[Upstream Update] Gateway API Inference Extension v1.3.0 → v1.3.1

[github-actions]

Summary

Gateway API Inference Extension has released v1.3.1 (published 2026-02-20), which includes bug fixes cherry-picked from recent PRs.

Current Pin: v1.3.0
Latest Release: v1.3.1 (2026-02-20)
Status: Repository is ~1 day behind latest release

Release Highlights

Bug Fixes (v1.3.1)

This patch cherry-picks fixes from:

• #2321
• #2300
• #2316

Notable Features in v1.3.0

• LoRA Syncer Deprecation: LoRA syncer images will no longer be released (use vLLM's file system resolver instead)
• Flow Control: Scale from/to Zero support for autoscaling
• Standalone EPP: Experimental feature for batch inference scenarios

Impact Assessment

Impact Level: MEDIUM

Affected Files:

• guides/prereq/gateway-provider/install-gateway-provider-dependencies.sh line 46
• All guides/*/helmfile.yaml.gotmpl files (InferencePool chart version)

Breaking Changes: None - this is a patch release with bug fixes

LoRA Syncer Impact: llm-d should migrate to vLLM's [file system resolver]((docs.vllm.ai/redacted) if using LoRA management

Recommended Action

Update Gateway API Inference Extension CRD version and InferencePool chart references:

1. CRD Installation Script:

   # In guides/prereq/gateway-provider/install-gateway-provider-dependencies.sh
   GATEWAY_API_INFERENCE_CRD_REVISION=\$\{GATEWAY_API_INFERENCE_CRD_REVISION:-"v1.3.1"}

2. Helm Charts (all guides/*/helmfile.yaml.gotmpl files):

   version: v1.3.1

Testing: Verify InferencePool functionality with existing model deployments, especially if using Flow Control or EPP features.

References

• Upstream Release: https://github.com/kubernetes-sigs/gateway-api-inference-extension/releases/tag/v1.3.1
• Current Pin: guides/prereq/gateway-provider/install-gateway-provider-dependencies.sh line 46
• vLLM File System Resolver: (docs.vllm.ai/redacted)

---

Generated by Upstream Dependency Monitor

Generated by Upstream Dependency Monitor