@shop-os/modelkit 0.1.8 → 0.1.10

package/CHANGELOG.md

@@ -0,0 +1,182 @@
+# Changelog
+
+All notable changes to `@shop-os/modelkit`. The format follows
+[Keep a Changelog](https://keepachangelog.com/en/1.1.0/). Dates are npm publish dates.
+
+A release is a PR that touches only `packages/modelkit/**` and bumps `version`; merging it to
+`dev` publishes. Add an entry under **Unreleased** with the change, and move it under the version
+header in the release PR.
+
+## [Unreleased]
+
+## [0.1.10] - unreleased (publish pending)
+
+### Added
+- Vendor logos on the ByteDance rows (Seedream 5 Pro, Seedance 2.0 Fast, Seedance 2.5) and the
+  Google rows that lacked one (Omni 1.1, Lyria 3). Rows without a logo are the picker's job to
+  cover with a provider mark.
+
+### Fixed
+- `usage.slug: "grok"` sits on the `grok-imagine` row; 0.1.9 shipped it on `kling-2.5` by mistake.
+- **0.1.9 merged the chat stream tuning into every text call.** Text rows carried it as
+  `providerOptions` and `defaults.temperature`, which the kit merges under the caller's options, so
+  Space Builder text-gen started sending Anthropic thinking budgets and Kimi temperature 1. The
+  tuning now lives in `display.chat` (`providerOptions`, `temperature`), which the kit never sends;
+  `capabilities.reasoning` stays. Consumers on 0.1.9 should move to 0.1.10.
+
+## [0.1.9] - 2026-09-12
+
+### Added
+- `display.sortOrder`: pickers show lower values first, then catalog order; set it to surface a
+  new model at the top.
+- `usage.slug` and `usage.providerLabel`: an app's historic usage-event names for a row, so
+  metering continuity no longer needs a per-model table in the app.
+- Text rows carry their stream tuning: `providerOptions` (Anthropic thinking / adaptive, OpenAI
+  reasoning summaries, DeepSeek thinking), `capabilities.reasoning`, and `defaults.temperature`
+  (Kimi rows).
+- `constraints.exactSizes` on the gpt-image-2 rows: the historic WxH per aspect ratio that Cowork
+  sends as `exactSize`.
+
+### Changed
+- `deepseek-flash` (DeepSeek V4.1 Flash) replaces the `deepseek-v4-flash` row; the old id is an
+  alias, so stored selections and plan lists keep resolving. DeepSeek retired V4 Flash upstream and
+  serves the old id from V4.1 Flash at Flash rates.
+- `resolveVideoOptions` enforces conditioning compatibility: a `sourceVideo` on a row without
+  `capabilities.acceptsSourceVideo`, or frames next to references on a row with
+  `framesExcludeReferences`, is an `unsupported_option` error, or is dropped with a warning
+  under `onUnsupported: "coerce"`.
+
+## [0.1.8] - 2026-09-11
+
+### Added
+- `exposure: "catalog" | "internal"` is required on every row. `catalog` rows are offered in
+  product pickers (Creative Studio, chat, Space Builder); `internal` rows are callable by id
+  only. 47 rows are `catalog`. A snapshot test pins the set.
+- `display.costTier` (1 to 3) and `display.recommended` on text rows, for the chat picker.
+- `constraints.qualityOptions` on rows that support `quality`; `constraints.durationPresets`
+  where the UI offers a short list over a duration range. `resolveOptions` ignores both.
+- `@shop-os/modelkit/catalog`: a browser-safe subpath with `Catalog`, `MODELS`, the per-type
+  arrays, `resolveOptions`, the error types and every type. No adapter, client, key handling
+  or provider SDK is reachable from it; a test walks the import graph.
+- Alias `meta-muse` on `muse-image-1.0`.
+
+### Changed
+- Labels: `gemini-3.1-flash-image-preview` is "Nano Banana 2" again and `gpt-image-2` is
+  "GPT-image-2" (the 0.1.7 labels read as different models in the picker).
+- `Catalog` rejects duplicate ids and aliases across rows.
+- `veo-3.1-lite` offers 720p and 1080p only; Google publishes no 4k rate for it.
+
+## [0.1.7] - 2026-09-10
+
+### Added
+- `display { provider, description, logo? }` and `capabilities` (`imageGenerate`,
+  `imageEdit`, `acceptsSourceVideo`, `framesExcludeReferences`, `vision`, `tools`,
+  `structuredOutput`, `contextWindow`) on every active row, so a UI can render a picker from
+  `kit.catalog.list()` without a hand-written catalog. The nine Cowork picker rows keep their
+  descriptions and logos verbatim.
