@oh-my-pi/pi-catalog 16.3.14 → 16.3.15

package/CHANGELOG.md

@@ -2,6 +2,24 @@
 
 ## [Unreleased]
 
+## [16.3.15] - 2026-07-09
+
+### Added
+
+- Added support for Grok 4.5 model
+- Added `gpt-5.6` base models and `gpt-5.6-{luna,sol,terra}-pro` variants
+- Added `meta/muse-spark-1.1` model support
+- Added support for thinking modes on `poolside/laguna` models
+- Added generated GPT-5.6 Pro aliases (`gpt-5.6-{luna,sol,terra}-pro`) on the `openai` and `openai-codex` providers: each alias sends the base model id on the wire (`requestModelId`) with the new `reasoningMode: "pro"` marker, and re-derives from the current base rows on every catalog regeneration.
+
+### Changed
+
+- Updated cache read costs for Grok models
+- Reduced max token limit for Grok 4.3 model
+- Enabled prompt cache affinity for Grok models via the x-grok-conv-id header in OpenAI compatible endpoints
+- Enabled prompt cache affinity for Grok models via the x-grok-conv-id header
+- Marked direct xAI Grok Chat Completions models for `x-grok-conv-id` prompt-cache affinity.
+
 ## [16.3.14] - 2026-07-09
 
 ### Added

package/dist/types/provider-models/openai-compat.d.ts

@@ -54,6 +54,18 @@ export interface OpenAIModelManagerConfig {
     fetch?: FetchImpl;
 }
 export declare function openaiModelManagerOptions(config?: OpenAIModelManagerConfig): ModelManagerOptions<"openai-responses">;
+/**
+ * Re-derive the generated pro-reasoning aliases (`gpt-5.6-*-pro`) for the
+ * first-party `openai`/`openai-codex` gpt-5.6 rows. Each alias inherits the
+ * base row's metadata, requests the base wire id via `requestModelId`, and
+ * sets `reasoningMode: "pro"` so Responses-family request builders emit
+ * `reasoning: { mode: "pro" }`. Called by the models.json generator after all
+ * sources merge: stale copies of the owned aliases (previous snapshot) are
+ * dropped and re-projected from the current base rows so alias metadata always
+ * tracks the base, while a real upstream model that occupies an alias id wins
+ * and suppresses the projection.
+ */
+export declare function projectOpenAIProReasoningAliases(models: readonly ModelSpec<Api>[]): ModelSpec<Api>[];
 export interface GroqModelManagerConfig {
     apiKey?: string;
     baseUrl?: string;

package/dist/types/types.d.ts

@@ -563,6 +563,13 @@ export interface Model<TApi extends Api = Api> {
      * everything local (selection, caching, usage attribution) keys on `id`.
      */
     requestModelId?: string;
+    /**
+     * `reasoning.mode` to send on OpenAI Responses-family requests. Set on
+     * generated pro aliases (`gpt-5.6-*-pro` on `openai`/`openai-codex`) that
+     * pair a base wire id (`requestModelId`) with OpenAI's pro reasoning
+     * serving path. Absent everywhere else; providers omit the wire field.
+     */
+    reasoningMode?: "pro";
     name: string;
     api: TApi;
     provider: Provider;

package/package.json

@@ -1,7 +1,7 @@
 {
 	"type": "module",
 	"name": "@oh-my-pi/pi-catalog",
-	"version": "16.3.14",
+	"version": "16.3.15",
 	"description": "Model catalog for omp: bundled model database, provider discovery descriptors, model identity, classification, and equivalence",
 	"homepage": "https://omp.sh",
 	"author": "Can Boluk",
@@ -34,12 +34,12 @@
 	},
 	"dependencies": {
 		"@bufbuild/protobuf": "^2.12.0",
-		"@oh-my-pi/pi-utils": "16.3.14",
+		"@oh-my-pi/pi-utils": "16.3.15",
 		"arktype": "^2.2.0",
 		"zod": "^4"
 	},
 	"devDependencies": {
-		"@oh-my-pi/pi-ai": "16.3.14",
+		"@oh-my-pi/pi-ai": "16.3.15",
 		"@types/bun": "^1.3.14"
 	},
 	"engines": {

