Question 1

Can I run Hugging Face alongside a frontier provider?

Accepted Answer

Yes. The provider is configuration, so routine questions can run on Hugging Face while heavier analysis routes to a frontier provider. A fallback chain with health checks handles the case where one provider is unavailable.

Question 2

What happens when a model is retired?

Accepted Answer

Retired model IDs are remapped to current equivalents automatically, and configured providers are validated against their live model listings. A model that disappears upstream does not appear as a failed agent run.

Question 3

Can I bring my own Hugging Face account?

Accepted Answer

Yes. Provider credentials are configured per tenant, so requests run through your own Hugging Face account and usage counts against your plan. A deployment with no override configured falls back to the default provider.

Question 4

Does using Hugging Face change what CommerceShip can do?

Accepted Answer

No. The agent tools, the data they read, and the actions they can take are the same whichever provider serves the model. The provider is a setting. What changes is cost, latency, and where inference physically runs.

Question 5

Which Hugging Face models can I use?

Accepted Answer

Any model served from a Hugging Face inference endpoint, including a fine-tune you host there. CommerceShip calls the endpoint through the model gateway, so a model tuned on your own contracts or SOPs runs the same way as a stock one.

Question 6

Which other models can I run alongside Hugging Face?

Accepted Answer

Any of them, and more than one at a time. CommerceShip runs ChatGPT, Claude, Gemini, Mistral, and Muse as frontier models, and DeepSeek, Gemma, GLM, Kimi, Llama, MiniMax, Nemotron, and Qwen as open-weight models, served through Ollama Cloud or on your own hardware. Route routine work to one model and heavier analysis to another, and if a provider is rate-limited or down, the run moves to a fallback and completes.

Question 7

What access does CommerceShip need to Hugging Face?

Accepted Answer

The connection uses scoped credentials limited to the record types the integration works with, and you choose which fields are included. Credentials are encrypted at rest and stored separately from your operational data. Revoke the connection from either side at any time and the access ends with it.

Question 8

How is my data protected?

Accepted Answer

TLS in transit and AES-256 at rest, with tenant isolation enforced by row-level security in the database. Role-based permissions control who on your team can see cost data or change carrier settings, and automated actions are written to an audit log. CommerceShip is pursuing SOC 2 Type II certification. The security page carries the full detail.

Run supply chain AI on Hugging Face

Hugging Face Integration Capabilities

How the Hugging Face integration works

Connect Hugging Face

Choose the models

Ask and act

Hugging Face Integration FAQs

Run your models on Hugging Face

Models You Can Run