+
+## [0.1.6] - 2026-09-10
+
+### Added
+- `MediaRef.label`: a text part before the media on Gemini image and Omni.
+- `ImageRequest.temperature` (Gemini image) and `ImageRequest.upscale` (Topaz enhance model,
+  subject detection, face enhancement with creativity and strength; face enhancement off by
+  default).
+- `VideoRequest.watermark` (Seedance) and `VideoRequest.safetyChecker` (fal MiniMax).
+- MiniMax base rows accept `lastFrame`; Seedream 4.5 gets the 4K tier and 9:21.
+
+### Changed
+- `onUnsupported: "error"` now covers aspect ratio and duration as well as size and
+  resolution. **Breaking** for callers that relied on silent snapping: pass `"coerce"`.
+- Aborts are never retried and are forwarded to OpenAI, Replicate, Prodia, Ark and reference
+  downloads. One retry policy across adapters.
+- Gateway token is required only for vendors that go through the gateway; the
+  enterprise-to-pro key fallback is logged.
+- Seedance poll keeps usage, duration, fps and resolution on `raw`; the token split lands on
+  the usage event. Grok sends `resolution` only when the caller set it.
+- The unused OpenRouter text route is removed.
+
+## [0.1.5] - 2026-09-10
+
+### Fixed
+- `size: "512"` reaches the adapter (Gemini 3.1 Flash requests were generated at 1K).
+- Every status-less SDK throw is a retryable transport error; five adapters had classified
+  them as permanent.
+- Grok `firstFrame` maps to the SDK's first frame (image-to-video); references capped at 7.
+- Omni: `delivery: "uri"` and `store: true`, `firstFrame` runs image-to-video, references run
+  reference-to-video, no `resolution` on 1.0, uploads deleted after create.
+
+### Changed
+- `constraints.supports` undefined means "supports nothing"; every row carries an explicit list.
+- `aspectRatio: "adaptive"` means "send no ratio" and is never coerced or defaulted.
+- Usage `unit` comes from the catalog row; adapters supply quantities.
+- Gemini `safetySettings` are per-call opt-in.
+- fal direct video is queue submit plus poll, only the submit is retried; the fal client takes
+  the kit's `fetch`. Cloudflare gateway slugs `anthropic/v1` and `grok/v1`.
+
+## [0.1.4] - 2026-09-10
+
+### Added
+- `topaz` row takes references and `scaleFactor` and returns `outputUrl` next to the bytes.
+- Gemini image usage events carry input and output token counts.
+- Aliases `claude-haiku-4` and `claude-sonnet-4-5-20250929`.
+
+## [0.1.3] - 2026-09-10
+
+### Fixed
+- The remaining medium findings from the adoption review.
+
+## [0.1.2] - 2026-09-10
+
+### Added
+- `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare` with the `xhigh` and `max` quality tiers.
+- Aliases `gpt-5.1-mini` and `gpt-5.2-mini` on `gpt-5.4-mini`.
+
+### Changed
+- Google text goes through the gateway at `/google-ai-studio/v1beta`.
+- Reasoning defaults are dropped from the text catalog.
+- Seedance and Kling Omni keep last frames and seeds; Google transport failures retry.
+
+### Fixed
+- The seven blocking findings from the adoption review.
+
+## [0.1.1] - 2026-09-09
+
+### Added
+- `generateText`, `streamText`, `generateObject`, `streamObject` with catalog models. The
+  package re-exports every `ai` export, so `from "ai"` becomes `from "@shop-os/modelkit"`.
+- `generateAudio` (Lyria 3, Lyria 2, MMAudio) and `prepareMediaPart` for video understanding.
+- `generateImage` accepts the `upscale` and `bg-remove` catalog types, `n`, exact `WxH` sizes.
+- Catalog rows for every id the apps still sent to direct SDKs; retired Claude 4.0 ids alias to
+  Sonnet 4.6 and Opus 4.8; fal passthrough with a gateway-stored key.
+
+### Fixed
+- Text keys fall back per provider; dotted Claude aliases; native DeepSeek, Moonshot and
+  Perplexity providers; AI SDK-shaped `providerOptions`.
