Skip to content

AI Model Gateway v1.4.4

Latest

Choose a tag to compare

@github-actions github-actions released this 22 May 23:13
· 113 commits to main since this release

AI Model Gateway v1.4.4

AI Model Gateway is a self-hosted LLM operations gateway for teams that want provider routing, fallback, telemetry, benchmarking, config publishing, diagnostics, updates, and rollback inside their own environment.

This release keeps the public project surface consistent across repository metadata, README, issue templates, Admin UI copy, release assets, and promotion material so new visitors can understand the project faster.

Highlights

  • Clearer project positioning around a self-hosted LLM operations gateway rather than a hosted model marketplace.
  • Repository metadata, issue templates, README, release notes, and promotion assets are aligned around the same message.
  • Admin UI naming and language fallback are cleaned up so users do not see untranslated placeholders.
  • Old CLI i18n dead code and local generated noise were removed.
  • Embedded Admin UI assets were refreshed so release bundles and repository assets stay consistent.
  • Provider fallback and health documentation now includes an executable demo that proves a primary 429 can be served through a fallback provider with route_mode=model_fallback telemetry.
  • The new project roadmap makes current direction, near-term priorities, and contribution areas explicit for evaluators.
  • The new 15-minute evaluation path gives high-intent visitors a short fit, runtime, fallback-demo, and operations-workflow trial.
  • Repository topics were aligned around LLM gateway discovery terms such as llm-gateway, openai-compatible, rate-limiting, and config-management.

Start Here

Downloads

  • Windows x64: ai-model-gateway-windows-amd64.zip
  • Linux x64: ai-model-gateway-linux-amd64.tar.gz
  • Linux arm64: ai-model-gateway-linux-arm64.tar.gz
  • macOS arm64: ai-model-gateway-darwin-arm64.tar.gz
  • Checksums: SHA256SUMS.txt

Verification And Upgrade