@shop-os/modelkit 0.1.8 → 0.1.9

package/dist/index.d.cts

@@ -175,6 +175,11 @@ interface ModelConstraints {
     durationPresets?: number[];
     /** Curated quality choices for rows whose `supports` includes "quality". */
     qualityOptions?: string[];
+    /**
+     * Exact pixel size per aspect ratio for rows an app must size verbatim
+     * (gpt-image-2 keeps its historic WxH contract); callers pass it as `exactSize`.
+     */
+    exactSizes?: Record<string, string>;
     references?: ReferenceLimits;
     requires?: RequiredInput[];
     supports?: SupportedOption[];
@@ -216,6 +221,11 @@ interface CatalogDisplay {
     costTier?: 1 | 2 | 3;
     /** Chat picker highlight flag. */
     recommended?: boolean;
+    /**
+     * Picker position: lower comes first; rows without one keep catalog order
+     * after the ordered ones. Set it to surface a new model at the top.
+     */
+    sortOrder?: number;
 }
 /** Capability flags a UI keys off. Absent means false / unknown. */
 interface CatalogCapabilities {
@@ -227,7 +237,8 @@ interface CatalogCapabilities {
     acceptsSourceVideo?: boolean;
     /** Video rows: first/last frames and reference images are mutually exclusive inputs. */
     framesExcludeReferences?: boolean;
-    /** Text rows. */
+    /** Text rows. `reasoning`: the model streams reasoning the UI should surface. */
+    reasoning?: boolean;
     vision?: boolean;
     tools?: boolean;
     structuredOutput?: boolean;
@@ -250,11 +261,21 @@ interface CatalogModel {
      */
     exposure: "catalog" | "internal";
     constraints: ModelConstraints;
-    defaults: Partial<Omit<ImageRequest, "model" | "prompt"> & Omit<VideoRequest, "model" | "prompt"> & Omit<AudioRequest, "model" | "prompt">>;
+    defaults: Partial<Omit<ImageRequest, "model" | "prompt"> & Omit<VideoRequest, "model" | "prompt"> & Omit<AudioRequest, "model" | "prompt"> & {
+        temperature?: number;
+    }>;
+    /** Provider options merged under the caller's; text rows carry their reasoning / thinking tuning here. */
     providerOptions?: ProviderOptions;
+    /**
+     * `slug` and `providerLabel` keep an app's historic usage-event names for a
+     * row (e.g. cowork emits `image_gen_<slug>` from provider `<providerLabel>`);
+     * absent means the app derives them from the id and display.provider.
+     */
     usage: {
         unit: UsageUnit;
         modality: UsageModality;
+        slug?: string;
+        providerLabel?: string;
     };
     pricingKind: PricingKind;
     /** Required on every `active` row so a picker can render it without a hand-written catalog. */

package/dist/index.d.ts

@@ -175,6 +175,11 @@ interface ModelConstraints {
     durationPresets?: number[];
     /** Curated quality choices for rows whose `supports` includes "quality". */
     qualityOptions?: string[];
+    /**
+     * Exact pixel size per aspect ratio for rows an app must size verbatim
+     * (gpt-image-2 keeps its historic WxH contract); callers pass it as `exactSize`.
+     */
+    exactSizes?: Record<string, string>;
     references?: ReferenceLimits;
     requires?: RequiredInput[];
     supports?: SupportedOption[];
@@ -216,6 +221,11 @@ interface CatalogDisplay {
     costTier?: 1 | 2 | 3;
     /** Chat picker highlight flag. */
     recommended?: boolean;
+    /**
+     * Picker position: lower comes first; rows without one keep catalog order
+     * after the ordered ones. Set it to surface a new model at the top.
+     */
+    sortOrder?: number;
 }
 /** Capability flags a UI keys off. Absent means false / unknown. */
 interface CatalogCapabilities {
@@ -227,7 +237,8 @@ interface CatalogCapabilities {
     acceptsSourceVideo?: boolean;
     /** Video rows: first/last frames and reference images are mutually exclusive inputs. */
     framesExcludeReferences?: boolean;
-    /** Text rows. */
+    /** Text rows. `reasoning`: the model streams reasoning the UI should surface. */
+    reasoning?: boolean;
     vision?: boolean;
     tools?: boolean;
     structuredOutput?: boolean;
@@ -250,11 +261,21 @@ interface CatalogModel {
      */
     exposure: "catalog" | "internal";
     constraints: ModelConstraints;
-    defaults: Partial<Omit<ImageRequest, "model" | "prompt"> & Omit<VideoRequest, "model" | "prompt"> & Omit<AudioRequest, "model" | "prompt">>;
+    defaults: Partial<Omit<ImageRequest, "model" | "prompt"> & Omit<VideoRequest, "model" | "prompt"> & Omit<AudioRequest, "model" | "prompt"> & {
+        temperature?: number;
+    }>;
+    /** Provider options merged under the caller's; text rows carry their reasoning / thinking tuning here. */
     providerOptions?: ProviderOptions;
+    /**
+     * `slug` and `providerLabel` keep an app's historic usage-event names for a
+     * row (e.g. cowork emits `image_gen_<slug>` from provider `<providerLabel>`);
+     * absent means the app derives them from the id and display.provider.
+     */
     usage: {
         unit: UsageUnit;
         modality: UsageModality;
+        slug?: string;
+        providerLabel?: string;
     };
     pricingKind: PricingKind;
     /** Required on every `active` row so a picker can render it without a hand-written catalog. */

