@openclaw/llama-cpp-provider 2026.6.8-beta.2 → 2026.6.9-beta.1

package/dist/index.js

@@ -1,16 +1,46 @@
 import { createRequire } from "node:module";
 import { definePluginEntry } from "openclaw/plugin-sdk/plugin-entry";
+import os from "node:os";
+import path from "node:path";
 import { pathToFileURL } from "node:url";
 import { createLocalEmbeddingProvider } from "openclaw/plugin-sdk/memory-core-host-engine-embeddings";
 //#region extensions/llama-cpp/src/embedding-provider.ts
 const LLAMA_CPP_EMBEDDING_PROVIDER_ID = "local";
 const DEFAULT_LLAMA_CPP_EMBEDDING_MODEL = "hf:ggml-org/embeddinggemma-300m-qat-q8_0-GGUF/embeddinggemma-300m-qat-Q8_0.gguf";
+const DEFAULT_LLAMA_CPP_EMBEDDING_MODEL_CACHE_FILE_NAME = "hf_ggml-org_embeddinggemma-300m-qat-Q8_0.gguf";
 function normalizeOptionalString(value) {
 	return typeof value === "string" && value.trim() ? value.trim() : void 0;
 }
 function readLocalOptions(options) {
 	return options.local ?? {};
 }
+function createLlamaCppCacheKeyData(model, outputDimensionality) {
+	return {
+		provider: LLAMA_CPP_EMBEDDING_PROVIDER_ID,
+		model,
+		...typeof outputDimensionality === "number" ? { outputDimensionality } : {}
+	};
+}
+function resolveLlamaCppModelIdentity(local, modelPath, outputDimensionality) {
+	const modelCacheDir = normalizeOptionalString(local.modelCacheDir) ?? path.join(os.homedir(), ".node-llama-cpp", "models");
+	const resolvedDefaultModelPath = path.resolve(modelCacheDir, DEFAULT_LLAMA_CPP_EMBEDDING_MODEL_CACHE_FILE_NAME);
+	const resolvedModelPath = /^(?:hf:|https?:\/\/)/i.test(modelPath) ? void 0 : path.resolve(modelCacheDir, modelPath);
+	if (modelPath !== "hf:ggml-org/embeddinggemma-300m-qat-q8_0-GGUF/embeddinggemma-300m-qat-Q8_0.gguf" && resolvedModelPath !== resolvedDefaultModelPath) return {
+		model: modelPath,
+		cacheKeyData: createLlamaCppCacheKeyData(modelPath, outputDimensionality),
+		aliases: []
+	};
+	const aliasModels = new Set([resolvedDefaultModelPath, DEFAULT_LLAMA_CPP_EMBEDDING_MODEL_CACHE_FILE_NAME]);
+	if (modelPath !== "hf:ggml-org/embeddinggemma-300m-qat-q8_0-GGUF/embeddinggemma-300m-qat-Q8_0.gguf") aliasModels.add(modelPath);
+	return {
+		model: DEFAULT_LLAMA_CPP_EMBEDDING_MODEL,
+		cacheKeyData: createLlamaCppCacheKeyData(DEFAULT_LLAMA_CPP_EMBEDDING_MODEL, outputDimensionality),
+		aliases: Array.from(aliasModels, (aliasModel) => ({
+			model: aliasModel,
+			cacheKeyData: createLlamaCppCacheKeyData(aliasModel, outputDimensionality)
+		}))
+	};
+}
 function textFromEmbeddingInput(input) {
 	return typeof input === "string" ? input : input.text;
 }
@@ -56,16 +86,24 @@ function adaptMemoryEmbeddingProvider(provider) {
 		close: provider.close
 	};
 }
-async function createLlamaCppEmbeddingProvider(options, runtimeOptions = {}) {
-	const result = await createLlamaCppMemoryEmbeddingProvider(buildMemoryCreateOptions(options, options.dimensions), runtimeOptions);
-	if (!result.provider) throw new Error("llama.cpp local embedding provider was unavailable");
-	return adaptMemoryEmbeddingProvider(result.provider);
-}
 async function createLlamaCppMemoryEmbeddingProvider(options, runtimeOptions = {}) {
-	const provider = await createLocalEmbeddingProvider(buildMemoryCreateOptions(options, options.outputDimensionality), { nodeLlamaCppImportUrl: runtimeOptions.nodeLlamaCppImportUrl ?? resolveNodeLlamaCppImportUrl() });
+	const createOptions = buildMemoryCreateOptions(options, options.outputDimensionality);
+	const local = readLocalOptions(createOptions);
+	const provider = await createLocalEmbeddingProvider(createOptions, { nodeLlamaCppImportUrl: runtimeOptions.nodeLlamaCppImportUrl ?? resolveNodeLlamaCppImportUrl() });
+	const identity = resolveLlamaCppModelIdentity(local, provider.model, createOptions.outputDimensionality);
+	return {
+		provider: identity.model === provider.model ? provider : {
+			...provider,
+			model: identity.model
+		},
+		runtime: createLlamaCppEmbeddingProviderRuntime(identity)
+	};
+}
+async function createLlamaCppEmbeddingProviderResult(options, runtimeOptions = {}) {
+	const result = await createLlamaCppMemoryEmbeddingProvider(buildMemoryCreateOptions(options, options.dimensions), runtimeOptions);
 	return {
-		provider,
-		runtime: createLlamaCppEmbeddingProviderRuntime(provider)
+		provider: result.provider ? adaptMemoryEmbeddingProvider(result.provider) : null,
+		runtime: result.runtime
 	};
 }
 function buildMemoryCreateOptions(options, outputDimensionality) {
@@ -88,15 +126,13 @@ function buildMemoryCreateOptions(options, outputDimensionality) {
 		outputDimensionality
 	};
 }
-function createLlamaCppEmbeddingProviderRuntime(provider) {
+function createLlamaCppEmbeddingProviderRuntime(identity) {
 	return {
 		id: LLAMA_CPP_EMBEDDING_PROVIDER_ID,
 		inlineQueryTimeoutMs: 5 * 6e4,
 		inlineBatchTimeoutMs: 10 * 6e4,
-		cacheKeyData: {
-			provider: LLAMA_CPP_EMBEDDING_PROVIDER_ID,
-			model: provider.model
-		}
+		cacheKeyData: identity.cacheKeyData,
+		...identity.aliases.length > 0 ? { indexIdentityAliases: identity.aliases } : {}
 	};
 }
 const llamaCppEmbeddingProviderAdapter = {
@@ -104,13 +140,12 @@ const llamaCppEmbeddingProviderAdapter = {
 	defaultModel: DEFAULT_LLAMA_CPP_EMBEDDING_MODEL,
 	transport: "local",
 	formatSetupError: formatLlamaCppSetupError,
-	create: async (options) => {
-		const provider = await createLlamaCppEmbeddingProvider(options);
-		return {
-			provider,
-			runtime: createLlamaCppEmbeddingProviderRuntime(provider)
-		};
-	}
+	resolveIndexIdentity: (options) => {
+		const createOptions = buildMemoryCreateOptions(options, options.dimensions);
+		const local = readLocalOptions(createOptions);
+		return resolveLlamaCppModelIdentity(local, normalizeOptionalString(local.modelPath) ?? "hf:ggml-org/embeddinggemma-300m-qat-q8_0-GGUF/embeddinggemma-300m-qat-Q8_0.gguf", createOptions.outputDimensionality);
+	},
+	create: async (options) => await createLlamaCppEmbeddingProviderResult(options)
 };
 //#endregion
 //#region extensions/llama-cpp/index.ts

