What are the hardware requirements for Spark?

Spark requires a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Spark and how large is it?

Spark is developed by SurjoLabs, a model with 7.1m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Spark perform on standard benchmarks?

Spark has the following published scores: LMSYS ELO 1489. Compare against reference models on this page or on the LLM Explorer leaderboards.

Spark by SurjoLabs — 2K context

Name: Spark
Author: SurjoLabs

Spark is an open-source language model by SurjoLabs. Features: 7.1m LLM, Context: 2K, License: mit, LLM Explorer Score: 0.43, ELO: 1489.

Conversational Custom-architecture Custom code En Region:us Safetensors Slm Small-language-model Spark

Model Card on HF 🤗: https://huggingface.co/SurjoLabs/Spark

Spark Benchmarks

LMSys ELO 1489 vs 1272 (so35) 17.1%

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Spark Parameters and Internals

LLM Name	Spark
Repository 🤗	https://huggingface.co/SurjoLabs/Spark
Model Size	7.1m
Required VRAM	0 GB
Updated	2026-08-23
Maintainer	SurjoLabs
Model Type	spark
Model Files	0.0 GB 0.0 GB 0.0 GB
Supported Languages	en
Model Architecture	SparkForCausalLM
License	mit
Context Length	2048
Model Max Length	2048
Transformers Version	5.15.1
Vocabulary Size	4096

Rank the Spark 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