package/src/compat/openai.ts

@@ -541,7 +541,7 @@ export function buildOpenAICompat(spec: ModelSpec<"openai-completions">): Resolv
 			MINIMAX_PROVIDER_OR_ID_PATTERN.test(provider) || MINIMAX_PROVIDER_OR_ID_PATTERN.test(spec.id),
 		emptyLengthFinishIsContextError: provider === "ollama",
 		usesOpenAIToolCallIdLimit: provider === "openai",
-		promptCacheSessionHeader: undefined,
+		promptCacheSessionHeader: isGrok ? "x-grok-conv-id" : undefined,
 		dropThinkingWhenReasoningEffort: provider === "fireworks",
 	};
 

package/src/identity/family.ts

@@ -78,7 +78,7 @@ export const isMimoModelIdOrName = memo((value: string): boolean => {
 	return value.toLowerCase().includes("mimo");
 });
 
-const GROK_EFFORT_CAPABLE_PREFIXES = ["grok-3-mini", "grok-4.20-multi-agent", "grok-4.3"] as const;
+const GROK_EFFORT_CAPABLE_PREFIXES = ["grok-3-mini", "grok-4.20-multi-agent", "grok-4.3", "grok-4.5"] as const;
 
 /**
  * Grok SKUs that expose the wire `reasoning.effort` dial. Other Grok reasoners

