Earth-2 and FourCastNet are not directly represented by any of the candidate sources. The NVIDIA source 5 is a Brev.dev console page, and its indexed actions reference Earth2Studio, the NVIDIA Earth-2 toolkit for weather and climate forecasting, which can run deterministic forecast models. The sources do not show FourCastNet itself, so the closest verifiable route is NVIDIA's Earth-2 tooling via 5, with Hugging Face 9 as a general place to host and run open weather models.
- 1
NVIDIAnvidia.comverified and signed
Its indexed Earth2Studio action builds deterministic weather forecast scripts (model, data source, inference), which is the Earth-2 forecasting workflow relevant to FourCastNet 5.
official:MCP server · NVIDIA CUDA Docs+84 community tools built for it
for agents: forecasting · forecasting · earth2studio-deterministic-forecast
- 2
Hugging Facehuggingface.coverified and signed
Hosts open machine learning models and datasets for download and collaboration, where open weather forecasting models can be found and used 9.
official:MCP server · @huggingface/mcp-clientSDK · huggingface-hubCLI · text-embeddings-inferencerepository · pwc-cliskill · hf-cli+23 community tools built for it
for agents: huggingface-vision-trainer · deploy_model · models
- 3
Replicatereplicate.comverified and signed
Runs open-source machine learning models through a cloud API, which can serve a model such as FourCastNet if it is packaged there 10.
+14 community tools built for it
for agents: ai-forever/kandinsky-2 · black-forest-labs/flux-2-max · fofr/latent-consistency-model
Sources
- 5Brev.dev Consolebrev.nvidia.com · read 2026-10-29
- 9Hugging 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
model-recommendation · model-recommendation verified and signed
Analyze chatmode or prompt files and recommend optimal AI models based on task complexity, required capabilities, and cost-efficiency
get_forecast · Python MCP Weather Server with OAuth 2.1 Authentication found in code, not yet checked
Get weather forecast for a location
train_link_prediction_model · neo4j-gds found in code, not yet checked
Train a link prediction model on a projected graph and store it in the GDS model catalog under modelName.
4.0 · looker-sdk verified and signed
forecasting · Earth-2 FourCastNet verified and signed
Task: Forecasting
python · fastcore verified and signed
maps_elevation · MCP-Brave-Search found in code, not yet checked
Get elevation data for locations on the earth
workers-ai-post-run-model · Cloudflare API verified and signed
Run a Workers AI model
huggingface-vision-trainer · huggingface-vision-trainer verified and signed
Trains and fine-tunes vision models for object detection (D-FINE, RT-DETR v2, DETR, YOLOS), image classification (timm models — MobileNetV3, MobileViT, ResNet, ViT/DINOv3 — plus any Transformers classifier), and SAM/SAM2 segmentation using Hugging Face Transformers on Hugging Face Jobs cloud GPUs. C
ai-forever/kandinsky-2 · ai-forever/kandinsky-2 verified and signed
text2img model trained on LAION HighRes and fine-tuned on internal datasets
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.