YOLO or RF-DETR: which to choose for business
Both detect objects in video in real time and run on a CPU. Ultralytics YOLO is the most widespread detector, very small (2.4M to 68M) and fine on weak hardware, but its AGPL-3.0 license means opening your product code or buying an enterprise license. The catalog describes RF-DETR as an open YOLO alternative without AGPL: sizes Nano to Large are Apache 2.0, XL and 2XL use a Roboflow license; it supports segmentation, and keypoints are in preview since 2026-06. The latest YOLO release is YOLO26 (2026-01).
Comparison based on catalog data
| Parameter | YOLO (Ultralytics) | RF-DETR |
|---|---|---|
| Category | Computer vision | Computer vision |
| Developer | Ultralytics, USA | Roboflow, USA |
| Releases | Jan 2023 – Jan 2026 | Mar 2025 – Jun 2026 |
| Sizes | 2.4M – 68M | Nano – 2XL |
| Hardware | Laptop | Laptop, 1 GPU |
| Commercial use | Commercial use with conditions | Commercial use allowed |
| License | AGPL-3.0 (you would have to open your product's code) or a paid Ultralytics enterprise license | Apache 2.0 for Nano to Large sizes; XL and 2XL under Roboflow's own license |
| Russian | Not applicable | Not applicable |
| Ollama | No | No |
| Without GPU | Yes | Yes |
| Tasks |
|
|
Choose YOLO (Ultralytics) if
- You need proven scenarios: counting people and cars, helmets, defects, shelf layout
- The model must run on the weakest hardware: from 2.4M parameters
- You can open your product code or have budget for an Ultralytics license
Choose RF-DETR if
- You cannot open your code and do not want to buy a license: Apache 2.0
- You need precise outlines of parts and defects (segmentation)
- You want to fine-tune the detector on your own object classes
Other comparisons
- YOLO or Grounding DINO: which detector for video analytics
- Florence-2 or Grounding DINO: which to choose for object detection
- BRIA RMBG or BiRefNet: which to use for background removal
- Qwen or GigaChat: which to choose for business
- Qwen or Llama: which to choose for business
- GigaChat or YandexGPT: which to choose for business
- DeepSeek or Qwen: which to choose for business
- Gemma or Llama: which to choose for business
- Mistral or Llama: which to choose for business
- Whisper or GigaAM: which to choose for business
- Whisper or Parakeet: which to choose for business
- Silero or Piper: which to choose for business
Need a model for your task?
An open model can run on your own server: data stays in-house, there is no per-request fee, and the model can be fine-tuned on your documents.
- SelectThe model and size for your task and hardware budget
- DeployOn your server or in a closed network, with an API
- Fine-tuneOn your data, or connect a knowledge base
- IntegrateInto your CRM, ERP, bot, website or team chat