package/src/models.json

@@ -32288,7 +32288,7 @@
 			"api": "openai-completions",
 			"provider": "kilo",
 			"baseUrl": "https://api.kilo.ai/api/gateway",
-			"reasoning": false,
+			"reasoning": true,
 			"input": [
 				"text"
 			],
@@ -32299,7 +32299,17 @@
 				"cacheWrite": 0
 			},
 			"contextWindow": 262144,
-			"maxTokens": 32768
+			"maxTokens": 32768,
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				]
+			}
 		},
 		"poolside/laguna-m.1:free": {
 			"id": "poolside/laguna-m.1:free",
@@ -32374,7 +32384,7 @@
 			"api": "openai-completions",
 			"provider": "kilo",
 			"baseUrl": "https://api.kilo.ai/api/gateway",
-			"reasoning": false,
+			"reasoning": true,
 			"input": [
 				"text"
 			],
@@ -32385,7 +32395,17 @@
 				"cacheWrite": 0
 			},
 			"contextWindow": 262144,
-			"maxTokens": 32768
+			"maxTokens": 32768,
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				]
+			}
 		},
 		"poolside/laguna-xs.2:free": {
 			"id": "poolside/laguna-xs.2:free",
@@ -45679,6 +45699,25 @@
 			"contextWindow": 328000,
 			"maxTokens": 65536
 		},
+		"meta/muse-spark-1.1": {
+			"id": "meta/muse-spark-1.1",
+			"name": "meta/muse-spark-1.1",
+			"api": "openai-completions",
+			"provider": "nanogpt",
+			"baseUrl": "https://nano-gpt.com/api/v1",
+			"reasoning": false,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1048576,
+			"maxTokens": 1048576
+		},
 		"microsoft/MAI-DS-R1-FP8": {
 			"id": "microsoft/MAI-DS-R1-FP8",
 			"name": "microsoft/MAI-DS-R1-FP8",
@@ -48497,6 +48536,174 @@
 			},
 			"contextPromotionTarget": "nanogpt/openai/gpt-5.4"
 		},
+		"openai/gpt-5.6-luna": {
+			"id": "openai/gpt-5.6-luna",
+			"name": "GPT-5.6 Luna",
+			"api": "openai-completions",
+			"provider": "nanogpt",
+			"baseUrl": "https://nano-gpt.com/api/v1",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 1,
+				"output": 6,
+				"cacheRead": 0.09999999999999999,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1050000,
+			"maxTokens": 128000,
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				],
+				"effortMap": {
+					"minimal": "low",
+					"low": "medium",
+					"medium": "high",
+					"high": "xhigh",
+					"xhigh": "max"
+				}
+			}
+		},
+		"openai/gpt-5.6-luna-pro": {
+			"id": "openai/gpt-5.6-luna-pro",
+			"name": "GPT-5.6 Luna Pro",
+			"api": "openai-completions",
+			"provider": "nanogpt",
+			"baseUrl": "https://nano-gpt.com/api/v1",
+			"reasoning": false,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1050000,
+			"maxTokens": 128000
+		},
+		"openai/gpt-5.6-sol": {
+			"id": "openai/gpt-5.6-sol",
+			"name": "GPT-5.6 Sol",
+			"api": "openai-completions",
+			"provider": "nanogpt",
+			"baseUrl": "https://nano-gpt.com/api/v1",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 5,
+				"output": 30,
+				"cacheRead": 0.5,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1050000,
+			"maxTokens": 128000,
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				],
+				"effortMap": {
+					"minimal": "low",
+					"low": "medium",
+					"medium": "high",
+					"high": "xhigh",
+					"xhigh": "max"
+				}
+			}
+		},
+		"openai/gpt-5.6-sol-pro": {
+			"id": "openai/gpt-5.6-sol-pro",
+			"name": "GPT-5.6 Sol Pro",
+			"api": "openai-completions",
+			"provider": "nanogpt",
+			"baseUrl": "https://nano-gpt.com/api/v1",
+			"reasoning": false,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1050000,
+			"maxTokens": 128000
+		},
+		"openai/gpt-5.6-terra": {
+			"id": "openai/gpt-5.6-terra",
+			"name": "GPT-5.6 Terra",
+			"api": "openai-completions",
+			"provider": "nanogpt",
+			"baseUrl": "https://nano-gpt.com/api/v1",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 2.5,
