What are the hardware requirements for AReaL SEA 235B A22B?

AReaL SEA 235B A22B requires approximately 95.8 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed AReaL SEA 235B A22B and how large is it?

AReaL SEA 235B A22B is developed by inclusionAI, a model with 235b 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 AReaL SEA 235B A22B?

AReaL SEA 235B A22B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

AReaL SEA 235B A22B by inclusionAI — VRAM 95.8GB, 256K context

Name: AReaL SEA 235B A22B
Author: inclusionAI

AReaL SEA 235B A22B is an open-source language model by inclusionAI. Features: 235b LLM, VRAM: 95.8GB, Context: 256K, License: apache-2.0, LLM Explorer Score: 0.23.

Arxiv:2505.24298 Arxiv:2601.22607 Agent Areal Base model:finetune:qwen/qwen3... Base model:qwen/qwen3-235b-a22... Conversational En Grpo Multi-turn Qwen3 moe Region:us Safetensors Sharded Tau2-bench Tensorflow Tool-use

Model Card on HF 🤗: https://huggingface.co/inclusionAI/AReaL-SEA-235B-A22B

AReaL SEA 235B A22B Parameters and Internals

LLM Name	AReaL SEA 235B A22B
Repository 🤗	https://huggingface.co/inclusionAI/AReaL-SEA-235B-A22B
Base Model(s)	Qwen3 235B A22B Thinking 2507 Qwen/Qwen3-235B-A22B-Thinking-2507
Model Size	235b
Required VRAM	95.8 GB
Updated	2026-08-08
Maintainer	inclusionAI
Model Type	qwen3_moe
Model Files	2.2 GB: 1-of-196 1.1 GB: 2-of-196 1.1 GB: 3-of-196 1.1 GB: 4-of-196 1.1 GB: 5-of-196 1.1 GB: 6-of-196 1.1 GB: 7-of-196 1.1 GB: 8-of-196 1.1 GB: 9-of-196 1.1 GB: 10-of-196 1.1 GB: 11-of-196 2.2 GB: 12-of-196 2.8 GB: 13-of-196 2.8 GB: 14-of-196 2.8 GB: 15-of-196 2.4 GB: 16-of-196 2.4 GB: 17-of-196 2.4 GB: 18-of-196 2.4 GB: 19-of-196 2.4 GB: 20-of-196 2.4 GB: 21-of-196 2.4 GB: 22-of-196 2.4 GB: 23-of-196 2.4 GB: 24-of-196 2.4 GB: 25-of-196 2.4 GB: 26-of-196 2.4 GB: 27-of-196 2.4 GB: 28-of-196 2.4 GB: 29-of-196 2.4 GB: 30-of-196 2.4 GB: 31-of-196 2.4 GB: 32-of-196 2.4 GB: 33-of-196 2.4 GB: 34-of-196 2.4 GB: 35-of-196 2.4 GB: 36-of-196 2.4 GB: 37-of-196 2.4 GB: 38-of-196 2.4 GB: 39-of-196 2.4 GB: 40-of-196 2.4 GB: 41-of-196 2.4 GB: 42-of-196 2.4 GB: 43-of-196 2.4 GB: 44-of-196 2.4 GB: 45-of-196
Supported Languages	en
Model Architecture	Qwen3MoeForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Transformers Version	4.57.1
Vocabulary Size	151936

Rank the AReaL SEA 235B A22B 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 AReaL SEA 235B A22B

Best Alternatives	Context / RAM	Downloads	Likes
...en3 235B A22B Instruct 2507 1M	986K / 171.9 GB	4	1
Qwen3 235B A22B Instruct 2507	256K / 167.9 GB	28	10
Qwen3 235B A22B Thinking 2507	256K / 167.9 GB	14	3
Qwen3 235B A22B Instruct 2507	256K / 171.9 GB	59472	793
Vivo C V1	256K / 156.4 GB	415	2
... 235B A22B Thinking 2507 NVFP4	256K / 139.2 GB	22828	8
... 235B A22B Instruct 2507 NVFP4	256K / 139.2 GB	14322	10
Qwen3 235B A22B Thinking 2507	256K / 171.9 GB	20387	408
...n3 235B A22B Instruct 2507 FP8	256K / 236.5 GB	54126	147
MiroThinker 1.7 Mini	256K / 61.1 GB	584	104

Note: green Score (e.g. "73.2") means that the model is better than inclusionAI/AReaL-SEA-235B-A22B.