SAVRN
Search Contact SAVRN

Open-weight model · Image and text to text

MemGUI-8B-RL

by Anonymous memgui-rl/MemGUI-8B-RL

Anonymous release for the ICLR 2027 submission MemGUI-RL: Reinforcement Learning for Proactive Context Management in Long-Horizon Mobile GUI Agents.

Parameters8.8B
Context262,144
Weights17.5 GB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads—

Runs On

What it takes to serve MemGUI-8B-RL (8.8B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

PrecisionWeightsMemory neededCheapest setupPer hourAlso fits
16-bit 17.5 GB 21.0 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
8-bit 8.8 GB 10.5 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
4-bit 4.4 GB 5.3 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

Model Card

By Anonymous, published under apache-2.0, revision cf08cdb9a59f.

Anonymous release for the ICLR 2027 submission MemGUI-RL: Reinforcement Learning for Proactive Context Management in Long-Horizon Mobile GUI Agents. Project page: https://memgui-rl-anonymous.github.io/ MemGUI-8B-RL is MemGUI-8B-SFT (Qwen3-VL-8B-Instruct supervised on MemGUI-3K) post-trained for 100 optimizer steps with FARPO (Folding-Aware Reward-decoupled Policy Optimization, span-to-step ratio rho = 9). The policy speaks the ConAct (Context-as-Action) interface of MemGUI-Agent: every response contains a folding directive for its own history, an optional memory operation and the next GUI action. The checkpoint is a standard Qwen3VLForConditionalGeneration model (weights in bf16, ~17.5 GB).…

Read Anonymous's full model card

Anonymous release for the ICLR 2027 submission MemGUI-RL: Reinforcement Learning for Proactive Context Management in Long-Horizon Mobile GUI Agents. Project page: https://memgui-rl-anonymous.github.io/

MemGUI-8B-RL is MemGUI-8B-SFT (Qwen3-VL-8B-Instruct supervised on MemGUI-3K) post-trained for 100 optimizer steps with FARPO (Folding-Aware Reward-decoupled Policy Optimization, span-to-step ratio rho = 9). The policy speaks the ConAct (Context-as-Action) interface of MemGUI-Agent: every response contains a folding directive for its own history, an optional memory operation and the next GUI action.

benchmark MemGUI-8B-SFT MemGUI-8B-RL
MemGUI-Bench Pass@1 / Pass@3 / IRR (128 tasks) 23.4 / 35.9 / 30.2 24.2 / 43.0 / 35.0
MobileWorld GUI-only success rate (117 tasks, OOD) 17.9 19.7

Usage

The checkpoint is a standard Qwen3VLForConditionalGeneration model (weights in bf16, ~17.5 GB). Use it with the ConAct system prompt and the MemGUI-Agent runtime unchanged:

from transformers import AutoProcessor, Qwen3VLForConditionalGeneration
model = Qwen3VLForConditionalGeneration.from_pretrained("memgui-rl-anonymous/MemGUI-8B-RL", torch_dtype="bfloat16", device_map="auto")
processor = AutoProcessor.from_pretrained("memgui-rl-anonymous/MemGUI-8B-RL")

Training code and launch scripts: https://github.com/memgui-rl-anonymous/MemGUI-RL Training data: https://huggingface.co/datasets/memgui-rl/MemGUI-3K-Verl Evaluation logs: https://huggingface.co/datasets/memgui-rl/MemGUI-RL-Eval

Configuration

Architecture
Qwen3VLForConditionalGeneration
Context length (tokens)
262,144
Layers
36
Hidden size
4,096
Feed-forward size
12,288
Attention heads
32
Key/value heads
8
Head dimension
128
Vocabulary size
151,936
RoPE base
5,000,000
Model type
qwen3_vl

Identity and Version

Repository
memgui-rl/MemGUI-8B-RL
Publisher
Anonymous
Task
Image and text to text
Modality
Image and text
Library
Not stated by the source
Parameters
8.8B parameters
Languages
Not stated by the source
Revision
cf08cdb9a59fcc989f74d523ab07a62dd991cccf
First published
2026-09-18
Last updated
2026-09-18

Files and Weights

18 files, 17.6 GB in total. The weights are 4 files totalling 17.5 GB in safetensors.

Weights4 files · 17.5 GB
Configuration7 files · 72.5 KB
Tokenizer4 files · 15.9 MB
Documentation1 file · 1.8 KB
Other1 file · 5.3 KB
Repository1 file · 1.6 KB
Every file
FileTypeSizeSHA-256
model-00001-of-00004.safetensorsWeights4.9 GB 3c2339d4fbeb
model-00002-of-00004.safetensorsWeights5.0 GB 867a5207319d
model-00003-of-00004.safetensorsWeights3.9 GB 7e31053aad83
model-00004-of-00004.safetensorsWeights3.8 GB 3ad54a032c07
added_tokens.jsonConfiguration707 B —
config.jsonConfiguration1.6 KB —
generation_config.jsonConfiguration147 B —
model.safetensors.index.jsonConfiguration67.8 KB —
preprocessor_config.jsonConfiguration782 B —
special_tokens_map.jsonConfiguration613 B —
video_preprocessor_config.jsonConfiguration817 B —
README.mdDocumentation1.8 KB —
chat_template.jinjaOther5.3 KB —
.gitattributesRepository1.6 KB —
merges.txtTokenizer1.7 MB —
tokenizer.jsonTokenizer11.4 MB aeb13307a71a
tokenizer_config.jsonTokenizer5.5 KB —
vocab.jsonTokenizer2.8 MB —

License and Download

License
apache-2.0
Access
Open weights, no gate
Download size
17.5 GB
Download from Anonymous

Released by Anonymous through its official repository on Hugging Face. Read the license.

