What are the hardware requirements for SpatialBlock 7B Direct?

SpatialBlock 7B Direct requires approximately 16.6 GB of VRAM and supports a context window of 125K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed SpatialBlock 7B Direct and how large is it?

SpatialBlock 7B Direct is developed by rsoohyun, a model with 7b 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 SpatialBlock 7B Direct?

SpatialBlock 7B Direct is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

SpatialBlock 7B Direct by rsoohyun — VRAM 16.6GB, 125K context

Name: SpatialBlock 7B Direct
Author: rsoohyun

SpatialBlock 7B Direct is an open-source language model by rsoohyun. Features: 7b LLM, VRAM: 16.6GB, Context: 125K, License: apache-2.0, Instruction-Based.

Arxiv:2609.07064 Base model:finetune:qwen/qwen2... Base model:qwen/qwen2.5-vl-7b-... Conversational Dataset:rsoohyun/spatialblock-... Endpoints compatible Image-text-to-text Instruct Qwen2 5 vl Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/rsoohyun/SpatialBlock-7B-direct

SpatialBlock 7B Direct Parameters and Internals

LLM Name	SpatialBlock 7B Direct
Repository 🤗	https://huggingface.co/rsoohyun/SpatialBlock-7B-direct
Base Model(s)	Qwen/Qwen2.5-VL-7B-Instruct Qwen/Qwen2.5-VL-7B-Instruct
Model Size	7b
Required VRAM	16.6 GB
Updated	2026-09-13
Maintainer	rsoohyun
Model Type	qwen2_5_vl
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-4 5.0 GB: 2-of-4 4.9 GB: 3-of-4 1.7 GB: 4-of-4
Model Architecture	Qwen2_5_VLForConditionalGeneration
License	apache-2.0
Context Length	128000
Model Max Length	128000
Transformers Version	4.57.6
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	152064
Errors	replace

Rank the SpatialBlock 7B Direct 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 SpatialBlock 7B Direct

Best Alternatives	Context / RAM	Downloads	Likes
OmniLong Qwen2.5 VL 7B	512K / 16.6 GB	7	2
Qwen2.5 VL 7B Instruct	125K / 16.7 GB	8740905	1666
OlmOCR 2 7B 1025 FP8	125K / 10 GB	540426	253
OlmOCR 2 7B 1025	125K / 16.6 GB	244728	157
RolmOCR	125K / 16.6 GB	224280	589
Qwen2.5 VL 7B Instruct FP4	125K / 7.2 GB	49165	3
SpaSEViLA Cold Start 7B	125K / 16.6 GB	25	0
Cosmos Reason1 7B	125K / 16.6 GB	62913	244
OlmOCR 7B 0825 FP8	125K / 10 GB	38011	10
Zoom IQA 7B	125K / 16.6 GB	14	1

Note: green Score (e.g. "73.2") means that the model is better than rsoohyun/SpatialBlock-7B-direct.