package/npm-shrinkwrap.json

@@ -1,12 +1,12 @@
 {
   "name": "@openclaw/llama-cpp-provider",
-  "version": "2026.6.8-beta.2",
+  "version": "2026.6.9-beta.1",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "@openclaw/llama-cpp-provider",
-      "version": "2026.6.8-beta.2",
+      "version": "2026.6.9-beta.1",
       "dependencies": {
         "node-llama-cpp": "3.18.1"
       }
@@ -1363,9 +1363,9 @@
       }
     },
     "node_modules/semver": {
-      "version": "7.8.2",
-      "resolved": "https://registry.npmjs.org/semver/-/semver-7.8.2.tgz",
-      "integrity": "sha512-c8jsqUZm3omBOI66G90z1Dyw5z622G8oLG+omfsHBJf3CWQTlOcwOjvOG6wtiNfW6anKm/eA39LMwMtMez2TiQ==",
+      "version": "7.8.4",
+      "resolved": "https://registry.npmjs.org/semver/-/semver-7.8.4.tgz",
+      "integrity": "sha512-rUCObTnP32Q08R2uuIrt7r9PlEonuTmtuXYcW6s5kjdlj3xbnwe+21yXptAUYcMAABLkYYTtnmzb3w3EDZfueA==",
       "license": "ISC",
       "bin": {
         "semver": "bin/semver.js"
@@ -1543,9 +1543,9 @@
       }
     },
     "node_modules/tar": {
-      "version": "7.5.15",
-      "resolved": "https://registry.npmjs.org/tar/-/tar-7.5.15.tgz",
-      "integrity": "sha512-dzGK0boVlC4W5QFuQN1EFSl3bIDYsk7Tj40U6eIBnK2k/8ml7TZ5agbI5j5+qnoVcAA+rNtBml8SEiLxZpNqRQ==",
+      "version": "7.5.16",
+      "resolved": "https://registry.npmjs.org/tar/-/tar-7.5.16.tgz",
+      "integrity": "sha512-56adEpPMouktRlBLXiaYFFzZ/3+JXa8P9n7WbR+ibIjtviN55mEaOkiysCnPnWm+7kkui1Dn8J9l+g6zV8731w==",
       "license": "BlueOak-1.0.0",
       "dependencies": {
         "@isaacs/fs-minipass": "^4.0.0",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@openclaw/llama-cpp-provider",
-  "version": "2026.6.8-beta.2",
+  "version": "2026.6.9-beta.1",
   "description": "OpenClaw llama.cpp embedding provider plugin",
   "repository": {
     "type": "git",
@@ -23,10 +23,10 @@
       "minHostVersion": ">=2026.6.2"
     },
     "compat": {
-      "pluginApi": ">=2026.6.8-beta.2"
+      "pluginApi": ">=2026.6.9-beta.1"
     },
     "build": {
-      "openclawVersion": "2026.6.8-beta.2"
+      "openclawVersion": "2026.6.9-beta.1"
     },
     "release": {
       "bundleRuntimeDependencies": false,
@@ -44,7 +44,7 @@
     "README.md"
   ],
   "peerDependencies": {
-    "openclaw": ">=2026.6.8-beta.2"
+    "openclaw": ">=2026.6.9-beta.1"
   },
   "peerDependenciesMeta": {
     "openclaw": {