Use YOLOS Small for Free

YOLOS Small is an AI model released 2021 by hustvl, a leading organization in large language model research. Transformer-based YOLO — detects and labels objects in images using the COCO 80-class taxonomy. Compact and fast. With ~6M and a a large context window context window, YOLOS Small is designed to excel across a range of demanding tasks — from software engineering and mathematical problem-solving to multilingual content generation and long-document analysis.

YOLOS Small uses a dense transformer architecture with ~6M of parameters, all of which are active for every input token. Dense architectures provide predictable, consistent behavior across diverse task types — there are no "cold" experts to warm up, and quality is uniform across prompts. hustvl's training pipeline combined large-scale pretraining on a large-scale multilingual pretraining dataset with supervised fine-tuning and alignment techniques (RLHF or DPO), producing YOLOS Small's characteristic balance of capability, helpfulness, and safety.

As of 2026-10-16, YOLOS Small by hustvl demonstrates measurable performance on standard AI evaluation benchmarks. On COCO AP — which measures model performance on specialized evaluation tasks — it achieves 36.1. These scores place YOLOS Small — at ~6M of parameters — in a competitive position within its model class. Note that benchmark scores reflect controlled evaluation conditions; real-world performance varies by task, prompt design, and domain specificity.

The primary strengths of YOLOS Small are: Minimal detection head on DeiT ViT — clean architecture; Apache 2.0; Research model for ViT-based detection. These make it particularly well-suited for Research on ViT-based object detection, Educational demonstrations. Users working on complex, multi-step problems benefit most from YOLOS Small's ability to maintain coherence across long inputs and produce structured, verifiable outputs.

Like all current AI systems, YOLOS Small has limitations: Not production-ready for speed-critical detection; DETR and RTDETR outperform on accuracy. For production deployments, validating outputs against authoritative sources — especially for domain-specific or time-sensitive information — remains best practice. hustvl continues to improve YOLOS Small through ongoing research.

YOLOS Small is available on LITHOVEX AI with a free account. Unlike direct API access — which requires billing setup, API key management, and per-token charges — LITHOVEX AI provides a seamless browser interface with no setup required, making YOLOS Small accessible to students, researchers, developers, and early-stage teams globally.

As of 2026-10-16, YOLOS Small by hustvl demonstrates measurable performance on standard AI evaluation benchmarks. On COCO AP — which measures model performance on specialized evaluation tasks — it achieves 36.1. These scores place YOLOS Small — at ~6M of parameters — in a competitive position within its model class. Note that benchmark scores reflect controlled evaluation conditions; real-world performance varies by task, prompt design, and domain specificity.

YOLOS: 2D object detection on Vision Transformer, small variant

The training pipeline for YOLOS Small involves multiple stages: large-scale pretraining on diverse multilingual text and code corpora to develop broad world knowledge, followed by supervised fine-tuning on instruction-response pairs, and alignment techniques such as RLHF or DPO. Each stage refines the model's helpfulness, accuracy, and safety. hustvl's research team — based in China — has optimized YOLOS Small for the specific capability profile described above.

Inference efficiency: YOLOS Small at ~6M is designed to deliver quality outputs at practical serving speeds. Its context capacity supports complex multi-turn conversations without quality degradation.

Getting started with YOLOS Small on LITHOVEX AI takes under two minutes. Follow these steps:

For best results with YOLOS Small: provide specific, context-rich prompts; include examples when generating code; specify output format when structure matters; break complex tasks into focused sub-questions.

hustvl is a leading organization in AI research and large language model development. Their team has built expertise across natural language processing, model architecture design, and alignment research. hustvl's models are used by researchers, developers, and enterprises worldwide. Their commitment to advancing AI capability while maintaining responsible deployment practices has made them a respected contributor to the global AI ecosystem. The organization continues to publish research and release models that push the boundaries of what AI systems can achieve.

Founded: 2022, headquartered in China.

Visit the official hustvl website →

YOLOS Small by hustvl competes in a field with several strong alternatives. Below is a detailed side-by-side comparison against commonly listed alternatives, using benchmark data and capability profiles in the catalog.

Both YOLOS Small and the models above are available free on LITHOVEX AI — test them side by side on your real tasks before committing to a subscription with any provider.

YOLOS Small is released under Apache 2.0, which means you can download and run it on your own hardware. Below are the quickest ways to get started.

Real-time object detection is a universal computer vision need — from retail inventory management in Japan to traffic monitoring in Singapore to warehouse automation in Germany. YOLOS Small's transformer-based approach provides a reference implementation for modern detection architectures.

YOLOS Small by hustvl has the following version history and notable updates:

For the complete changelog and latest release notes, visit the official hustvl model page.