What are the hardware requirements for Spark X2.5 4B?

Spark X2.5 4B requires approximately 8.3 GB of VRAM and supports a context window of 1024K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Spark X2.5 4B and how large is it?

Spark X2.5 4B is developed by XHToken, 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 Spark X2.5 4B?

Spark X2.5 4B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Spark X2.5 4B by XHToken — VRAM 8.3GB, 1024K context

Name: Spark X2.5 4B
Author: XHToken

Spark X2.5 4B is an open-source language model by XHToken. Features: 4b LLM, VRAM: 8.3GB, Context: 1024K, License: apache-2.0, LLM Explorer Score: 0.39.

Base model:finetune:xhtoken/sp... Base model:xhtoken/spark-x2.5-... Conversational Custom code En Region:us Safetensors Sharded Spark2 5 Sparkx2 5 Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/XHToken/Spark-X2.5-4B

Spark X2.5 4B Parameters and Internals

LLM Name	Spark X2.5 4B
Repository 🤗	https://huggingface.co/XHToken/Spark-X2.5-4B
Base Model(s)	XHToken/Spark-X2.5-4B-Base XHToken/Spark-X2.5-4B-Base
Model Size	4b
Required VRAM	8.3 GB
Updated	2026-09-02
Maintainer	XHToken
Model Type	spark2_5
Model Files	2.0 GB: 1-of-5 2.0 GB: 2-of-5 2.0 GB: 3-of-5 2.0 GB: 4-of-5 0.3 GB: 5-of-5
Supported Languages	en zh
Model Architecture	Spark2_5ForCausalLM
License	apache-2.0
Context Length	1048576
Model Max Length	1048576
Transformers Version	4.57.1
Tokenizer Class	PreTrainedTokenizerFast
Beginning of Sentence Token	<｜start▁of▁sentence｜>
End of Sentence Token	<｜end▁of▁sentence｜>
Unk Token	<unk>
Vocabulary Size	131072

Rank the Spark X2.5 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

Quantized Models of the Spark X2.5 4B

Model	Likes	Downloads	VRAM
Spark X2.5 4B MLX 8bit	3	858	4 GB
Spark X2.5 4B MLX 4bit	2	617	2 GB

Best Alternatives to Spark X2.5 4B

Best Alternatives	Context / RAM	Downloads	Likes
Spark X2.5 4B Base	1024K / 8.3 GB	16	5
Spark X2.5 4B MLX 8bit	1024K / 4.4 GB	858	3
Spark X2.5 4B MLX 4bit	1024K / 2.3 GB	617	2