N Neurarch Architectures Models Checks Data Docs Open the app

Comparisons / LLaMA-3 Block vs Mixtral MoE Block

LLaMA-3 Block vs Mixtral MoE Block

A dense feed-forward against a mixture of experts, at the block level.

Mixtral MoE Block has 749M more parameters than LLaMA-3 Block: 1 layer added, 2 removed, 2 changed.

Baseline

LLaMA-3 Block

Layers
8
Parameters
703M
Input
1 × 2048
Output
1 × 2048 × 4096
Forward-passes
yes
Est. train cost
$14.27
Compared

Mixtral MoE Block

Layers
7
Parameters
1.5B
Input
1 × 4096 × 4096
Output
1 × 4096 × 4096
Forward-passes
yes
Est. train cost
$63.07

The deltas

Every number is Mixtral MoE Block relative to LLaMA-3 Block.

Parameters
+749M (+107%)
Layers
-1
Added
1
Removed
2
Changed
2
Unchanged
6

Which GPUs each one fits

Each side is measured at its own declared input (1 × 2048 against 1 × 4096 × 4096). Both columns are right about their own model; the difference between them is not a fact about the designs.

GPULLaMA-3 BlockMixtral MoE Block
T4 16GBfitsno
A100 40GBfitsfits
H100 80GBfitsfits

Layer by layer

Aligned in topological order. 6 of 11 rows are the same layer with the same parameters.

Hide all 11 rows
LLaMA-3 BlockMixtral MoE Block
LayerParamsOutputLayerParamsOutput
1changed
shape
tokens
Input
1 × 2048hidden_states
Input
1 × 4096 × 4096
2changed
maxSeqLen
rope
Rope
rotary_emb
Rope
3removedembed
Embedding
525M1 × 2048 × 4096
4–7same4 unchanged layers
8removedswiglu_ffn
Swiglu
135M1 × 2048 × 4096
9addedblock_sparse_moe
Moe Layer
1.4B1 × 4096 × 4096
10sameresidual_2
Add
1 × 2048 × 4096moe_residual
Add
1 × 4096 × 4096
11samehidden_state
Output
1 × 2048 × 4096hidden_out
Output
1 × 4096 × 4096

What this is not

Take it further

Open either graph in the editor, change it, and check it again: LLaMA-3 Block · Mixtral MoE Block

Compare any two models of your own, including anything on Hugging Face: the comparison tool.

Machine-readable: this page as markdown · the pair index · POST https://www.neurarch.com/api/v1/plan for a graph of your own.

Related comparisons

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.
LLaMA-3 Block vs Phi-3 Mini Block
A block from a large model against a whole small one.