What are the hardware requirements for Agnes 2.5 Pro Alpha?

Agnes 2.5 Pro Alpha requires approximately 271.9 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Agnes 2.5 Pro Alpha and how large is it?

Agnes 2.5 Pro Alpha is developed by Agnes-AI, a model with 397b 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 Agnes 2.5 Pro Alpha?

Agnes 2.5 Pro Alpha is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Agnes 2.5 Pro Alpha by Agnes-AI — VRAM 271.9GB

Name: Agnes 2.5 Pro Alpha
Author: Agnes-AI

Agnes 2.5 Pro Alpha is an open-source language model by Agnes-AI. Features: 397b LLM, VRAM: 271.9GB, License: apache-2.0, LLM Explorer Score: 0.29.

Agnes-ai Api Base model:finetune:qwen/qwen3... Base model:qwen/qwen3.5-397b-a... Conversational En Image-text-to-text Long-context Multimodal Qwen3 5 moe Reasoning Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/Agnes-AI/Agnes-2.5-Pro-Alpha

Agnes 2.5 Pro Alpha Parameters and Internals

LLM Name	Agnes 2.5 Pro Alpha
Repository 🤗	https://huggingface.co/Agnes-AI/Agnes-2.5-Pro-Alpha
Base Model(s)	Qwen/Qwen3.5-397B-A17B Qwen/Qwen3.5-397B-A17B
Model Size	397b
Required VRAM	271.9 GB
Updated	2026-08-21
Maintainer	Agnes-AI
Model Type	qwen3_5_moe
Model Files	4.3 GB: 1-of-122 4.3 GB: 2-of-122 8.6 GB: 3-of-122 4.6 GB: 4-of-122 8.6 GB: 5-of-122 4.6 GB: 6-of-122 8.6 GB: 7-of-122 4.3 GB: 8-of-122 8.6 GB: 9-of-122 4.8 GB: 10-of-122 8.6 GB: 11-of-122 4.6 GB: 12-of-122 8.6 GB: 13-of-122 4.6 GB: 14-of-122 8.6 GB: 15-of-122 4.3 GB: 16-of-122 8.6 GB: 17-of-122 4.8 GB: 18-of-122 8.6 GB: 19-of-122 4.6 GB: 20-of-122 8.6 GB: 21-of-122 4.6 GB: 22-of-122 8.6 GB: 23-of-122 4.3 GB: 24-of-122 8.6 GB: 25-of-122 4.8 GB: 26-of-122 8.6 GB: 27-of-122 4.6 GB: 28-of-122 8.6 GB: 29-of-122 4.6 GB: 30-of-122 8.6 GB: 31-of-122 4.6 GB: 32-of-122 8.6 GB: 33-of-122 4.3 GB: 34-of-122 8.6 GB: 35-of-122 4.8 GB: 36-of-122 8.6 GB: 37-of-122 4.6 GB: 38-of-122 8.6 GB: 39-of-122 4.6 GB: 40-of-122 8.6 GB: 41-of-122 4.3 GB: 42-of-122
Supported Languages	en zh
Model Architecture	Qwen3_5MoeForConditionalGeneration
License	apache-2.0
Transformers Version	5.12.1

Rank the Agnes 2.5 Pro Alpha 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 Agnes 2.5 Pro Alpha

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3.5 397B A17B	0K / 369.8 GB	331990	1549
Qwen3.5 397B A17B NVFP4	0K / 251.1 GB	428565	102
Qwen3.5 397B A17B NVFP4 V2	0K / 243.5 GB	23781	5
Ornith 1.0 397B FP8	0K / 147.7 GB	597119	183
Ornith 1.5 397B	0K / 279.9 GB	672	66
Ornith 1.0 397B	0K / 288.5 GB	256425	260
XYZ Aquila Pro	0K / 225 GB	1805	375
Ornith 1.5 397B NVFP4	0K / 237.9 GB	2533	15
Qwen3.5 397B A17B	0K / 369.8 GB	53	13
Qwen3.5 397B A17B FP8	0K / 184.9 GB	825515	182

Note: green Score (e.g. "73.2") means that the model is better than Agnes-AI/Agnes-2.5-Pro-Alpha.