/Qwen3-8B |
| Dataset | nabin2004/qwen-Manimator-1-sft-data |
| Epochs | 3 |
| Max Length | 4500 |
| Learning Rate | 1e-4 |
| LoRA | r=16, alpha=32 |
| Optimizer | paged_adamw_8bit |
| Hardware | Kaggle T4×2 |
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model_id = "nabin2004/qwen-Manimator-1-merged"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.bfloat16, device_map="auto") messages = [ {"role": "system", "content": "You are an expert Python programmer and mathematics educator specializing in ManimCE and Manim Voiceover..."}, {"role": "user", "content": "Create a Manim animation explaining Fourier series with audio narration."},]ids = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to("cuda")out = model.generate(ids, max_new_tokens=2048, temperature=0.2, top_p=0.9)print(tokenizer.decode(out[0][ids.shape[-1]:], skip_special_tokens=True))