What are the hardware requirements for Laguna S 2.1 MXFP4?

Laguna S 2.1 MXFP4 requires approximately 66 GB of VRAM and supports a context window of 1024K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Laguna S 2.1 MXFP4 and how large is it?

Laguna S 2.1 MXFP4 is developed by olka-fi, a model with 64.3b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Laguna S 2.1 MXFP4?

Laguna S 2.1 MXFP4 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Laguna S 2.1 MXFP4 by olka-fi — VRAM 66GB, 1024K context

Name: Laguna S 2.1 MXFP4
Author: olka-fi

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: 0.29.

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

Model Card on HF 🤗: https://huggingface.co/olka-fi/Laguna-S-2.1-MXFP4

Laguna S 2.1 MXFP4 Parameters and Internals

LLM Name	Laguna 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 Size	64.3b
Required VRAM	66 GB
Updated	2026-07-29
Maintainer	olka-fi
Model Type	laguna
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 Architecture	LagunaForCausalLM
License	openmdw-1.1
Context Length	1048576
Model Max Length	1048576
Vocabulary Size	100352
Torch Data Type	bfloat16

Rank the Laguna S 2.1 MXFP4 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

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