What are the hardware requirements for Laguna S 2.1 FP8?

Laguna S 2.1 FP8 requires approximately 116.8 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 FP8 and how large is it?

Laguna S 2.1 FP8 is developed by poolside, a model with 117.6b 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 FP8?

Laguna S 2.1 FP8 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 FP8 by poolside — VRAM 116.8GB, 1024K context

Name: Laguna S 2.1 FP8
Author: poolside

Laguna S 2.1 FP8 is an open-source language model by poolside. Features: 117.6b LLM, VRAM: 116.8GB, Context: 1024K, License: openmdw-1.1, LLM Explorer Score: 0.37.

Base model:poolside/laguna-s-2... Base model:quantized:poolside/... Conversational Custom code Fp8 Laguna Laguna-s-2.1 Region:us Safetensors Sharded Tensorflow Vllm

Model Card on HF 🤗: https://huggingface.co/poolside/Laguna-S-2.1-FP8

Laguna S 2.1 FP8 Parameters and Internals

LLM Name	Laguna S 2.1 FP8
Repository 🤗	https://huggingface.co/poolside/Laguna-S-2.1-FP8
Base Model(s)	poolside/Laguna-S-2.1 poolside/Laguna-S-2.1
Model Size	117.6b
Required VRAM	116.8 GB
Updated	2026-08-08
Maintainer	poolside
Model Type	laguna
Model Files	0.2 GB: 1-of-49 0.4 GB: 2-of-49 4.1 GB: 3-of-49 5.1 GB: 4-of-49 5.1 GB: 5-of-49 5.0 GB: 6-of-49 4.9 GB: 7-of-49 3.6 GB: 8-of-49 2.6 GB: 9-of-49 2.6 GB: 10-of-49 2.6 GB: 11-of-49 2.6 GB: 12-of-49 2.6 GB: 13-of-49 2.6 GB: 14-of-49 2.6 GB: 15-of-49 2.6 GB: 16-of-49 2.6 GB: 17-of-49 2.6 GB: 18-of-49 2.6 GB: 19-of-49 2.6 GB: 20-of-49 2.6 GB: 21-of-49 2.6 GB: 22-of-49 2.6 GB: 23-of-49 2.6 GB: 24-of-49 2.6 GB: 25-of-49 2.6 GB: 26-of-49 2.6 GB: 27-of-49 2.6 GB: 28-of-49 2.6 GB: 29-of-49 2.6 GB: 30-of-49 2.6 GB: 31-of-49 2.6 GB: 32-of-49 2.6 GB: 33-of-49 2.6 GB: 34-of-49 2.6 GB: 35-of-49 2.6 GB: 36-of-49 2.6 GB: 37-of-49 2.6 GB: 38-of-49 2.6 GB: 39-of-49 2.6 GB: 40-of-49 2.6 GB: 41-of-49 2.6 GB: 42-of-49
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 FP8 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

Best Alternatives to Laguna S 2.1 FP8

Best Alternatives	Context / RAM	Downloads	Likes
Laguna S 2.1	1024K / 214.2 GB	56445	708
Laguna S 2.1 INT4	1024K / 100.2 GB	191066	26
Laguna S 2.1	1024K / 214.2 GB	569	7
Laguna S 2.1 Heretic	1024K / 219.3 GB	45	0
Laguna S 2.1 OptiQ 2bit	1024K / 44.2 GB	380	2
Laguna S 2.1 MLX 4bit	1024K / 65.8 GB	493	3
Laguna S 2.1 MLX 6bit	1024K / 95.4 GB	457	2
Laguna S 2.1 AWQ INT4	1024K / 78.2 GB	533	4

Note: green Score (e.g. "73.2") means that the model is better than poolside/Laguna-S-2.1-FP8.