Laguna S 2.1
Active Parameters
118B
Context Length
1.05M
Modality
Code
Architecture
Mixture of Experts (MoE)
License
openmdw-1.1
Release Date
13 Jul 2026
Knowledge Cutoff
-
System Requirements
VRAM requirements for different quantization methods and context sizes
1,024 tokens
Consumer
13x RTX 4090
24GB VRAM
Datacenter
4x NVIDIA A100
80GB VRAM
Apple Silicon
3x Apple M3 Max
128GB VRAM
1,048,576 tokens
Consumer
25x RTX 4090
24GB VRAM
Datacenter
7x NVIDIA A100
80GB VRAM
Apple Silicon
5x Apple M3 Max
128GB VRAM
Architecture Diagram
Evaluation Benchmarks
No evaluation benchmarks for Laguna S 2.1 available.
Rankings
Overall Rank
-
Coding Rank
-
About Laguna S 2.1
Laguna S 2.1 is an open-weight sparse mixture-of-experts coding agent model created by Poolside. It activates 8B parameters out of 118B total to achieve top-tier performance on complex terminal and software engineering benchmarks.
Technical Specifications
Attention
Attention Structure
Grouped-Query Attention
Attention Heads
48
Key-Value Heads
8
Attention Head Dimension
128
Position Embedding
ROPE
RoPE Theta
500,000
Sliding Window Attention
Yes
Sliding Window Size
512
Sliding Window Ratio
75.0%
Linear Attention
No
Linear Attention Ratio
-
Normalization
RMS Normalization
Activation Function
SwigLU
Dimensions
Hidden Dimension Size
3,072
Number of Layers
48
FFN Intermediate Size (Dense)
12,288
Multi-Token Prediction Heads
-
Tokenizer
Vocabulary Size
100,352
Mixture of Experts
Total Expert Parameters
8.0B
Number of Experts
256
Active Experts
10
Shared Experts
1
FFN Intermediate Size (per Expert)
1,024
Dense Layers Before MoE
1
Resources
About Laguna
The Laguna model family developed by Poolside.
Other Laguna Models
- No related models available