From 178c52a9bf87396fde1adcaa7b4880f5b2cfd7e8 Mon Sep 17 00:00:00 2001 From: Claude Code Date: Sat, 7 Feb 2026 11:41:27 +0000 Subject: [PATCH] docs: update for PR #53 - Add changelog entries for fal.ai queue migration, caching, and bug fixes - Document 48-hour localStorage caching for models and schemas - Document manual refresh button in model browser - Add troubleshooting note for stale model lists - Update models-pricing guide with caching information Co-Authored-By: Claude Sonnet 4.5 --- pages/changelog.mdx | 20 ++++++++++++++++++++ pages/guides/models-pricing.mdx | 2 ++ pages/guides/troubleshooting.mdx | 2 ++ 3 files changed, 24 insertions(+) diff --git a/pages/changelog.mdx b/pages/changelog.mdx index 99af4b8..6358b1a 100644 --- a/pages/changelog.mdx +++ b/pages/changelog.mdx @@ -4,6 +4,26 @@ All notable changes to Node Banana are documented here. --- +## Unreleased + +### Added + +- Model browser refresh button to manually clear model and schema caches ([#53](https://github.com/shrimbly/node-banana/pull/53)) +- 48-hour localStorage caching for model lists and schema parameters to reduce API calls ([#53](https://github.com/shrimbly/node-banana/pull/53)) +- 20MB upload size guard to prevent memory spikes when uploading images ([#53](https://github.com/shrimbly/node-banana/pull/53)) + +### Changed + +- fal.ai generation now uses async queue-based execution with polling instead of blocking requests ([#53](https://github.com/shrimbly/node-banana/pull/53)) +- Base64 images are now uploaded to fal.ai CDN before submission to avoid large inline payloads ([#53](https://github.com/shrimbly/node-banana/pull/53)) + +### Fixed + +- WaveSpeed video generation now saves with correct file extension ([#53](https://github.com/shrimbly/node-banana/pull/53)) +- Updated Kie and WaveSpeed provider badge icons for visual consistency ([#53](https://github.com/shrimbly/node-banana/pull/53)) + +--- + ## 1.3.0 — 2026-01-31 ### Added diff --git a/pages/guides/models-pricing.mdx b/pages/guides/models-pricing.mdx index 5babcbf..571f83b 100644 --- a/pages/guides/models-pricing.mdx +++ b/pages/guides/models-pricing.mdx @@ -66,6 +66,8 @@ Replicate and fal.ai provide access to a wide variety of image generation models - **Image-to-Image** — Many models support image input for style transfer and modifications - **Pricing** — Varies by model; check provider documentation for current rates - **Recently Used** — Quick access to your frequently used models +- **Caching** — Model lists and schemas are cached for 48 hours to reduce API calls +- **Manual Refresh** — Use the refresh button in the model browser to clear caches and reload fresh data Use the model selector in Generate Image nodes to browse and search available models from all providers. diff --git a/pages/guides/troubleshooting.mdx b/pages/guides/troubleshooting.mdx index b2ba534..e121fe6 100644 --- a/pages/guides/troubleshooting.mdx +++ b/pages/guides/troubleshooting.mdx @@ -242,6 +242,8 @@ import { Callout } from 'nextra/components' **Timeout errors**: Some models (especially video) may exceed default timeouts. The app includes extended timeout support for video generation. +**Stale model list**: If models appear outdated or missing, use the refresh button in the model browser (next to the capability dropdown) to clear caches and reload. + ### Video generation issues **Video won't play**: Ensure the output is connected to an Output node, which supports video playback.