+				"output": 15,
+				"cacheRead": 0.25,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1050000,
+			"maxTokens": 128000,
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				],
+				"effortMap": {
+					"minimal": "low",
+					"low": "medium",
+					"medium": "high",
+					"high": "xhigh",
+					"xhigh": "max"
+				}
+			}
+		},
+		"openai/gpt-5.6-terra-pro": {
+			"id": "openai/gpt-5.6-terra-pro",
+			"name": "GPT-5.6 Terra Pro",
+			"api": "openai-completions",
+			"provider": "nanogpt",
+			"baseUrl": "https://nano-gpt.com/api/v1",
+			"reasoning": false,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1050000,
+			"maxTokens": 128000
+		},
 		"openai/gpt-chat-latest": {
 			"id": "openai/gpt-chat-latest",
 			"name": "GPT Chat Latest",
@@ -49105,41 +49312,61 @@
 		},
 		"poolside/laguna-m.1": {
 			"id": "poolside/laguna-m.1",
-			"name": "poolside/laguna-m.1",
+			"name": "Laguna M.1",
 			"api": "openai-completions",
 			"provider": "nanogpt",
 			"baseUrl": "https://nano-gpt.com/api/v1",
-			"reasoning": false,
+			"reasoning": true,
 			"input": [
 				"text"
 			],
 			"cost": {
-				"input": 0,
-				"output": 0,
+				"input": 0.2,
+				"output": 0.4,
 				"cacheRead": 0,
 				"cacheWrite": 0
 			},
 			"contextWindow": 262144,
-			"maxTokens": 32768
+			"maxTokens": 32768,
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				]
+			}
 		},
 		"poolside/laguna-xs.2": {
 			"id": "poolside/laguna-xs.2",
-			"name": "poolside/laguna-xs.2",
+			"name": "Laguna XS.2",
 			"api": "openai-completions",
 			"provider": "nanogpt",
 			"baseUrl": "https://nano-gpt.com/api/v1",
-			"reasoning": false,
+			"reasoning": true,
 			"input": [
 				"text"
 			],
 			"cost": {
-				"input": 0,
-				"output": 0,
+				"input": 0.2,
+				"output": 0.4,
 				"cacheRead": 0,
 				"cacheWrite": 0
 			},
 			"contextWindow": 262144,
-			"maxTokens": 32768
+			"maxTokens": 32768,
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				]
+			}
 		},
 		"qvq-max": {
 			"id": "qvq-max",
@@ -60096,9 +60323,9 @@
 			},
 			"contextPromotionTarget": "openai/gpt-5.4"
 		},
-		"gpt-5.6-luna": {
-			"id": "gpt-5.6-luna",
-			"name": "GPT-5.6 Luna",
+		"gpt-5.6": {
+			"id": "gpt-5.6",
+			"name": "GPT-5.6",
 			"api": "openai-responses",
 			"provider": "openai",
 			"baseUrl": "https://api.openai.com/v1",
@@ -60108,10 +60335,10 @@
 				"image"
 			],
 			"cost": {
-				"input": 1,
-				"output": 6,
-				"cacheRead": 0.1,
-				"cacheWrite": 1.25
+				"input": 5,
+				"output": 30,
+				"cacheRead": 0.5,
+				"cacheWrite": 6.25
 			},
 			"contextWindow": 1050000,
 			"maxTokens": 128000,
@@ -60134,6 +60361,84 @@
 				}
 			}
 		},
+		"gpt-5.6-luna": {
+			"id": "gpt-5.6-luna",
+			"name": "GPT-5.6 Luna",
+			"api": "openai-responses",
+			"provider": "openai",
+			"baseUrl": "https://api.openai.com/v1",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 1,
+				"output": 6,
+				"cacheRead": 0.1,
+				"cacheWrite": 1.25
+			},
+			"contextWindow": 1050000,
+			"maxTokens": 128000,
+			"applyPatchToolType": "freeform",
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				],
+				"effortMap": {
+					"minimal": "low",
+					"low": "medium",
+					"medium": "high",
+					"high": "xhigh",
+					"xhigh": "max"
+				}
+			}
+		},
+		"gpt-5.6-luna-pro": {
+			"id": "gpt-5.6-luna-pro",
+			"name": "GPT-5.6 Luna Pro",
+			"api": "openai-responses",
+			"provider": "openai",
+			"baseUrl": "https://api.openai.com/v1",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 1,
+				"output": 6,
+				"cacheRead": 0.1,
+				"cacheWrite": 1.25
+			},
+			"contextWindow": 1050000,
+			"maxTokens": 128000,