package/dist/index.js

@@ -130,7 +130,7 @@ 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 }
@@ -198,6 +198,15 @@ 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"],
@@ -205,7 +214,7 @@ 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",
@@ -225,6 +234,15 @@ 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"],
@@ -251,6 +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"],
@@ -296,7 +323,7 @@ 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",
@@ -713,6 +740,14 @@ 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",
@@ -791,6 +826,7 @@ var TEXT_MODELS = [
     }
   }),
   text("gpt-5.6-luna", "GPT-5.6 Luna", "gpt-5.6-luna", {
+    providerOptions: OPENAI_REASONING,
     exposure: "catalog",
     display: {
       provider: "openai",
@@ -798,6 +834,7 @@ var TEXT_MODELS = [
       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,
@@ -805,6 +842,7 @@ var TEXT_MODELS = [
     }
   }),
   text("gpt-5.6-terra", "GPT-5.6 Terra", "gpt-5.6-terra", {
+    providerOptions: OPENAI_REASONING,
     exposure: "catalog",
     display: {
       provider: "openai",
@@ -812,6 +850,7 @@ var TEXT_MODELS = [
       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,
@@ -819,6 +858,7 @@ var TEXT_MODELS = [
     }
   }),
   text("gpt-5.6-sol", "GPT-5.6 Sol", "gpt-5.6-sol", {
+    providerOptions: OPENAI_REASONING,
     exposure: "catalog",
     display: {
       provider: "openai",
@@ -826,6 +866,7 @@ var TEXT_MODELS = [
       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,
@@ -845,6 +886,7 @@ var TEXT_MODELS = [
   }),
   // Anthropic retired the dated 4.0 ids; callers that still send them get 4.6.
   text("claude-sonnet-4-6", "Claude Sonnet 4.6", "claude-sonnet-4-6", {
+    providerOptions: ANTHROPIC_THINKING,
     exposure: "catalog",
     aliases: [
       "claude-sonnet-4.6",
@@ -858,6 +900,7 @@ var TEXT_MODELS = [
       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,
@@ -865,6 +908,7 @@ var TEXT_MODELS = [
     }
   }),
   text("claude-sonnet-5", "Claude Sonnet 5", "claude-sonnet-5", {
+    providerOptions: ANTHROPIC_ADAPTIVE,
     exposure: "catalog",
     display: {
       provider: "anthropic",
@@ -872,6 +916,7 @@ var TEXT_MODELS = [
       description: "Anthropic Claude Sonnet 5 current balanced model; text and image input, 1M-token context."
     },
     capabilities: {
+      reasoning: true,
       vision: true,
       tools: true,
       structuredOutput: true,
@@ -879,6 +924,7 @@ var TEXT_MODELS = [
     }
   }),
   text("claude-opus-4-8", "Claude Opus 4.8", "claude-opus-4-8", {
+    providerOptions: ANTHROPIC_ADAPTIVE,
     exposure: "catalog",
     aliases: ["claude-opus-4.8", "claude-opus-4", "claude-opus-4.0", "claude-opus-4-20250514"],
     display: {
@@ -887,6 +933,7 @@ var TEXT_MODELS = [
       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,
@@ -894,6 +941,7 @@ var TEXT_MODELS = [
     }
   }),
   text("claude-haiku-4-5", "Claude Haiku 4.5", "claude-haiku-4-5", {
+    providerOptions: ANTHROPIC_THINKING,
     exposure: "catalog",
     aliases: ["claude-haiku-4.5", "claude-haiku-4-5-20251001", "claude-haiku-4"],
     display: {
@@ -901,7 +949,13 @@ var TEXT_MODELS = [
       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",
@@ -1093,15 +1147,21 @@ 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", {
+    providerOptions: DEEPSEEK_THINKING,
+    aliases: ["deepseek-v4-flash"],
     exposure: "catalog",
     display: {
       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,
@@ -1109,6 +1169,7 @@ var TEXT_MODELS = [
     }
   }),
   text("deepseek-v4-pro", "DeepSeek V4 Pro", "deepseek-v4-pro", {
+    providerOptions: DEEPSEEK_THINKING,
     exposure: "catalog",
     display: {
       provider: "deepseek",
@@ -1116,6 +1177,7 @@ var TEXT_MODELS = [
       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,
@@ -1139,6 +1201,7 @@ var TEXT_MODELS = [
   }),
   // Moonshot retired kimi-k2.5 on 2026-08-31 and no longer publishes its specs.
   text("kimi-k2.5", "Kimi K2.5", "kimi-k2.5", {
+    defaults: { temperature: 1 },
     exposure: "catalog",
     display: {
       provider: "moonshot",
@@ -1148,6 +1211,7 @@ var TEXT_MODELS = [
     capabilities: {}
   }),
   text("kimi-k3", "Kimi K3", "kimi-k3", {
+    defaults: { temperature: 1 },
     exposure: "catalog",
     display: {
       provider: "moonshot",
@@ -1155,6 +1219,7 @@ var TEXT_MODELS = [
       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,
@@ -1209,7 +1274,7 @@ 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",
@@ -1301,7 +1366,7 @@ var VIDEO_MODELS = [
       supports: []
     },
     defaults: {},
-    usage: { ...VIDEO_USAGE },
+    usage: { ...VIDEO_USAGE, slug: "gemini_omni" },
     pricingKind: "video_per_second",
     display: {
       provider: "google",
@@ -1360,7 +1425,7 @@ 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",
@@ -1482,7 +1547,7 @@ var VIDEO_MODELS = [
       cfgScale: 0.5,
       negativePrompt: NEGATIVE_PROMPT
     },
-    usage: { ...VIDEO_USAGE },
+    usage: { ...VIDEO_USAGE, slug: "grok" },
     pricingKind: "video_per_second",
     display: {
       provider: "kuaishou",
@@ -1933,6 +1998,40 @@ 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) {