openclaw-amem 1.4.2 → 1.4.3

package/dist/index.js

@@ -62,22 +62,37 @@ function getEmbeddingPooling() {
   const basename2 = getEmbeddingModel().split("/").pop()?.toLowerCase() ?? "";
   return CLS_POOLED_MODELS.has(basename2) ? "cls" : "mean";
 }
+function getEmbeddingDevice() {
+  return process.env.AMEM_EMBED_DEVICE?.trim() || void 0;
+}
+function getEmbeddingDtype() {
+  return process.env.AMEM_EMBED_DTYPE?.trim() || void 0;
+}
+function extractorKey() {
+  return `${getEmbeddingModel()}|${getEmbeddingDevice() ?? ""}|${getEmbeddingDtype() ?? ""}`;
+}
 async function getExtractor() {
-  const wanted = getEmbeddingModel();
-  if (extractor && loadedModelName === wanted) return extractor;
+  const wanted = extractorKey();
+  if (extractor && loadedKey === wanted) return extractor;
   if (!pipeline) {
     const mod = await import("@huggingface/transformers");
     pipeline = mod.pipeline;
   }
-  extractor = await pipeline("feature-extraction", wanted, {
-    revision: "main"
+  const device = getEmbeddingDevice();
+  const dtype = getEmbeddingDtype();
+  extractor = await pipeline("feature-extraction", getEmbeddingModel(), {
+    revision: "main",
+    // Omitted entirely when unset, so an unconfigured install gets exactly the
+    // library defaults it got before these existed.
+    ...device ? { device } : {},
+    ...dtype ? { dtype } : {}
   });
-  loadedModelName = wanted;
+  loadedKey = wanted;
   cachedDim = null;
   return extractor;
 }
 async function getEmbeddingDim() {
-  if (cachedDim !== null && loadedModelName === getEmbeddingModel()) return cachedDim;
+  if (cachedDim !== null && loadedKey === extractorKey()) return cachedDim;
   const probe = await encode("dimension probe");
   cachedDim = probe.length;
   return cachedDim;
@@ -140,13 +155,13 @@ function cosineSimilarity(a, b) {
   for (let i = 0; i < a.length; i++) dot += a[i] * b[i];
   return dot;
 }
-var pipeline, extractor, loadedModelName, cachedDim, DEFAULT_EMBEDDING_MODEL, CLS_POOLED_MODELS;
+var pipeline, extractor, loadedKey, cachedDim, DEFAULT_EMBEDDING_MODEL, CLS_POOLED_MODELS;
 var init_embedding = __esm({
   "../amem-core/src/embedding.ts"() {
     "use strict";
     pipeline = null;
     extractor = null;
-    loadedModelName = null;
+    loadedKey = null;
     cachedDim = null;
     DEFAULT_EMBEDDING_MODEL = "Xenova/paraphrase-multilingual-MiniLM-L12-v2";
     CLS_POOLED_MODELS = /* @__PURE__ */ new Set([
@@ -179,6 +194,20 @@ var init_auth = __esm({
 });
 
