Architectures / Comparisons
What is actually different between two models
A textual diff of model code shows renamed variables. It does not show what the change did to the shape of the network or to the bill. These 30 comparisons do, layer by layer, for graphs the verifier has already checked. 7 of them turn on a GPU one model fits and the other does not.
Comparing two models of your own, or anything on Hugging Face: open the comparison tool. Nothing to install and no account.
GPT-2 vs Qwen3-8B
What seven years of scaling actually changed inside the decoder.
LLaMA-3 Block vs Mixtral MoE Block
A dense feed-forward against a mixture of experts, at the block level.
Mixtral MoE Block vs DeepSeek-V3
Two mixture-of-experts designs that route very differently.
LLaMA-3 Block vs DeepSeek-V3
What a frontier open model adds to the block everyone started from.
Transformer Block vs LLaMA-3 Block
What changed in the transformer block between 2017 and now.
Transformer Block vs Mamba SSM Block
Attention against a state-space layer for the same job.
Mamba SSM Block vs Jamba
Pure state space against a hybrid that keeps some attention.
GPT-2 vs BERT Base
Decoder-only against encoder-only, same era, same size class.
BERT Base vs T5 Small
Encoder-only against encoder-decoder.
Qwen3-8B vs Phi-3 Mini Block
What gets cut to make a small model small.
LLaMA-3 Block vs Phi-3 Mini Block
A block from a large model against a whole small one.
Qwen3-8B vs Llama-4 Scout
Two 2026 open models with different attention budgets.
ResNet Block vs Simple CNN
What the residual connection costs and what it buys.
ResNet Block vs ViT-B/16
Convolution against attention for images.
ViT-B/16 vs Swin-Tiny
A flat vision transformer against a hierarchical one.
BERT Base vs ViT-B/16
The same transformer applied to text and to images.
U-Net vs Diffusion UNet
A segmentation U-Net against the one a diffusion model uses.
Diffusion UNet vs DiT-XL/2
Convolutional against transformer backbones for diffusion.
ViT-B/16 vs DiT-XL/2
A vision transformer against a diffusion transformer.
Simple RNN vs Mamba SSM Block
The recurrent layer everyone started with against its modern replacement.
1D CNN + LSTM vs PatchTST
Convolution-plus-recurrence against patched attention for time series.
EEGNet vs EEG Conformer
A compact EEG convnet against a convolution-plus-transformer one.
Whisper Small vs T5 Small
Speech-to-text against text-to-text, both encoder-decoder.
Neural Collaborative Filtering vs Neural Collaborative Filtering
Neural collaborative filtering against its generalised-matrix-factorisation hybrid.
Wide & Deep vs Neural Collaborative Filtering
The memorisation-plus-generalisation recommender against a purely neural one.
LightGCN vs GraphSAGE Recommender
A stripped graph convolution against a sampled one.
Two-Tower vs Neural Collaborative Filtering
Retrieval by dot product against a learned interaction.
DLRM vs Wide & Deep
Two production ranking models with different interaction layers.
Behavior Sequence Transformer vs SLi-Rec
Two sequential recommenders with different notions of time.
LLaVA-1.5 vs CLIP ViT-B/32
A vision-language model against the encoder it is built on.
For machines
/diff/index.json lists every comparison with its deltas. Each page has a
markdown twin at /diff/<key>.md. To compare two graphs of your own,
POST https://www.neurarch.com/api/v1/plan with one as model and the other as base.