+- gpt-image-2 edit sent `input_fidelity`, which the API rejects; a filtered Gemini prompt maps
+  to `content_blocked`; `@google/genai` no longer switches to Vertex from the host env.
+
+## [0.1.0] - 2026-09-09
+
+First publish, renamed from `@shop-os/model-sdk`.
+
+- Catalog of 74 models. The id is the pro-backend pricing key; every historical id is an alias.
+- `resolveOptions` fills defaults, snaps the aspect ratio, errors or coerces on unsupported
+  size, resolution and duration, trims references.
+- Ten adapters by provider API: AI SDK text, Gemini image, gpt-image-2, Google GenAI (Veo,
+  Omni, Lyria 3, video understanding), Ark, fal, Replicate, Prodia, Higgsfield, xAI video.
+- Keys per tier (pro, enterprise with per-provider fallback); one retry classifier, one error
+  type, one usage event with the canonical id.
+
+[Unreleased]: https://github.com/Shop-OS/shopos-backend-ts/compare/modelkit-0.1.10...dev
+[0.1.10]: https://github.com/Shop-OS/shopos-backend-ts/compare/f17e822e2...dev
+[0.1.9]: https://github.com/Shop-OS/shopos-backend-ts/compare/1958a5615...dev
+[0.1.8]: https://github.com/Shop-OS/shopos-backend-ts/commit/9661a109d
+[0.1.7]: https://github.com/Shop-OS/shopos-backend-ts/commit/2d3d19ac2
+[0.1.6]: https://github.com/Shop-OS/shopos-backend-ts/commit/f67fee08f
+[0.1.5]: https://github.com/Shop-OS/shopos-backend-ts/commit/7f3f745ec
+[0.1.4]: https://github.com/Shop-OS/shopos-backend-ts/commit/a2f3ffcc7
+[0.1.3]: https://github.com/Shop-OS/shopos-backend-ts/commit/785960dbd
+[0.1.2]: https://github.com/Shop-OS/shopos-backend-ts/commit/4581615e6
+[0.1.1]: https://github.com/Shop-OS/shopos-backend-ts/commit/c159c97d3
+[0.1.0]: https://github.com/Shop-OS/shopos-backend-ts/commit/6b70c91cf

package/README.md

@@ -181,3 +181,7 @@ Prompt construction, agent tools, node schemas and UI, S3 upload, credit reserve
 ## Live smoke test
 
 `pnpm smoke [filter...]` runs one cheap real call per adapter (about $5 for the full matrix, 15 minutes, mostly video polling). Keys load from the `trigger-workflows`, `cowork`, and `vibe-triggers` env files next to this monorepo. Outputs land in `scripts/smoke-out/`. Run it before every release and after any adapter change. Never wire it into CI.
+
+## Changelog
+
+See [CHANGELOG.md](./CHANGELOG.md). Add an entry under Unreleased with every change; the release PR moves it under the version header.

package/dist/catalog.cjs

@@ -165,7 +165,7 @@ function geminiImage(id, label, aliases, sizes, display, exposure) {
       supports: ["safetySettings", "temperature"]
     },
     defaults: { size: "1K", aspectRatio: "1:1" },
-    usage: { ...IMAGE_USAGE },
+    usage: { ...IMAGE_USAGE, slug: "gemini" },
     pricingKind: "image_size",
     display,
     capabilities: { ...IMAGE_GENERATE_EDIT }
@@ -233,6 +233,15 @@ var IMAGE_MODELS = [
     lifecycle: "active",
     exposure: "catalog",
     constraints: {
+      exactSizes: {
+        "1:1": "1024x1024",
+        "3:2": "1536x1024",
+        "2:3": "1024x1536",
+        "16:9": "2048x1152",
+        "9:16": "1152x2048",
+        "4:5": "1024x1280",
+        "5:4": "1280x1024"
+      },
       aspectRatios: ["1:1", "3:2", "2:3", "16:9", "9:16", "4:5", "5:4", "adaptive"],
       sizes: ["1K", "2K", "4K"],
       qualityOptions: ["auto", "low", "medium", "high"],
@@ -240,7 +249,7 @@ var IMAGE_MODELS = [
       supports: ["quality", "count"]
     },
     defaults: { size: "1K", aspectRatio: "adaptive", quality: "auto" },
-    usage: { ...IMAGE_USAGE },
+    usage: { ...IMAGE_USAGE, slug: "gpt" },
     pricingKind: "image_size",
     display: {
       provider: "openai",
@@ -260,6 +269,15 @@ var IMAGE_MODELS = [
     lifecycle: "active",