Built From

  • Derived from memgui-agent-anonymous/MemGUI-8B-SFT

Memory Requirements

PrecisionWeights in memory
As published17.5 GB
16-bit17.5 GB
8-bit8.8 GB
4-bit4.4 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About MemGUI-8B-RL

How much GPU memory does MemGUI-8B-RL need?

About 21 GB at 16-bit and 5.3 GB at 4-bit: the weights (8.8B parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run MemGUI-8B-RL on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use MemGUI-8B-RL commercially?

Yes. MemGUI-8B-RL is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

What is MemGUI-8B-RL's context length?

262,144 tokens, from the maximum position embeddings in its published configuration.

Similar Models

Model · Image and text to text

Qwen3-VL-8B-Instruct

Qwen

Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date. This generation delivers comprehensive upgrades across the board: superior text understanding & generation, deeper visual perception & reasoning, extended context length, enhanced spatial and video dynamics comprehension, and stronger agent interaction capabilities. Available in Dense and MoE architectures that scale from edge to cloud, with Instruct and reasoning‑enhanced Thinking editions for flexible, on‑demand deployment. Text Understanding on par with pure LLMs: Seamless text–vision fusion for lossless, unified comprehension. 1. Interleaved-MRoPE: Full‑frequency allocation over time, width, and height…

Open weights apache-2.0 8.8B parameters 262,144 tokens transformers
View model

Model · Image and text to text

Qwen3-VL-8B-Instruct-FP8

Qwen

Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date. This generation delivers comprehensive upgrades across the board: superior text understanding & generation, deeper visual perception & reasoning, extended context length, enhanced spatial and video dynamics comprehension, and stronger agent interaction capabilities. Available in Dense and MoE architectures that scale from edge to cloud, with Instruct and reasoning‑enhanced Thinking editions for flexible, on‑demand deployment. Text Understanding on par with pure LLMs: Seamless text–vision fusion for lossless, unified comprehension. 1. Interleaved-MRoPE: Full‑frequency allocation over time, width, and height…

Open weights apache-2.0 8.8B parameters 262,144 tokens transformers
View model

Model · Image and text to text

Qwen2.5-VL-7B-Instruct

Qwen

pipelinetag: image-text-to-text - multimodal libraryname: transformers In the past five months since Qwen2-VL’s release, numerous developers have built new models on the Qwen2-VL vision-language models, providing us with valuable feedback. During this period, we focused on building more useful vision-language models. Today, we are excited to introduce the latest addition to the Qwen family: Qwen2.5-VL. Understanding long videos and capturing events: Qwen2.5-VL can comprehend videos of over 1 hour, and this time it has a new ability of cpaturing event by pinpointing the relevant video segments. Capable of visual localization in different formats: Qwen2.5-VL can accurately localize objects in…

Open weights apache-2.0 8.3B parameters 128,000 tokens transformers
View model

Model · Image and text to text

Qwen2.5-VL-7B-Instruct-AWQ

Qwen

In the past five months since Qwen2-VL’s release, numerous developers have built new models on the Qwen2-VL vision-language models, providing us with valuable feedback. During this period, we focused on building more useful vision-language models. Today, we are excited to introduce the latest addition to the Qwen family: Qwen2.5-VL. Understanding long videos and capturing events: Qwen2.5-VL can comprehend videos of over 1 hour, and this time it has a new ability of cpaturing event by pinpointing the relevant video segments. Capable of visual localization in different formats: Qwen2.5-VL can accurately localize objects in an image by generating bounding boxes or points, and it can provide…

Open weights apache-2.0 8.3B parameters 128,000 tokens transformers
View model

Model · Image and text to text

Qwen2-VL-7B-Instruct-AWQ

Qwen

We're excited to unveil Qwen2-VL, the latest iteration of our Qwen-VL model, representing nearly a year of innovation. SoTA understanding of images of various resolution & ratio: Qwen2-VL achieves state-of-the-art performance on visual understanding benchmarks, including MathVista, DocVQA, RealWorldQA, MTVQA, etc. Understanding videos of 20min+: Qwen2-VL can understand videos over 20 minutes for high-quality video-based question answering, dialog, content creation, etc. Agent that can operate your mobiles, robots, etc.: with the abilities of complex reasoning and decision making, Qwen2-VL can be integrated with devices like mobile phones, robots, etc., for automatic operation based on…

Open weights apache-2.0 8.3B parameters 32,768 tokens transformers
View model

Model · Image and text to text

Qwen2-VL-7B-Instruct

Qwen

We're excited to unveil Qwen2-VL, the latest iteration of our Qwen-VL model, representing nearly a year of innovation. SoTA understanding of images of various resolution & ratio: Qwen2-VL achieves state-of-the-art performance on visual understanding benchmarks, including MathVista, DocVQA, RealWorldQA, MTVQA, etc. Understanding videos of 20min+: Qwen2-VL can understand videos over 20 minutes for high-quality video-based question answering, dialog, content creation, etc. Agent that can operate your mobiles, robots, etc.: with the abilities of complex reasoning and decision making, Qwen2-VL can be integrated with devices like mobile phones, robots, etc., for automatic operation based on…

Open weights apache-2.0 8.3B parameters 32,768 tokens transformers
View model