Teams, single sign-on, GPU selection, and database backup
Role-based access control
Users can now work together in teams. Agents, fine-tuning jobs, deployments, files, and vector databases belong to a team, and every member of the team can use them based on their role.
- Organization – The top-level container for every user, team, and resource in your deployment.
- Teams – Groups of users who share resources. Users can belong to more than one team and switch between them.
- Roles – Organizations have Owner and Member roles. Teams have Admin and Creator roles.
Each user also gets a personal workspace that only they can see. The upgrade moves each existing user’s resources into their personal workspace. For details, see Role-based access control.Organization setup and single sign-on for sfe
sfe org setup creates the first administrator and your organization in one step. If your deployment uses Keycloak, it also creates the Keycloak realm and binds it to the organization. New users who sign in through single sign-on then join that organization. sfe org setup can also create a break-glass administrator for signing in when single sign-on is unavailable.Administrators can sign in to sfe with their single sign-on account using sfe org login. For steps, see Create the first SeekrFlow administrator.GPU selection for fine-tuning
You can now choose which GPU types fine-tuning jobs run on. Set the types in your values file, then label your GPU nodes to match. For steps, see Create the SeekrFlow values file.Database backup and restore
You can now back up every SeekrFlow database with sfe db backup and restore it with sfe db restore. Reverting an upgrade doesn’t undo its database changes. Restoring a backup does. For steps, see Revert a SeekrFlow upgrade.Upgrade considerations
- Back up your databases first – The upgrade changes the database schema and moves stored files to team locations, which can take hours for a large file store.
- Use
sfe 3.7.0 or later – Backup, restore, sfe org setup, and sfe org login need it.
- Expect agents to be briefly unavailable – Agents are unavailable while their database migrates.
- Label your GPU nodes – Fine-tuning jobs stay pending until their nodes carry
seekr.io/gpu_type, or until global.gpu.nodeSelectorKey names your nodes’ existing label.
- Tools and AI-Ready Data now deploy by default – Set
enabled: false for any you don’t want. File search needs OpenSearch, and web search needs a Brave Search API key.
- Enable
sfe sign-in on existing realms – Run sfe org sso setup --reconcile.
- Remove
Explainability.images.runner – If your values file sets it, remove it. The image no longer exists.
Serve inference through AWS Bedrock and Azure AI Foundry
Cloud service provider inference adds support for Azure AI Foundry, alongside AWS Bedrock.AWS Bedrock support extended to ingestion pipelines
SeekrFlow ingestion pipelines now support AWS Bedrock for embedding generation and PDF-to-Markdown conversion.
- Embedding generation for vector databases and RAG workflows can now run on AWS Bedrock.
- PDF-to-Markdown conversion uses Bedrock-compatible vision models, supporting AI-ready data jobs for synthetic dataset generation and downstream fine-tuning.
- Using a
bedrock:-prefixed model automatically routes embedding requests to AWS Bedrock.
- Vector databases created with Bedrock models use Titan embeddings for indexing and retrieval.
This update expands platform availability and adds new capabilities across agents, orchestration, and deployment environments.New platform supportSeekrFlow now supports Oracle Cloud Infrastructure (OCI). Organizations can deploy SeekrFlow on OCI environments running NVIDIA GPUs, expanding deployment options for customers with Oracle cloud infrastructure requirements. OCI support currently includes:
- SeekrFlow UI
- SeekrFlow API
Agents
- Tool Library support for agents
- Multi-Agentic Orchestration (Beta) — agents can coordinate across multiple tools and tasks
Platform
- OCI deployment support via Helm
- vLLM upgrade improving inference performance on NVIDIA GPUs
Updates for self-hosted SeekrFlow deployments on AWS, AWS GovCloud, and Appliance environments.Agents and inference:
- AWS Bedrock inference support for agents with cloud service provider-hosted inference
- Full agent execution including tools, streaming, and promotion/demotion workflows
- Multi-LoRA inference with hot loading
vLLM upgrade for improved performance and stability- CSP inference routing to Bedrock endpoints based on region and model availability
Fine-tuning:
- LoRA support for reinforcement (GRPO) and instruction fine-tuning (NVIDIA-based)
Explainability:
- Inference-time data attribution
UI and documentation:
- Dark mode support
- Documentation updates covering new inference, agent, and fine-tuning capabilities
AWS Bedrock inference for agents
AWS Bedrock inference support for SeekrFlow Agents, providing an alternative to Seekr-hosted GPU inference. Agents can now use Claude 4.5 via AWS Bedrock for all intra-agent inference (planner, executor, evaluator, summarizer).Capabilities:
- Full end-to-end agent execution including Agent Chat in the UI, streaming and non-streaming runs, and multi-tool execution
- Bedrock models appear in existing agent model-selection UI
- Configure at deploy time via Helm (EKS-only)
- Deployments select either Bedrock inference or Seekr-hosted GPU inference
Current limitations:
- Agents only (Model Chat is disabled when Bedrock is enabled)
- File search and vector-backed retrieval not yet supported for Bedrock agents
- Reasoning and speed optimization parameters are ignored to ensure consistent execution
Last modified on October 16, 2026