+			"requestModelId": "gpt-5.6-luna",
+			"reasoningMode": "pro",
+			"applyPatchToolType": "freeform",
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				],
+				"effortMap": {
+					"minimal": "low",
+					"low": "medium",
+					"medium": "high",
+					"high": "xhigh",
+					"xhigh": "max"
+				}
+			}
+		},
 		"gpt-5.6-sol": {
 			"id": "gpt-5.6-sol",
 			"name": "GPT-5.6 Sol",
@@ -60172,6 +60477,46 @@
 				}
 			}
 		},
+		"gpt-5.6-sol-pro": {
+			"id": "gpt-5.6-sol-pro",
+			"name": "GPT-5.6 Sol Pro",
+			"api": "openai-responses",
+			"provider": "openai",
+			"baseUrl": "https://api.openai.com/v1",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 5,
+				"output": 30,
+				"cacheRead": 0.5,
+				"cacheWrite": 6.25
+			},
+			"contextWindow": 1050000,
+			"maxTokens": 128000,
+			"requestModelId": "gpt-5.6-sol",
+			"reasoningMode": "pro",
+			"applyPatchToolType": "freeform",
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				],
+				"effortMap": {
+					"minimal": "low",
+					"low": "medium",
+					"medium": "high",
+					"high": "xhigh",
+					"xhigh": "max"
+				}
+			}
+		},
 		"gpt-5.6-terra": {
 			"id": "gpt-5.6-terra",
 			"name": "GPT-5.6 Terra",
@@ -60210,6 +60555,46 @@
 				}
 			}
 		},
+		"gpt-5.6-terra-pro": {
+			"id": "gpt-5.6-terra-pro",
+			"name": "GPT-5.6 Terra Pro",
+			"api": "openai-responses",
+			"provider": "openai",
+			"baseUrl": "https://api.openai.com/v1",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 2.5,
+				"output": 15,
+				"cacheRead": 0.25,
+				"cacheWrite": 3.125
+			},
+			"contextWindow": 1050000,
+			"maxTokens": 128000,
+			"requestModelId": "gpt-5.6-terra",
+			"reasoningMode": "pro",
+			"applyPatchToolType": "freeform",
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				],
+				"effortMap": {
+					"minimal": "low",
+					"low": "medium",
+					"medium": "high",
+					"high": "xhigh",
+					"xhigh": "max"
+				}
+			}
+		},
 		"o1": {
 			"id": "o1",
 			"name": "o1",
@@ -61034,6 +61419,53 @@
 				}
 			}
 		},
+		"gpt-5.6-luna-pro": {
+			"id": "gpt-5.6-luna-pro",
+			"name": "GPT-5.6 Luna Pro",
+			"api": "openai-codex-responses",
+			"provider": "openai-codex",
+			"baseUrl": "https://chatgpt.com/backend-api",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 1,
+				"output": 6,
+				"cacheRead": 0.1,
+				"cacheWrite": 1.25
+			},
+			"remoteCompaction": {
+				"enabled": true,
+				"api": "openai-codex-responses",
+				"v2StreamingEnabled": true
+			},
+			"contextWindow": 372000,
+			"maxTokens": 128000,
+			"preferWebsockets": true,
+			"priority": 3,
+			"requestModelId": "gpt-5.6-luna",
+			"reasoningMode": "pro",
+			"applyPatchToolType": "freeform",
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				],
+				"effortMap": {
+					"minimal": "low",
+					"low": "medium",
+					"medium": "high",
+					"high": "xhigh",
+					"xhigh": "max"
+				}
+			}
+		},
 		"gpt-5.6-sol": {
 			"id": "gpt-5.6-sol",
 			"name": "GPT-5.6 Sol",
@@ -61079,6 +61511,53 @@
 				}
 			}
 		},
+		"gpt-5.6-sol-pro": {
+			"id": "gpt-5.6-sol-pro",
+			"name": "GPT-5.6 Sol Pro",
+			"api": "openai-codex-responses",
+			"provider": "openai-codex",
+			"baseUrl": "https://chatgpt.com/backend-api",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 5,
+				"output": 30,
+				"cacheRead": 0.5,
+				"cacheWrite": 6.25
+			},
+			"remoteCompaction": {
+				"enabled": true,
+				"api": "openai-codex-responses",
+				"v2StreamingEnabled": true
+			},
+			"contextWindow": 372000,
+			"maxTokens": 128000,
+			"preferWebsockets": true,
+			"priority": 1,
+			"requestModelId": "gpt-5.6-sol",
+			"reasoningMode": "pro",
+			"applyPatchToolType": "freeform",
