Cohere adds third-party open models to its Model Vault inference platform
Cohere added open models such as DeepSeek V4.1 Flash, Gemma 4, GPT-OSS and Nemotron to Model Vault, its single-tenant managed inference platform. Zero data retention is on by default.
By Cora Whitley, GTM AI Analyst, Chief of Staff
Image: Cohere press materials, from the announcement
What we are seeing
Cohere announced on November 3 that Model Vault and Model Vault Encrypted now run third-party open models alongside Cohere's own. Model Vault offers DeepSeek V4.1 Flash, GLM 5.3 Flash, MiMo V2.6 Pro, Gemma 4, NVIDIA Nemotron, GPT-OSS, Phi and Tencent's Hy3 in single-tenant, region-selected instances that Cohere manages. Model Vault Encrypted uses confidential computing so prompts and outputs stay encrypted during processing, and supports Gemma, GPT-OSS and Phi. Cohere targets banks, hospitals and government departments. Model Vault is available in the US, EU and Canada, while the encrypted version is US only for now. Pricing is hourly by GPU type, with reserved capacity through sales.
Discuss this with peers in the ForgeX CommunityReport an error
Vendors and readers can flag a factual error in this story, or ask for an image or logo to be removed. Email Cora, the editor, with the link. Image and logo removals happen within one business day. For factual corrections, Cora replies with next steps and a ForgeX lead approves the fix before it is published.
