| Vision tasks | Object detection, instance and semantic segmentation, classification, keypoints, OCR, depth estimation, and multimodal tasks | Legacy vision models; pre-trained detection, OCR, and moderation models remain, but the Nebius license excluded Clarifai's legacy computer vision models |
| Cloud inference | Hosted APIs with autoscaling | Yes, Compute Orchestration on serverless or dedicated GPU clusters, with OpenAI-compatible endpoints |
| Train your own model | Yes, RF-DETR, YOLO (v8 / v11 / v26), Detectron2, and more | Yes, training templates including MMDetection YOLOX, YOLOF, and Faster R-CNN; single-GPU training at $4/hr |
| AI-assisted labeling | Yes, Auto Label labels the classes you describe in a text prompt; Smart Polygon powered by Meta AI's Segment Anything; Label Assist pre-annotates with your own trained models | Yes, auto-annotation tasks run a model or workflow over your inputs |
| Labeling team workflow | Yes, Admin, Labeler, and Reviewer roles with assignment tracking; annotation history with version revert | Yes, labeling tasks assigned to workers, with manual or consensus review |
| Export and own model weights | Yes | Enterprise only; full model exports are listed on the Hybrid-Cloud AI Enterprise plan |
| Industrial integrations | PLC Reader, PLC Writer, and OPC UA blocks speak OPC UA, Modbus TCP, and EtherNet/IP | Not documented |
| Deployment options | Managed cloud, customer VPC, on-premises, and fully air-gapped / offline | Shared or dedicated cloud, your VPC, on-premises, or edge; air-gapped is optional on Enterprise |
| MCP | Yes, every trained model is exposed over MCP so frontier models can call it as a specialized vision tool | Yes, hosts custom MCP servers, and AI Runners connect local models, MCP servers, and agents through its API |
| Uptime SLA | Yes, on Enterprise, with a public status page | Yes, 99.99% SLA on Enterprise |
| Support | 24/7, with a support SLA and a named Customer Support Engineer on Enterprise Support | 24/7 dedicated support on Enterprise; email support on paid plans |
| Primary focus | Computer vision: labeling, training, and deploying vision models | Fast GPU inference and reasoning for large models such as Kimi K2.5 and Llama 4 Scout |
| Self-hosted inference | Yes, open-source inference server for custom and foundation models | Yes, Local Runners connect locally run models through Clarifai's API |
| SDKs and APIs | Yes, Python SDK plus REST/HTTP APIs | Yes, Python and Node.js SDKs, gRPC clients in seven more languages, and a REST channel |
| Compliance | SOC 2 Type II, HIPAA with BAAs, PCI DSS | SOC 2 Type II |
| Free tier | Yes, with free self-hosted inference, model weights download, and a private workspace | Yes, Community plan with a limited number of monthly requests on Clarifai's hosted API |
| Pricing | Published plans; Enterprise adds volume pricing and an uptime SLA | Published plans: Developer (promotional $1/mo), Essential from $30/mo, Professional from $300/mo; Enterprise by quote |