     exposure: "internal",
     constraints: {
+      exactSizes: {
+        "1:1": "1024x1024",
+        "3:2": "1536x1024",
+        "2:3": "1024x1536",
+        "16:9": "2048x1152",
+        "9:16": "1152x2048",
+        "4:5": "1024x1280",
+        "5:4": "1280x1024"
+      },
       aspectRatios: ["1:1", "3:2", "2:3", "16:9", "9:16", "4:5", "5:4", "adaptive"],
       sizes: ["1K", "2K", "4K"],
       qualityOptions: ["auto", "low", "medium", "high"],
@@ -286,6 +304,15 @@ var IMAGE_MODELS = [
     lifecycle: "active",
     exposure: "internal",
     constraints: {
+      exactSizes: {
+        "1:1": "1024x1024",
+        "3:2": "1536x1024",
+        "2:3": "1024x1536",
+        "16:9": "2048x1152",
+        "9:16": "1152x2048",
+        "4:5": "1024x1280",
+        "5:4": "1280x1024"
+      },
       aspectRatios: ["1:1", "3:2", "2:3", "16:9", "9:16", "4:5", "5:4", "adaptive"],
       sizes: ["1K", "2K", "4K"],
       qualityOptions: ["auto", "low", "medium", "high"],
@@ -331,7 +358,7 @@ var IMAGE_MODELS = [
       supports: ["count", "outputFormat"]
     },
     defaults: { size: "1K", aspectRatio: "1:1", outputFormat: "png", count: 1 },
-    usage: { ...IMAGE_USAGE },
+    usage: { ...IMAGE_USAGE, slug: "muse" },
     pricingKind: "image_size",
     display: {
       provider: "meta",
@@ -362,6 +389,7 @@ var IMAGE_MODELS = [
     pricingKind: "image_size",
     display: {
       provider: "bytedance",
+      logo: "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/3bba5d81-4e57-4e71-911b-cdfa05acca2c_compressed.png",
       description: "ByteDance Seedream 5.0 Pro text-to-image and multi-reference image-to-image (up to 10 inputs) with region editing; outputs 1K, 1.5K, or 2K."
     },
     capabilities: { ...IMAGE_GENERATE_EDIT }
@@ -567,6 +595,7 @@ var IMAGE_MODELS = [
 ];
 
 // src/catalog/other.ts
+var GOOGLE_LOGO = "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/192485d2-a76a-4657-9c06-930b2c70badd_compressed.png";
 var OTHER_MODELS = [
   {
     id: "lyria-3-clip-preview",
@@ -587,6 +616,7 @@ var OTHER_MODELS = [
     pricingKind: "flat",
     display: {
       provider: "google",
+      logo: GOOGLE_LOGO,
       description: "Google Lyria 3 Clip short music clips, loops, and previews; takes text and up to 10 image prompts and returns a fixed 30-second MP3 plus lyrics."
     },
     capabilities: {}
@@ -609,6 +639,7 @@ var OTHER_MODELS = [
     pricingKind: "flat",
     display: {
       provider: "google",
+      logo: GOOGLE_LOGO,
       description: "Google Lyria 3 Pro flagship music model for full-length songs with verses, choruses, and bridges; takes text and up to 10 image prompts and returns MP3 plus lyrics."
     },
     capabilities: {}
@@ -748,6 +779,14 @@ function text(id, label, providerModelId, extra) {
     ...extra
   };
 }
+var ANTHROPIC_THINKING = {
+  anthropic: { thinking: { type: "enabled", budgetTokens: 4e3 }, sendReasoning: true }
+};
+var ANTHROPIC_ADAPTIVE = {
+  anthropic: { thinking: { type: "adaptive", display: "summarized" }, sendReasoning: true }
+};
+var OPENAI_REASONING = { openai: { reasoningEffort: "medium", reasoningSummary: "auto" } };
+var DEEPSEEK_THINKING = { deepseek: { thinking: { type: "enabled" } } };
 var TEXT_MODELS = [
   text("gpt-4o", "GPT-4o", "gpt-4o", {
     exposure: "internal",
@@ -828,11 +867,13 @@ var TEXT_MODELS = [
   text("gpt-5.6-luna", "GPT-5.6 Luna", "gpt-5.6-luna", {
     exposure: "catalog",
     display: {
+      chat: { providerOptions: OPENAI_REASONING },
       provider: "openai",
       costTier: 1,
       description: "OpenAI GPT-5.6 Luna cost-optimized tier of the GPT-5.6 reasoning family; text and image input, 1.05M-token context."