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				],
+				"effortMap": {
+					"minimal": "low",
+					"low": "medium",
+					"medium": "high",
+					"high": "xhigh",
+					"xhigh": "max"
+				}
+			}
+		},
 		"gpt-5.6-terra": {
 			"id": "gpt-5.6-terra",
 			"name": "GPT-5.6 Terra",
@@ -61123,6 +61602,53 @@
 					"xhigh": "max"
 				}
 			}
+		},
+		"gpt-5.6-terra-pro": {
+			"id": "gpt-5.6-terra-pro",
+			"name": "GPT-5.6 Terra Pro",
+			"api": "openai-codex-responses",
+			"provider": "openai-codex",
+			"baseUrl": "https://chatgpt.com/backend-api",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 2.5,
+				"output": 15,
+				"cacheRead": 0.25,
+				"cacheWrite": 3.125
+			},
+			"remoteCompaction": {
+				"enabled": true,
+				"api": "openai-codex-responses",
+				"v2StreamingEnabled": true
+			},
+			"contextWindow": 372000,
+			"maxTokens": 128000,
+			"preferWebsockets": true,
+			"priority": 2,
+			"requestModelId": "gpt-5.6-terra",
+			"reasoningMode": "pro",
+			"applyPatchToolType": "freeform",
+			"thinking": {
+				"mode": "effort",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				],
+				"effortMap": {
+					"minimal": "low",
+					"low": "medium",
+					"medium": "high",
+					"high": "xhigh",
+					"xhigh": "max"
+				}
+			}
 		}
 	},
 	"opencode": {
@@ -64125,7 +64651,7 @@
 			"cost": {
 				"input": 0.65,
 				"output": 3.41,
-				"cacheRead": 0.14,
+				"cacheRead": 0.15,
 				"cacheWrite": 0
 			},
 			"contextWindow": 262144,
@@ -64155,7 +64681,7 @@
 				"input": 5,
 				"output": 30,
 				"cacheRead": 0.5,
-				"cacheWrite": 0
+				"cacheWrite": 6.25
 			},
 			"contextWindow": 1050000,
 			"maxTokens": 128000,
@@ -67262,7 +67788,7 @@
 				"cacheWrite": 0
 			},
 			"contextWindow": 1048576,
-			"maxTokens": 512000,
+			"maxTokens": 131072,
 			"thinking": {
 				"mode": "effort",
 				"efforts": [
@@ -67926,7 +68452,7 @@
 			"cost": {
 				"input": 0.65,
 				"output": 3.41,
-				"cacheRead": 0.14,
+				"cacheRead": 0.15,
 				"cacheWrite": 0
 			},
 			"contextWindow": 262144,
@@ -67983,8 +68509,8 @@
 			],
 			"cost": {
 				"input": 0.72,
-				"output": 3.5,
-				"cacheRead": 0.15,
+				"output": 3.49,
+				"cacheRead": 0.159,
 				"cacheWrite": 0
 			},
 			"contextWindow": 262144,
@@ -69547,7 +70073,7 @@
 				"input": 1,
 				"output": 6,
 				"cacheRead": 0.09999999999999999,
-				"cacheWrite": 0
+				"cacheWrite": 1.25
 			},
 			"contextWindow": 1050000,
 			"maxTokens": 128000,
@@ -69584,7 +70110,7 @@
 				"input": 1,
 				"output": 6,
 				"cacheRead": 0.09999999999999999,
-				"cacheWrite": 0
+				"cacheWrite": 1.25
 			},
 			"contextWindow": 1050000,
 			"maxTokens": 128000,
@@ -69621,7 +70147,7 @@
 				"input": 5,
 				"output": 30,
 				"cacheRead": 0.5,
-				"cacheWrite": 0
+				"cacheWrite": 6.25
 			},
 			"contextWindow": 1050000,
 			"maxTokens": 128000,
@@ -69658,7 +70184,7 @@
 				"input": 5,
 				"output": 30,
 				"cacheRead": 0.5,
-				"cacheWrite": 0
+				"cacheWrite": 6.25
 			},
 			"contextWindow": 1050000,
 			"maxTokens": 128000,
@@ -69695,7 +70221,7 @@
 				"input": 2.5,
 				"output": 15,
 				"cacheRead": 0.25,
-				"cacheWrite": 0
+				"cacheWrite": 3.125
 			},
 			"contextWindow": 1050000,
 			"maxTokens": 128000,
@@ -69732,7 +70258,7 @@
 				"input": 2.5,
 				"output": 15,
 				"cacheRead": 0.25,
-				"cacheWrite": 0
+				"cacheWrite": 3.125
 			},
 			"contextWindow": 1050000,
 			"maxTokens": 128000,
@@ -77243,6 +77769,138 @@
 				]
 			}
 		},
+		"openai-gpt-56-luna": {
+			"id": "openai-gpt-56-luna",
+			"name": "openai-gpt-56-luna",
+			"api": "openai-completions",
+			"provider": "venice",
+			"baseUrl": "https://api.venice.ai/api/v1",
