@shop-os/modelkit 0.7.6 → 0.7.8

package/CHANGELOG.md

@@ -9,6 +9,18 @@ header in the release PR.
 
 ## [Unreleased]
 
+## [0.7.8] - Unreleased
+
+ENG-3324: Ray 3.2 offers 10 s clips.
+
+### Changed
+- `ray-3.2` offers `durations: [5, 10]` (was `[5]`); the default stays 5 s. The Luma adapter sends the requested length (`"10s"`) instead of a fixed `"5s"`, and a generate reports the block it asked for (5 or 10) as `durationSeconds`, never the measured frames. A generate of any other length is refused with `unsupported_option` before a request, since nothing prices it. Luma bills a 10 s clip at 3x the 5 s price; pro-backend prices it through its billed-seconds map (`{"10": 15}`), so usage keeps the real 10 s. Consumers must not offer this version before that pricing is deployed, or a 10 s clip bills 2x.
+
+## [0.7.7] - Unreleased
+
+### Fixed
+- Google `prepareMediaPart` inlines bytes only up to 10 MiB (was 20 MB) and uploads larger media through the Files API. Inline bytes are base64-encoded, so a 15 MB video made a ~20 MB request and Gemini refused it or timed out; on dev no inline video between 10.9 MB and 20 MB succeeded. Media of 10-20 MB now uploads (a few seconds slower, same tokens) and the returned `cleanup` deletes the file as before.
+
 ## [0.7.6] - Unreleased
 
 ENG-3323: one xAI reference-mode rule.

package/README.md

@@ -309,7 +309,7 @@ const defaultKit = createDefaultModelKit({ keys: { pro: keysFromEnv(process.env)
 | Retry       | One classifier, one policy. Never on content blocks, aborts, or 4xx except 429; SDKs with their own retry (Replicate) run under a single kit attempt; a 429 on a status poll keeps polling unless `CallContext.maxRetries` is `0`. See durable media execution above.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   |
 | Errors      | `ModelKitError` with `code`, `provider`, `status`, `body`. Gemini `blockReason` and fal `body.detail` are lifted into `body`. A throw with no HTTP status (fetch failed, reset, DNS) is a retryable `network` error on every adapter. An out-of-credit provider account is `provider_billing` and is never retried: HTTP 402, a 4xx whose structured error code is a billing code (`insufficient_quota`, ...), or a 403/429 whose text names the exhausted account ("Exhausted balance", "Insufficient credit"). Text calls go through `ai` and surface the SDK's own error instead.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   |
 | Usage       | One event per call with the canonical model id and the row's `usage.unit` (adapters supply the quantity in that unit). Gemini image events carry the token counts too (`inputTokens`/`outputTokens`); Google's raw `usageMetadata` rides on `raw.providerMetadata.google`. Video usage is emitted on completion and carries the resolution the provider renders (the default it falls back to when the request has none), because pricing is keyed by it; Seedance keeps `usage`, `duration`, `framespersecond` and `resolution` from the poll on `raw`. Audio: Lyria 3 returns no duration, so it emits `unit: "images", quantity: 1` (token counts ride along as `inputTokens`/`outputTokens`); Replicate audio (Lyria 2, MMAudio) emits `unit: "compute_seconds"` with the prediction time.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   |
-| Media parts | `prepareMediaPart` turns a `MediaRef` into an AI SDK file part. Google inlines bytes up to 20 MB and uploads larger payloads through the Files API; call the returned `cleanup` after the request. Other providers get the reference inline.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            |
+| Media parts | `prepareMediaPart` turns a `MediaRef` into an AI SDK file part. Google inlines bytes up to 10 MiB (base64 makes them ~4/3 larger inside Gemini's 20 MB request cap) and uploads larger payloads through the Files API; call the returned `cleanup` after the request. Other providers get the reference inline.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            |
 
 ## Add a model
 

package/dist/catalog.cjs

@@ -5201,8 +5201,10 @@ var RAY_RESOLUTIONS = ["360p", "720p", "1080p"];
 var RAY_PROMPT = { referenceSyntax: "numbered", maxChars: 6e3 };
 var LUMA_VIDEO_MODELS = [
   {
-    // https://docs.agents.lumalabs.ai: type "video" with a fixed 5 s duration;
-    // firstFrame and lastFrame ride as start_frame and end_frame.
+    // https://docs.agents.lumalabs.ai: type "video" with a 5 s or 10 s duration;
+    // firstFrame and lastFrame ride as start_frame and end_frame. Luma bills a
+    // 10 s clip at 3x the 5 s price; pro-backend prices it that way, so this row
+    // must not offer 10 s before that pricing is deployed (ENG-3324).
     id: "ray-3.2",
     aliases: [],
     label: "Ray 3.2",
@@ -5214,7 +5216,7 @@ var LUMA_VIDEO_MODELS = [
     constraints: {
       aspectRatios: ["adaptive", ...RAY_RATIOS],
       resolutions: [...RAY_RESOLUTIONS],
-      durations: [5],
+      durations: [5, 10],
       supports: ["lastFrame"],
       prompt: { ...RAY_PROMPT },
       audio: "never"
@@ -5224,7 +5226,7 @@ var LUMA_VIDEO_MODELS = [
     pricingKind: "video_per_second",
     display: {
       provider: "luma",
-      description: "Luma Ray 3.2 text- and image-to-video with optional first and last frames, 5-second clips at up to 1080p.",
+      description: "Luma Ray 3.2 text- and image-to-video with optional first and last frames, 5- or 10-second clips at up to 1080p.",
       logo: LUMA_LOGO2
     },
     capabilities: {}