     },
     capabilities: {
+      reasoning: true,
       vision: true,
       tools: true,
       structuredOutput: true,
@@ -842,11 +883,13 @@ var TEXT_MODELS = [
   text("gpt-5.6-terra", "GPT-5.6 Terra", "gpt-5.6-terra", {
     exposure: "catalog",
     display: {
+      chat: { providerOptions: OPENAI_REASONING },
       provider: "openai",
       costTier: 2,
       description: "OpenAI GPT-5.6 Terra balanced tier of the GPT-5.6 reasoning family; text and image input, 1.05M-token context."
     },
     capabilities: {
+      reasoning: true,
       vision: true,
       tools: true,
       structuredOutput: true,
@@ -856,11 +899,13 @@ var TEXT_MODELS = [
   text("gpt-5.6-sol", "GPT-5.6 Sol", "gpt-5.6-sol", {
     exposure: "catalog",
     display: {
+      chat: { providerOptions: OPENAI_REASONING },
       provider: "openai",
       costTier: 3,
       description: "OpenAI GPT-5.6 Sol flagship tier of the GPT-5.6 reasoning family; text and image input, 1.05M-token context."
     },
     capabilities: {
+      reasoning: true,
       vision: true,
       tools: true,
       structuredOutput: true,
@@ -888,11 +933,13 @@ var TEXT_MODELS = [
       "claude-sonnet-4-20250514"
     ],
     display: {
+      chat: { providerOptions: ANTHROPIC_THINKING },
       provider: "anthropic",
       costTier: 2,
       description: "Anthropic Claude Sonnet 4.6 (legacy) balanced model; text and image input, 1M-token context."
     },
     capabilities: {
+      reasoning: true,
       vision: true,
       tools: true,
       structuredOutput: true,
@@ -902,11 +949,13 @@ var TEXT_MODELS = [
   text("claude-sonnet-5", "Claude Sonnet 5", "claude-sonnet-5", {
     exposure: "catalog",
     display: {
+      chat: { providerOptions: ANTHROPIC_ADAPTIVE },
       provider: "anthropic",
       costTier: 3,
       description: "Anthropic Claude Sonnet 5 current balanced model; text and image input, 1M-token context."
     },
     capabilities: {
+      reasoning: true,
       vision: true,
       tools: true,
       structuredOutput: true,
@@ -917,11 +966,13 @@ var TEXT_MODELS = [
     exposure: "catalog",
     aliases: ["claude-opus-4.8", "claude-opus-4", "claude-opus-4.0", "claude-opus-4-20250514"],
     display: {
+      chat: { providerOptions: ANTHROPIC_ADAPTIVE },
       provider: "anthropic",
       costTier: 3,
       description: "Anthropic Claude Opus 4.8 (legacy) frontier model; text and image input, 1M-token context."
     },
     capabilities: {
+      reasoning: true,
       vision: true,
       tools: true,
       structuredOutput: true,
@@ -932,11 +983,18 @@ var TEXT_MODELS = [
     exposure: "catalog",
     aliases: ["claude-haiku-4.5", "claude-haiku-4-5-20251001", "claude-haiku-4"],
     display: {
+      chat: { providerOptions: ANTHROPIC_THINKING },
       provider: "anthropic",
       costTier: 1,
       description: "Anthropic Claude Haiku 4.5 fast model with near-frontier intelligence; text and image input, 200K-token context."
     },
-    capabilities: { vision: true, tools: true, structuredOutput: true, contextWindow: 2e5 }
+    capabilities: {
+      reasoning: true,
+      vision: true,
+      tools: true,
+      structuredOutput: true,
+      contextWindow: 2e5
+    }
   }),
   text("claude-opus-5", "Claude Opus 5", "claude-opus-5", {
     exposure: "internal",
@@ -1128,15 +1186,21 @@ var TEXT_MODELS = [
       contextWindow: 128e3
     }
   }),
-  text("deepseek-v4-flash", "DeepSeek V4 Flash", "deepseek-v4-flash", {
+  // DeepSeek retired V4 Flash upstream and routes the old id to V4.1 Flash at
+  // Flash rates (https://api-docs.deepseek.com/news/news260910/); the old id
+  // stays an alias so stored selections keep resolving.
