What are the hardware requirements for Agents A1?

Agents A1 requires approximately 70.6 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Agents A1 and how large is it?

Agents A1 is developed by InternScience, a model with 35.1b 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 Agents A1?

Agents A1 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Agents A1 by InternScience — VRAM 70.6GB

Name: Agents A1
Author: InternScience

Agents A1 is an open-source language model by InternScience. Features: 35.1b LLM, VRAM: 70.6GB, License: apache-2.0, LLM Explorer Score: 0.41.

Arxiv:2606.30616 Agentic Conversational Endpoints compatible Eval-results Image-text-to-text Moe Qwen3 5 moe Region:us Safetensors Sharded Tensorflow Vision Vlm

Model Card on HF 🤗: https://huggingface.co/InternScience/Agents-A1

Agents A1 Parameters and Internals

LLM Name	Agents A1
Repository 🤗	https://huggingface.co/InternScience/Agents-A1
Model Size	35.1b
Required VRAM	70.6 GB
Updated	2026-08-12
Maintainer	InternScience
Model Type	qwen3_5_moe
Model Files	5.4 GB: 0-of-14 5.4 GB: 1-of-14 5.4 GB: 2-of-14 5.4 GB: 3-of-14 5.4 GB: 4-of-14 5.4 GB: 5-of-14 5.4 GB: 6-of-14 5.4 GB: 7-of-14 5.4 GB: 8-of-14 5.4 GB: 9-of-14 5.4 GB: 10-of-14 5.4 GB: 11-of-14 5.4 GB: 12-of-14 0.4 GB: 13-of-14
Model Architecture	Qwen3_5MoeForConditionalGeneration
License	apache-2.0
Model Max Length	262144
Transformers Version	4.57.0.dev0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Agents A1 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

Quantized Models of the Agents A1

Model	Likes	Downloads	VRAM
Agents A1 4bit	4	1952	20 GB
Agents A1 8bit	4	1206	37 GB
Agents A1 OptiQ 4bit	3	891	22 GB
Agents A1 6bit	3	443	28 GB
Agents A1 5bit	3	359	24 GB

Best Alternatives to Agents A1

Best Alternatives	Context / RAM	Downloads	Likes
Nex N2.5 Mini	0K / 70.4 GB	3121	685
Nex N2 Mini	0K / 70.4 GB	2289	294
Thomson 1.0 Small	0K / 70.5 GB	214	109
Huihui Agents A1 Abliterated	0K / 70.6 GB	3339	29
Occamy 1.0 FP8	0K / 36.8 GB	270	4
Affine 5g4yy75zuz T6	0K / 70.2 GB	756	0
...yJ5ua3Sebfyr8kcVwLAdaiJLUC8pkW	0K / 70.4 GB	1894	0
Affine 5efg6cm3yl King	0K / 70.3 GB	578	0
Nex N2.5 Mini OQ5	0K / 25.4 GB	112	1
Affine 5gedzafcvg Queen	0K / 70.4 GB	718	0

Note: green Score (e.g. "73.2") means that the model is better than InternScience/Agents-A1.