The request names a specific timm checkpoint (ConvNeXt-Small, ImageNet-22k pretrained, fine-tuned on ImageNet-1k at 384px), not a product, so the answer is a general choice of where to run or fetch it. Hugging Face hosts model weights and provides the inference and hub tooling for them 4, and Replicate runs open-source models behind a cloud API 10. Since the request does not say whether you want to download the weights or run inference on a platform, a short question would clarify the best f
A question back: Do you want to download the weights and run them locally, or run inference through a hosted API, and if hosted, which platform do you prefer?
- 1
Hugging Facehuggingface.coverified and signed
Hosts open models and datasets and offers an SDK and CLI to load and run them, including image classification models 4.
official:MCP server · @huggingface/mcp-clientSDK · huggingface-hubCLI · text-embeddings-inferencerepository · pwc-cliskill · hf-cli+23 community tools built for it
for agents: image_classification · pretrained-models · hf-mem
- 2
Replicatereplicate.comverified and signed
Runs open-source machine learning models such as image models through a cloud API without managing your own hardware 10.
+14 community tools built for it
for agents: lightweight-ai/model1 · microsoft/phi-3-mini-4k-instruct · replicate/resnet
Sources
- 4Hugging Face – The AI community building the future.huggingface.co · read 2026-10-29
- 10Replicate - Run AI with an APIreplicate.com · read 2026-10-29
For agents: the 10 actions behind this answer
nano-banana-pro-openrouter · nano-banana-pro-openrouter verified and signed
Generate or edit images via OpenRouter with the Gemini 3 Pro Image model. Use for prompt-only image generation, image edits, and multi-image compositing; supports 1K/2K/4K output.
agent-platform-deploy · agent-platform-deploy verified and signed
Deploy open models or custom weights from Model Garden to Agent Platform endpoints, check the status of an in-progress deployment operation, or clean up resources by undeploying models and deleting endpoints. Use when asked to actively deploy a model, list the Model Garden CATALOG of available model
deleteModel · OpenAI API verified and signed
Delete a fine-tuned model. You must have the Owner role in your organization.
image_classification · Hugging Face (Activepieces) verified and signed
Image Classification — Classify images with pre-trained models or custom categories - perfect for content moderation, automated tagging, and smart asset management
tao-train-image-classification · tao-train-image-classification verified and signed
PyTorch-based TAO image classification. Supports a wide range of backbones (FAN, EfficientNet, ResNet, etc.) with distillation and quantization for deployment. Use when training, evaluating, distilling, quantizing, exporting, or running inference for a TAO image-classification (PyT) model. Trigger p
train_node_classification_model · neo4j-gds found in code, not yet checked
Train a node classification model on a projected graph and store it in the GDS model catalog under modelName.
codec · ebcdic verified and signed
MXNet · sagemaker verified and signed
python · fastcore verified and signed
lightweight-ai/model1 · lightweight-ai/model1 verified and signed
flux_schnell model img2img inference
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.