+  text("deepseek-flash", "DeepSeek V4.1 Flash", "deepseek-flash", {
+    aliases: ["deepseek-v4-flash"],
     exposure: "catalog",
     display: {
+      chat: { providerOptions: DEEPSEEK_THINKING },
       provider: "deepseek",
       costTier: 1,
       recommended: true,
-      description: "DeepSeek V4 Flash thinking/non-thinking model with vision, tool calls, and JSON output; 1M-token context and up to 384K output tokens."
+      description: "DeepSeek V4.1 Flash thinking/non-thinking model with vision, tool calls, and JSON output; 1M-token context and up to 384K output tokens."
     },
     capabilities: {
+      reasoning: true,
       vision: true,
       tools: true,
       structuredOutput: true,
@@ -1146,11 +1210,13 @@ var TEXT_MODELS = [
   text("deepseek-v4-pro", "DeepSeek V4 Pro", "deepseek-v4-pro", {
     exposure: "catalog",
     display: {
+      chat: { providerOptions: DEEPSEEK_THINKING },
       provider: "deepseek",
       costTier: 2,
       description: "DeepSeek V4 Pro thinking/non-thinking text model with tool calls and JSON output (no vision); 1M-token context and up to 384K output tokens."
     },
     capabilities: {
+      reasoning: true,
       vision: false,
       tools: true,
       structuredOutput: true,
@@ -1176,6 +1242,7 @@ var TEXT_MODELS = [
   text("kimi-k2.5", "Kimi K2.5", "kimi-k2.5", {
     exposure: "catalog",
     display: {
+      chat: { temperature: 1 },
       provider: "moonshot",
       costTier: 2,
       description: "Moonshot Kimi K2.5; the provider retired the model on 2026-08-31 and calls now return a model-not-found error."
@@ -1185,11 +1252,13 @@ var TEXT_MODELS = [
   text("kimi-k3", "Kimi K3", "kimi-k3", {
     exposure: "catalog",
     display: {
+      chat: { temperature: 1 },
       provider: "moonshot",
       costTier: 2,
       description: "Moonshot Kimi K3 flagship model with native visual understanding and always-on thinking; text, image, and video input, 1M-token context."
     },
     capabilities: {
+      reasoning: true,
       vision: true,
       tools: true,
       structuredOutput: true,
@@ -1220,7 +1289,8 @@ var H3_RATIOS = ["21:9", "16:9", "4:3", "1:1", "3:4", "9:16"];
 var H3_REFERENCE_RATIOS = ["adaptive", ...H3_RATIOS];
 var NEGATIVE_PROMPT = "blur, distort, and low quality";
 var FRAMES_EXCLUDE_REFERENCES = { framesExcludeReferences: true };
-var GOOGLE_LOGO = "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/192485d2-a76a-4657-9c06-930b2c70badd_compressed.png";
+var BYTEDANCE_LOGO = "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/3bba5d81-4e57-4e71-911b-cdfa05acca2c_compressed.png";
+var GOOGLE_LOGO2 = "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/192485d2-a76a-4657-9c06-930b2c70badd_compressed.png";
 var VIDEO_MODELS = [
   {
     id: "veo-3.1",
@@ -1244,12 +1314,12 @@ var VIDEO_MODELS = [
       supports: ["lastFrame"]
     },
     defaults: { durationSeconds: 8, resolution: "720p", aspectRatio: "16:9" },
-    usage: { ...VIDEO_USAGE },
+    usage: { ...VIDEO_USAGE, slug: "veo" },
     pricingKind: "video_per_second",
     display: {
       provider: "google",
       description: "Google's highest-quality video generation model. Supports up to 4K resolution and 8-second clips with reference image conditioning.",
-      logo: GOOGLE_LOGO
+      logo: GOOGLE_LOGO2
     },
     capabilities: { ...FRAMES_EXCLUDE_REFERENCES }
   },
@@ -1336,12 +1406,12 @@ var VIDEO_MODELS = [
       supports: []
     },
     defaults: {},
-    usage: { ...VIDEO_USAGE },
+    usage: { ...VIDEO_USAGE, slug: "gemini_omni" },
     pricingKind: "video_per_second",
     display: {
       provider: "google",
       description: "Google's conversational video model with native audio, human support, and natural language editing. Aspect ratio only \u2014 resolution and duration are model-determined.",
-      logo: GOOGLE_LOGO
+      logo: GOOGLE_LOGO2
     },
     capabilities: { acceptsSourceVideo: true }
   },
@@ -1366,6 +1436,7 @@ var VIDEO_MODELS = [
     pricingKind: "video_per_second",
     display: {
       provider: "google",
+      logo: GOOGLE_LOGO2,
       description: "Google Gemini Omni Flash 1.1 conversational video generation and editing; text, image, and up to 10-second video input, 3-10 second output at up to 4K and 24 FPS."
