What are the hardware requirements for MemReranker 4B?

MemReranker 4B requires approximately 8.8 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 MemReranker 4B and how large is it?

MemReranker 4B is developed by IAAR-Shanghai, a model with 4b 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 MemReranker 4B?

MemReranker 4B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

MemReranker 4B by IAAR-Shanghai — VRAM 8.8GB, 40K context

Name: MemReranker 4B
Author: IAAR-Shanghai

MemReranker 4B is an open-source language model by IAAR-Shanghai. Features: 4b LLM, VRAM: 8.8GB, Context: 40K, License: apache-2.0, LLM Explorer Score: 0.24.

Arxiv:2605.06132 Agent Base model:finetune:qwen/qwen3... Base model:qwen/qwen3-reranker... Cross-encoder En Endpoints compatible Memory Qwen3 Region:us Reranker Safetensors Sentence-transformers Text-embeddings-inference Zh

Model Card on HF 🤗: https://huggingface.co/IAAR-Shanghai/MemReranker-4B

MemReranker 4B Parameters and Internals

LLM Name	MemReranker 4B
Repository 🤗	https://huggingface.co/IAAR-Shanghai/MemReranker-4B
Base Model(s)	Qwen3 Reranker 4B Qwen/Qwen3-Reranker-4B
Model Size	4b
Required VRAM	8.8 GB
Updated	2026-07-29
Maintainer	IAAR-Shanghai
Model Type	qwen3
Model Files	8.8 GB 0.0 GB
Supported Languages	en zh
Model Architecture	Qwen3ForCausalLM
License	apache-2.0
Context Length	40960
Model Max Length	40960
Transformers Version	5.5.4
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151669
Errors	replace

Rank the MemReranker 4B 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 MemReranker 4B

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3 4B Instruct 2507	256K / 8.1 GB	3109972	911
GRPO 4 70	256K / 8.1 GB	12	0
FastContext 1.0 4B SFT	256K / 8.1 GB	5735	357
...2.5 Flash Lite Preview Distill	256K / 8.1 GB	31	1
Lightning 4B	256K / 8.1 GB	13	6
Dwight 4B	256K / 8.1 GB	374	10
SRA RiskGate 4B	256K / 8 GB	2662	1
Qwen3 4B Thinking 2507	256K / 8.1 GB	291023	606
Qwen3 4B Deslop V4	256K / 8.1 GB	128	8
...orm3200 Impl Verl S11 Step 039	256K / 8.1 GB	454	1

Note: green Score (e.g. "73.2") means that the model is better than IAAR-Shanghai/MemReranker-4B.