Laguna S 2.1 MXFP4 by olka-fi

 »  All LLMs  »  olka-fi  »  Laguna S 2.1 MXFP4   URL Share it on

Laguna S 2.1 MXFP4 is an open-source language model by olka-fi. Features: 64.3b LLM, VRAM: 66GB, Context: 1024K, License: openmdw-1.1, LLM Explorer Score: 0.32.

  8-bit Base model:poolside/laguna-s-2... Base model:quantized:poolside/...   Compressed-tensors   Conversational   Custom code   Laguna   Laguna-s-2.1   Moe   Mxfp4   Quantized   Region:us   Safetensors   Sharded   Tensorflow   Vllm

Laguna S 2.1 MXFP4 Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Laguna S 2.1 MXFP4 (olka-fi/Laguna-S-2.1-MXFP4)

Laguna S 2.1 MXFP4 Parameters and Internals

LLM NameLaguna S 2.1 MXFP4
Repository 🤗https://huggingface.co/olka-fi/Laguna-S-2.1-MXFP4 
Base Model(s)  poolside/Laguna-S-2.1   poolside/Laguna-S-2.1
Model Size64.3b
Required VRAM66 GB
Updated2026-07-27
Maintainerolka-fi
Model Typelaguna
Model Files  5.1 GB: 1-of-46   3.5 GB: 2-of-46   1.4 GB: 3-of-46   1.4 GB: 4-of-46   1.4 GB: 5-of-46   1.4 GB: 6-of-46   1.4 GB: 7-of-46   1.4 GB: 8-of-46   1.4 GB: 9-of-46   1.4 GB: 10-of-46   1.4 GB: 11-of-46   1.4 GB: 12-of-46   1.4 GB: 13-of-46   1.4 GB: 14-of-46   1.4 GB: 15-of-46   1.4 GB: 16-of-46   1.4 GB: 17-of-46   1.4 GB: 18-of-46   1.4 GB: 19-of-46   1.4 GB: 20-of-46   1.4 GB: 21-of-46   1.4 GB: 22-of-46   1.4 GB: 23-of-46   1.4 GB: 24-of-46   1.4 GB: 25-of-46   1.4 GB: 26-of-46   1.4 GB: 27-of-46   1.4 GB: 28-of-46   1.4 GB: 29-of-46   1.4 GB: 30-of-46   1.4 GB: 31-of-46   1.4 GB: 32-of-46   1.4 GB: 33-of-46   1.4 GB: 34-of-46   1.4 GB: 35-of-46   1.4 GB: 36-of-46   1.4 GB: 37-of-46   1.4 GB: 38-of-46   1.4 GB: 39-of-46   1.4 GB: 40-of-46   1.4 GB: 41-of-46   1.4 GB: 42-of-46   1.4 GB: 43-of-46
Model ArchitectureLagunaForCausalLM
Licenseopenmdw-1.1
Context Length1048576
Model Max Length1048576
Vocabulary Size100352
Torch Data Typebfloat16

Rank the Laguna S 2.1 MXFP4 Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 55446 in total.

Our Social Media →  
Original data from HuggingFace, Arena and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a