     },
     capabilities: { acceptsSourceVideo: true }
@@ -1395,12 +1466,12 @@ var VIDEO_MODELS = [
       aspectRatio: "16:9",
       generateAudio: true
     },
-    usage: { ...VIDEO_USAGE },
+    usage: { ...VIDEO_USAGE, slug: "seedance2", providerLabel: "byteplus" },
     pricingKind: "video_per_second",
     display: {
       provider: "bytedance",
       description: "High-quality AI video generation with reference image conditioning. Supports 4-15 second clips with human preprocessing.",
-      logo: "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/3bba5d81-4e57-4e71-911b-cdfa05acca2c_compressed.png"
+      logo: BYTEDANCE_LOGO
     },
     capabilities: { acceptsSourceVideo: true, ...FRAMES_EXCLUDE_REFERENCES }
   },
@@ -1431,6 +1502,7 @@ var VIDEO_MODELS = [
     pricingKind: "video_per_second",
     display: {
       provider: "bytedance",
+      logo: BYTEDANCE_LOGO,
       description: "Speed- and cost-optimized ByteDance Seedance 2.0 Fast; text plus image, video, and audio references to generate or edit 4-15 second clips."
     },
     capabilities: { acceptsSourceVideo: true, ...FRAMES_EXCLUDE_REFERENCES }
@@ -1463,6 +1535,7 @@ var VIDEO_MODELS = [
     pricingKind: "video_per_second",
     display: {
       provider: "bytedance",
+      logo: BYTEDANCE_LOGO,
       description: "ByteDance Seedance 2.5 for long-form storytelling and omni-reference generation/editing; up to 50 reference assets and 4-30 second clips at up to 1080p."
     },
     capabilities: { acceptsSourceVideo: true, ...FRAMES_EXCLUDE_REFERENCES }
@@ -1487,7 +1560,7 @@ var VIDEO_MODELS = [
     },
     // No resolution default: xAI picks its own unless the caller asks.
     defaults: { durationSeconds: 8, aspectRatio: "16:9" },
-    usage: { unit: "reported_cost_usd", modality: "VIDEO" },
+    usage: { unit: "reported_cost_usd", modality: "VIDEO", slug: "grok" },
     pricingKind: "video_per_second",
     display: {
       provider: "xai",
@@ -1968,6 +2041,40 @@ function resolveVideoOptions(model, request, onUnsupported = "error") {
         model: model.id
       });
   }
+  if (merged.sourceVideo !== void 0 && !model.capabilities?.acceptsSourceVideo) {
+    if (onUnsupported === "error") {
+      throw new ModelKitError({
+        code: "unsupported_option",
+        message: `${model.id} does not take a source video`,
+        model: model.id
+      });
+    }
+    warnings.push({
+      code: "option_ignored",
+      field: "sourceVideo",
+      message: `${model.id} does not take a source video; dropped`
+    });
+    delete merged.sourceVideo;
+  }
+  const hasFrames = merged.firstFrame !== void 0 || merged.lastFrame !== void 0;
+  const hasRefs = compact(merged.referenceImages).length + compact(merged.referenceVideos).length + compact(merged.referenceAudios).length > 0;
+  if (hasFrames && hasRefs && model.capabilities?.framesExcludeReferences) {
+    if (onUnsupported === "error") {
+      throw new ModelKitError({
+        code: "unsupported_option",
+        message: `${model.id} takes frames or references, not both`,
+        model: model.id
+      });
+    }
+    warnings.push({
+      code: "option_ignored",
+      field: "referenceImages",
+      message: `${model.id} takes frames or references, not both; references dropped`
+    });
+    delete merged.referenceImages;
+    delete merged.referenceVideos;
+    delete merged.referenceAudios;
+  }
   for (const input of c.ratioIgnoredWith ?? []) {
     if (merged[input] !== void 0 && merged.aspectRatio !== void 0) {
       if (request.aspectRatio !== void 0) {