What are the hardware requirements for ArcANE 32B SFT?

ArcANE 32B SFT requires approximately 65.8 GB of VRAM and supports a context window of 40K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed ArcANE 32B SFT and how large is it?

ArcANE 32B SFT is developed by holi-lab, a model with 32b 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 ArcANE 32B SFT?

ArcANE 32B SFT is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

ArcANE 32B SFT by holi-lab — VRAM 65.8GB, 40K context

Name: ArcANE 32B SFT
Author: holi-lab

ArcANE 32B SFT is an open-source language model by holi-lab. Features: 32b LLM, VRAM: 65.8GB, Context: 40K, License: apache-2.0, LLM Explorer Score: 0.33.

Arxiv:2606.05553 Base model:finetune:qwen/qwen3... Base model:qwen/qwen3-32b Character Conversational Dataset:holi-lab/arcane-data En Endpoints compatible Qwen3 Region:us Role-playing Safetensors Sft Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/holi-lab/ArcANE-32B-SFT

ArcANE 32B SFT Parameters and Internals

LLM Name	ArcANE 32B SFT
Repository 🤗	https://huggingface.co/holi-lab/ArcANE-32B-SFT
Base Model(s)	Qwen3 32B Qwen/Qwen3-32B
Model Size	32b
Required VRAM	65.8 GB
Updated	2026-09-06
Maintainer	holi-lab
Model Type	qwen3
Model Files	4.9 GB: 1-of-14 4.9 GB: 2-of-14 4.9 GB: 3-of-14 4.9 GB: 4-of-14 4.9 GB: 5-of-14 4.9 GB: 6-of-14 4.9 GB: 7-of-14 4.9 GB: 8-of-14 4.9 GB: 9-of-14 4.9 GB: 10-of-14 4.9 GB: 11-of-14 4.9 GB: 12-of-14 4.9 GB: 13-of-14 2.1 GB: 14-of-14
Supported Languages	en
Model Architecture	Qwen3ForCausalLM
License	apache-2.0
Context Length	40960
Model Max Length	40960
Transformers Version	4.57.6
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
Errors	replace

Rank the ArcANE 32B SFT 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 ArcANE 32B SFT

Best Alternatives	Context / RAM	Downloads	Likes
...ll Qwen3 32B Preview7 QAT 200K	195K / 65.8 GB	10	3
...ll Qwen3 32B Preview6 QAT 200K	195K / 65.8 GB	6	2
...ll Qwen3 32B Preview4 QAT 200K	195K / 65.8 GB	13	2
Qwen3 32B AWorld	128K / 65.8 GB	37	15
Qwen3 32B	40K / 65.6 GB	6155624	715
Qwen3 32B	40K / 65.8 GB	5411	15
ArcANE 32B DPO	40K / 65.8 GB	241	1
OpenThinkerAgent 32B	40K / 65.8 GB	9022	9
Awakening Goedel V2 32B Sft	40K / 65.8 GB	569	0
Awakening Goedel V2 32B Rl	40K / 65.8 GB	565	0

Note: green Score (e.g. "73.2") means that the model is better than holi-lab/ArcANE-32B-SFT.