v3.0.0

Released September 8, 2026

A major release: use Ask Sage with the OpenAI and Anthropic SDKs, plus a reworked administration, access‑control, and token‑governance experience.

Forward‑only upgrade. This release includes a database change and cannot be rolled back to the previous version once applied. Self‑hosted and Ask Sage Edge deployments should plan accordingly; managed (SaaS) tenants are upgraded by Ask Sage.

Confirming this release. The OpenAI‑ and Anthropic‑compatible API endpoints accept requests from the standard SDKs, and role‑based access controls are available to administrators.

✨ New

Use Ask Sage with the OpenAI, Anthropic, and Google SDKs. New OpenAI‑ and Anthropic‑compatible API gateways let you point the stock SDKs (and Claude Code) straight at Ask Sage — with tool‑calling, vision, and true streaming — alongside an updated Google (Gemini) passthrough.
Response provenance. Each answer can show the model, dataset, persona, and agent that produced it, carried into Copy and Save‑as‑DOCX exports.
Classification‑tier and data‑control gating. On government/DoD deployments, model availability can be enforced by the deployment’s security tier.
Role‑based access control. A new user role model with optional server‑side enforcement for Agent Builder, plugins, API passthrough, and MCP tools.

💬 Chat & workspace

Shareable chat deep links, in‑place editing of saved prompt‑library templates, clearer error messages, a model picker that only shows models your deployment can serve, and more reliable PowerPoint export.

🛠️ Administration & governance

Per‑user token ceilings within a shared pool, a Purchased‑vs‑Allocated view with oversubscription, real‑time 80%/90% usage alerts, and organization‑level token defaults.
Password‑change hardening, active/disabled/suspended account status, organization IP allowlists, pre‑registration, and richer user management. API keys can authenticate against admin endpoints with per‑key scopes and self‑service revocation.
Role‑aware dataset permissions, per‑connector integration status with Box added as a content provider, a privileged‑action audit trail, deployment‑wide and per‑organization model/policy controls, and editable per‑widget branding.

Bug fixes

Reliability and billing‑accuracy improvements across the platform.

🔒 Security & platform

Ask Sage services are delivered as FIPS‑hardened, digest‑pinned, and cosign‑signed images within the FedRAMP Class D authorization boundary, and are continuously rebuilt against the latest patched base images as part of Ask Sage’s continuous‑monitoring commitment.