+			"reasoning": false,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1000000,
+			"maxTokens": 128000,
+			"compat": {
+				"supportsUsageInStreaming": false
+			}
+		},
+		"openai-gpt-56-luna-pro": {
+			"id": "openai-gpt-56-luna-pro",
+			"name": "openai-gpt-56-luna-pro",
+			"api": "openai-completions",
+			"provider": "venice",
+			"baseUrl": "https://api.venice.ai/api/v1",
+			"reasoning": false,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1000000,
+			"maxTokens": 128000,
+			"compat": {
+				"supportsUsageInStreaming": false
+			}
+		},
+		"openai-gpt-56-sol": {
+			"id": "openai-gpt-56-sol",
+			"name": "openai-gpt-56-sol",
+			"api": "openai-completions",
+			"provider": "venice",
+			"baseUrl": "https://api.venice.ai/api/v1",
+			"reasoning": false,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1000000,
+			"maxTokens": 128000,
+			"compat": {
+				"supportsUsageInStreaming": false
+			}
+		},
+		"openai-gpt-56-sol-pro": {
+			"id": "openai-gpt-56-sol-pro",
+			"name": "openai-gpt-56-sol-pro",
+			"api": "openai-completions",
+			"provider": "venice",
+			"baseUrl": "https://api.venice.ai/api/v1",
+			"reasoning": false,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1000000,
+			"maxTokens": 128000,
+			"compat": {
+				"supportsUsageInStreaming": false
+			}
+		},
+		"openai-gpt-56-terra": {
+			"id": "openai-gpt-56-terra",
+			"name": "openai-gpt-56-terra",
+			"api": "openai-completions",
+			"provider": "venice",
+			"baseUrl": "https://api.venice.ai/api/v1",
+			"reasoning": false,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1000000,
+			"maxTokens": 128000,
+			"compat": {
+				"supportsUsageInStreaming": false
+			}
+		},
+		"openai-gpt-56-terra-pro": {
+			"id": "openai-gpt-56-terra-pro",
+			"name": "openai-gpt-56-terra-pro",
+			"api": "openai-completions",
+			"provider": "venice",
+			"baseUrl": "https://api.venice.ai/api/v1",
+			"reasoning": false,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1000000,
+			"maxTokens": 128000,
+			"compat": {
+				"supportsUsageInStreaming": false
+			}
+		},
 		"openai-gpt-oss-120b": {
 			"id": "openai-gpt-oss-120b",
 			"name": "OpenAI GPT OSS 120B",
@@ -80352,6 +81010,35 @@
 			"contextWindow": 128000,
 			"maxTokens": 8192
 		},
+		"meta/muse-spark-1.1": {
+			"id": "meta/muse-spark-1.1",
+			"name": "Muse Spark 1.1",
+			"api": "anthropic-messages",
+			"provider": "vercel-ai-gateway",
+			"baseUrl": "https://ai-gateway.vercel.sh",
+			"reasoning": true,
+			"input": [
+				"text"
+			],
+			"cost": {
+				"input": 1.25,
+				"output": 4.25,
+				"cacheRead": 0.15,
+				"cacheWrite": 0
+			},
+			"contextWindow": 1048576,
+			"maxTokens": 1048576,
+			"thinking": {
+				"mode": "budget",
+				"efforts": [
+					"minimal",
+					"low",
+					"medium",
+					"high",
+					"xhigh"
+				]
+			}
+		},
 		"minimax/minimax-m2": {
 			"id": "minimax/minimax-m2",
 			"name": "MiniMax M2",
@@ -84999,6 +85686,34 @@
 				"omitReasoningEffort": false
 			}
 		},
+		"grok-4.5": {
+			"id": "grok-4.5",
+			"name": "Grok 4.5",
+			"api": "openai-responses",
+			"provider": "xai-oauth",
+			"baseUrl": "https://api.x.ai/v1",
+			"reasoning": true,
+			"input": [
+				"text",
+				"image"
+			],
+			"cost": {
+				"input": 0,
+				"output": 0,
+				"cacheRead": 0,
+				"cacheWrite": 0
+			},
+			"contextWindow": 500000,
+			"maxTokens": 500000,
+			"compat": {
+				"reasoningEffortMap": {
+					"minimal": "low"
+				},
+				"includeEncryptedReasoning": false,
+				"filterReasoningHistory": true,
+				"omitReasoningEffort": true
+			}
+		},
 		"grok-build": {
 			"id": "grok-build",
 			"name": "Grok Build",

