Book a demo

A 30-minute walkthrough on your real providers. We'll deploy CodeVector against your OpenAI or Anthropic key, configure Claude Code or Cursor live, and answer the questions your security team will ask. No slides.

Request deployment access

Tell us about your deployment and we will send Docker registry credentials for the private image and a license token for your environment.

CodeVector release notes

What changed in each version of CodeVector. We skip patch releases when nothing changes for your team.

  1. v1.11.0

    New

    • Change your own password Members can change their password from the account menu without asking an admin or going through email.

  2. v1.10.0

    Kubernetes deployment, backend health monitoring, and a tougher gateway under failure.

    New

    • Helm chart for Kubernetes Install the gateway and run database migrations on any Kubernetes cluster with one Helm release. Read the docs
    • Backend health view Watch the live status of every provider backend, and bring a quarantined backend back online from the console.
    • CLI profile switching Point the CLI at several gateways from one workstation without signing in again each time. Read the docs
    • License-gated container images Pulling the gateway container now checks your license, so only your team can fetch new builds.

    Improved

    • Sturdier OpenAI Responses API streaming Streams close cleanly when an upstream errors out, and unsupported fields are dropped before the request reaches the provider.
    • Repo and machine CLI scopes The CLI configure command keeps repo-level setup separate from machine-wide setup, and updates each one on its own.

    Fixed

    • AWS Bedrock Anthropic compatibility Provider options, cache control, and model routing now match the standard Anthropic API when running through Bedrock.
    • Circuit-breaker reasons in alerts When the gateway trips a circuit on a backend, the alert names the specific failure reason instead of saying the call failed.

  3. v1.9.0

    New

    • Group-scoped usage reports Filter and export usage at the group level, alongside the existing per-user reports. Read the docs
    • PDF export for usage reports Save any usage report as a PDF from the console for sharing or printing.

    Improved

    • Names and emails in reports Usage reports show developer name and email next to the user ID.

  4. v1.8.0

    New

    • OpenAI Responses API support The gateway proxies the OpenAI Responses API, including streaming, on top of chat completions and messages.

    Improved

    • Per-day numbers on the Activity Trend chart Each point on the chart is the value for that day, not a running total from the start of the window.

  5. v1.7.0

    Improved

    • Live audit log with diff view The audit log refreshes every five seconds, and each entry opens a modal with the full before and after diff. Read the docs

    Fixed

    • Rate limits and budget caps actually enforce Per-user caps on requests, tokens, concurrency, and spend block traffic at the gateway. Previously they were only logged. Read the docs

  6. v1.6.0

    Notifications and cache visibility.

    New

    • Notification rules and email alerts Route alerts to email by rule, so each notification lands in the right inbox.
    • Cache and spend visibility Track prompt cache hits and the spend you save on providers that support caching.

    Fixed

    • Stable API key list API keys stay in a consistent order in the list, and rotating a key keeps its name.
    • Steadier retries and responses Upstream retries follow a clearer policy, and gateway responses stay consistent when traffic spikes.

  7. v1.5.0

    New

    • OpenAI-compatible providers Add any OpenAI-compatible endpoint as a provider, including self-hosted models and regional gateways. Read the docs

  8. v1.4.0

    New

    • CLI configure for supported coding tools One command sets up Claude Code, opencode, qwen-code, and codex to use the gateway. Read the docs

  9. v1.3.0

    New

    • Audit log UI Browse and search every admin change from the console. Read the docs
    • Secret rule management Write and edit regex rules from the console to block secrets in prompts before they leave your network. Read the docs
    • Reliable notification delivery Notifications go through an outbox, so a transient email failure retries on its own without dropping the alert.

  10. v1.2.0

    New

    • Admin-set user passwords Admins can set or rotate a user password from the console and copy it to the clipboard. No invite email needed.
    • Auto-generated passwords on user creation When you create a user, the console generates a password you can hand them right away, so they skip the invite acceptance step.

  11. v1.1.0

    First major console expansion. A full admin surface for users, groups, models, providers, and access policies.

    New

    • Groups Create groups, add members, and apply access policies at the group level instead of one user at a time. Read the docs
    • Models and backends in the console Manage the user-facing model names and the upstream provider endpoints they route to. Read the docs
    • API keys self-service Developers create, name, and rotate their own gateway API keys. Read the docs
    • Access grants, rate limits, and budgets Per-user and per-group panels for model access, traffic caps, and spend caps. Read the docs
    • Usage dashboard Charts for requests, tokens, and spend across the team. Read the docs

  12. v1.0.0

    First public release. A self-hosted AI gateway with per-seat governance for engineering teams.