What are the hardware requirements for Sekai 1?

Sekai 1 requires approximately 0.4 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Sekai 1 and how large is it?

Sekai 1 is developed by 8BitStudio, a model with 400b 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 Sekai 1?

Sekai 1 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Sekai 1 by 8BitStudio — VRAM 0.4GB, 2K context

Name: Sekai 1
Author: 8BitStudio

Sekai 1 is an open-source language model by 8BitStudio. Features: 400b LLM, VRAM: 0.4GB, Context: 2K, License: apache-2.0, LLM Explorer Score: 0.35.

Chatml Conversational Dataset:allenai/dolma3.5 pool Dataset:allenai/dolma3 dolmino... Dataset:allenai/dolmino-mix-11... Dataset:allenai/tulu-3-sft-mix... Dataset:bigcode/commitpackft Dataset:huggingfaceh4/ultracha... Dataset:huggingfacetb/cosmoped... Dataset:huggingfacetb/finemath Dataset:huggingfacetb/smollm-c... Dataset:huggingfacetb/smoltalk Dataset:huggingfacetb/smoltalk... Dataset:ifm/txt360-v2 Dataset:karpathy/climbmix-400b... Dataset:meta-math/metamathqa Dataset:microsoft/orca-math-wo... Dataset:nvidia/nemotron-mind Dataset:nvidia/openmathinstruc... Dataset:openbmb/ultradata-math Dataset:openbmb/ultrax-preview Dataset:opencoder-llm/opc-anne... En Endpoints compatible Gqa Muon Qwen3 Region:us Rope Safetensors Swiglu

Model Card on HF 🤗: https://huggingface.co/8BitStudio/Sekai-1

Sekai 1 Parameters and Internals

LLM Name	Sekai 1
Repository 🤗	https://huggingface.co/8BitStudio/Sekai-1
Model Size	400b
Required VRAM	0.4 GB
Updated	2026-11-10
Maintainer	8BitStudio
Model Type	qwen3
Model Files	0.4 GB
Supported Languages	en
Model Architecture	Qwen3ForCausalLM
License	apache-2.0
Context Length	2048
Model Max Length	2048
Transformers Version	5.15.1
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<\|endoftext\|>
Vocabulary Size	24576
Torch Data Type	float32

Rank the Sekai 1 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 Sekai 1

Best Alternatives	Context / RAM	Downloads	Likes
Mantis 50M	1K / 0.2 GB	36	0