GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
NIyueeE
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
NIyueeE
Model Tree
Base
this model
Input Modalities
Text
Image
Video
Output Modalities
Text
Supported Functionality
Dedicated Endpoints
Google Colab (colab.research.google.com) with NVIDIA A100-SXM4-40GB.
markdown
Unsloth 2026.5.5: Fast Qwen3_5 patching. Transformers: 5.5.0.NVIDIA A100-SXM4-40GB. Num GPUs = 1. Max memory: 39.494 GB.Torch: 2.10.0+cu128. CUDA: 8.0. CUDA Toolkit: 12.8. Triton: 3.6.0Bfloat16 = TRUE. FA [Xformers = 0.0.35. FA2 = False]Num examples = 1,227 | Num Epochs = 7 | Total steps = 50Batch size per device = 128 | Gradient accumulation steps = 1Total batch size (128 x 1 x 1) = 128Trainable parameters = 13,181,952 of 866,167,872 (1.52% trained)
See finetune_cocreator_coclab.ipynb for the complete fine-tuning notebook.
| Parameter | Value |
|---|---|
| LoRA r | 16 |
| LoRA alpha | 16 |
| LoRA dropout | 0 |
| Target modules | all-linear |
| Fine-tuned layers | vision + language + attention + MLP |
| Optimizer | adamw_8bit |
| Learning rate | 5e-5 (cosine schedule) |
| Max steps | 50 |
| Epochs | 7 |
python
from transformers import AutoModel, AutoTokenizerimport torchmodel = AutoModel.from_pretrained("NIyueeE/Qwen3.5-0.8B-cocreator",torch_dtype=torch.bfloat16,trust_remote_code=True,)tokenizer = AutoTokenizer.from_pretrained("NIyueeE/Qwen3.5-0.8B-cocreator",trust_remote_code=True,)
This model is fine-tuned for driving scene causal understanding. It takes multi-frame driving images as input and generates causal relationship text descriptions. The primary use case is as a feature extractor in the ReCogDrive autonomous driving VLA training pipeline.
Apache 2.0
| Gradient checkpointing | unsloth |
| Resolution | 800×450 (resized) |