package/src/provider-models/openai-compat.ts

@@ -821,6 +821,62 @@ export function openaiModelManagerOptions(config?: OpenAIModelManagerConfig): Mo
 	};
 }
 
+/** First-party gpt-5.6 SKUs that accept `reasoning: { mode: "pro" }` on the Responses APIs. */
+const OPENAI_PRO_REASONING_BASE_IDS: Record<string, true> = {
+	"gpt-5.6-luna": true,
+	"gpt-5.6-sol": true,
+	"gpt-5.6-terra": true,
+};
+const OPENAI_PRO_REASONING_PROVIDERS: Record<string, true> = { openai: true, "openai-codex": true };
+
+/**
+ * A row this generator pass owns: one of the derived `gpt-5.6-*-pro` alias ids
+ * on `openai`/`openai-codex` that carries the generated `reasoningMode` marker.
+ * A real upstream model occupying the same id has no `reasoningMode` and is
+ * never touched.
+ */
+function isGeneratedOpenAIProReasoningAlias(model: ModelSpec<Api>): boolean {
+	return (
+		OPENAI_PRO_REASONING_PROVIDERS[model.provider] === true &&
+		model.reasoningMode !== undefined &&
+		model.id.endsWith("-pro") &&
+		OPENAI_PRO_REASONING_BASE_IDS[model.id.slice(0, -"-pro".length)] === true
+	);
+}
+
+/**
+ * Re-derive the generated pro-reasoning aliases (`gpt-5.6-*-pro`) for the
+ * first-party `openai`/`openai-codex` gpt-5.6 rows. Each alias inherits the
+ * base row's metadata, requests the base wire id via `requestModelId`, and
+ * sets `reasoningMode: "pro"` so Responses-family request builders emit
+ * `reasoning: { mode: "pro" }`. Called by the models.json generator after all
+ * sources merge: stale copies of the owned aliases (previous snapshot) are
+ * dropped and re-projected from the current base rows so alias metadata always
+ * tracks the base, while a real upstream model that occupies an alias id wins
+ * and suppresses the projection.
+ */
+export function projectOpenAIProReasoningAliases(models: readonly ModelSpec<Api>[]): ModelSpec<Api>[] {
+	const kept = models.filter(model => !isGeneratedOpenAIProReasoningAlias(model));
+	const ids = new Set(kept.map(model => `${model.provider}/${model.id}`));
+	const out = [...kept];
+	for (const model of kept) {
+		if (!OPENAI_PRO_REASONING_PROVIDERS[model.provider]) continue;
+		if (!OPENAI_PRO_REASONING_BASE_IDS[model.id]) continue;
+		const aliasId = `${model.id}-pro`;
+		const aliasKey = `${model.provider}/${aliasId}`;
+		if (ids.has(aliasKey)) continue;
+		ids.add(aliasKey);
+		out.push({
+			...model,
+			id: aliasId,
+			name: `${model.name} Pro`,
+			requestModelId: model.id,
+			reasoningMode: "pro",
+		});
+	}
+	return out;
+}
+
 // ---------------------------------------------------------------------------
 // 2. Groq
 // ---------------------------------------------------------------------------
@@ -983,6 +1039,7 @@ export const XAI_OAUTH_CURATED_MODELS: readonly XAICuratedModel[] = [
 		input: ["text", "image"],
 	},
 	{ id: "grok-4.3", contextWindow: 1_000_000, name: "Grok 4.3", input: ["text", "image"] },
+	{ id: "grok-4.5", contextWindow: 500_000, name: "Grok 4.5", input: ["text", "image"] },
 	// grok-4.20-multi-agent-0309 is text-only per the bundled catalog; omit `input` for the default.
 	{ id: "grok-4.20-multi-agent-0309", contextWindow: 2_000_000, name: "Grok 4.20 (Multi-Agent)" },
 	{

package/src/types.ts

@@ -691,6 +691,13 @@ export interface Model<TApi extends Api = Api> {
 	 * everything local (selection, caching, usage attribution) keys on `id`.
 	 */
 	requestModelId?: string;
+	/**
+	 * `reasoning.mode` to send on OpenAI Responses-family requests. Set on
+	 * generated pro aliases (`gpt-5.6-*-pro` on `openai`/`openai-codex`) that
+	 * pair a base wire id (`requestModelId`) with OpenAI's pro reasoning
+	 * serving path. Absent everywhere else; providers omit the wire field.
+	 */
+	reasoningMode?: "pro";
 	name: string;
 	api: TApi;
 	provider: Provider;