 // ../amem-core/src/storage.ts
+function migrationHint(collection, targetModel) {
+  return `migrate to a new collection:
+
+  AMEM_EMBED_MODEL=${targetModel} \\
+    npx --package=@amemhq/core amem-migrate --to ${collection}_v2
+
+That is a dry run; add --apply to write. "${collection}" is only read, so nothing is lost either way. When it looks right, point whatever names this collection at the new one \u2014 AMEM_COLLECTION, or the plugin's "collection" setting if this agent has its own. See https://amem.owo.lc/reference/embedding-models.`;
+}
+async function recordCollectionModel(collection, model) {
+  try {
+    await qdrant("PATCH", `/collections/${collection}`, { metadata: { embedding_model: model } });
+  } catch {
+  }
+}
 async function qdrant(method, path6, body) {
   const res = await fetch(`${QDRANT_URL}${path6}`, {
     method,
@@ -219,6 +248,14 @@ async function ensureCollection(collectionName) {
         throw new EmbeddingDimensionMismatchError(col, collectionDim, modelDim, getEmbeddingModel());
       }
     }
+    const recorded = existing.config?.metadata?.embedding_model;
+    const current = getEmbeddingModel();
+    if (typeof recorded === "string" && recorded !== current) {
+      throw new EmbeddingModelMismatchError(col, recorded, current);
+    }
+    if (recorded === void 0) {
+      await recordCollectionModel(col, current);
+    }
     markReady();
     return;
   }
@@ -230,6 +267,7 @@ async function ensureCollection(collectionName) {
   } catch (err) {
     if (!(err instanceof Error) || !err.message.includes("already exists")) throw err;
   }
+  await recordCollectionModel(col, getEmbeddingModel());
   await qdrant("PUT", `/collections/${col}/index`, {
     field_name: "agent_id",
     field_schema: "keyword"
@@ -275,6 +313,7 @@ async function collectionDimRaw(collection) {
 }
 async function createCollectionRaw(collection, size) {
   await qdrant("PUT", `/collections/${collection}`, { vectors: { size, distance: "Cosine" } });
+  await recordCollectionModel(collection, getEmbeddingModel());
   for (const field_name of ["agent_id", "hash", "topics", "subjects"]) {
     await qdrant("PUT", `/collections/${collection}/index`, { field_name, field_schema: "keyword" });
   }
@@ -667,7 +706,7 @@ async function invalidateNote(id, callerAgentId) {
 async function patchNotePayload(id, fields) {
   return makeCrud(getCollection()).patchNotePayload(id, fields);
 }
-var QDRANT_URL, getCollection, EmbeddingDimensionMismatchError, _collectionReady, _collectionReadyMap;
+var QDRANT_URL, getCollection, EmbeddingDimensionMismatchError, EmbeddingModelMismatchError, _collectionReady, _collectionReadyMap;
 var init_storage = __esm({
   "../amem-core/src/storage.ts"() {
     "use strict";
@@ -679,7 +718,7 @@ var init_storage = __esm({
       constructor(collection, collectionDim, modelDim, model) {
         super(
           `Collection "${collection}" stores ${collectionDim}-dimension vectors, but the embedding model "${model}" produces ${modelDim}. Qdrant fixes a collection's vector size at creation and cannot change it, so writes and searches would both fail.
-Either set AMEM_EMBED_MODEL back to the model this collection was built with, or migrate: build a new collection with the new model, backfill it, then point AMEM_COLLECTION at it. See docs/reference/embedding-models.md.`
+Either set AMEM_EMBED_MODEL back to the model this collection was built with, or ${migrationHint(collection, model)}`
         );
         this.collection = collection;
         this.collectionDim = collectionDim;
@@ -692,6 +731,21 @@ Either set AMEM_EMBED_MODEL back to the model this collection was built with, or
       modelDim;
       model;
     };
+    EmbeddingModelMismatchError = class extends Error {
+      constructor(collection, collectionModel, configuredModel) {
+        super(
+          `Collection "${collection}" was built with the embedding model "${collectionModel}", but this process is configured for "${configuredModel}". Both produce vectors of the same width, so nothing would fail \u2014 searches would just quietly compare vectors from two different models.
+Either set AMEM_EMBED_MODEL back to "${collectionModel}", or ` + migrationHint(collection, configuredModel)
+        );
+        this.collection = collection;
+        this.collectionModel = collectionModel;
+        this.configuredModel = configuredModel;
+        this.name = "EmbeddingModelMismatchError";
+      }
+      collection;
+      collectionModel;
+      configuredModel;
+    };
     _collectionReady = false;
     _collectionReadyMap = /* @__PURE__ */ new Map();
   }
@@ -2403,6 +2457,7 @@ __export(src_exports, {
   DEFAULT_CRUD_UPDATE_MIN_SIM: () => DEFAULT_CRUD_UPDATE_MIN_SIM,
   DEFAULT_EMBEDDING_MODEL: () => DEFAULT_EMBEDDING_MODEL,
   EmbeddingDimensionMismatchError: () => EmbeddingDimensionMismatchError,
+  EmbeddingModelMismatchError: () => EmbeddingModelMismatchError,
   addEpisodic: () => addEpisodic,
   addMemory: () => addMemory,
   canRead: () => canRead,
@@ -2417,7 +2472,9 @@ __export(src_exports, {
   encode: () => encode,
   ensureCollection: () => ensureCollection,
   generateReviewBatch: () => generateReviewBatch,
+  getEmbeddingDevice: () => getEmbeddingDevice,
   getEmbeddingDim: () => getEmbeddingDim,
+  getEmbeddingDtype: () => getEmbeddingDtype,
   getEmbeddingModel: () => getEmbeddingModel,
   getEmbeddingPooling: () => getEmbeddingPooling,
   getNote: () => getNote,
@@ -2552,7 +2609,7 @@ function register(api) {
     `openclaw-amem: registered (native TS, Qdrant, default agent_id=${defaultScope.agentId}, default collection=${pluginConfig.collection ?? "amem_notes (default)"}, per-agent scope resolved per call)`
   );
   ensureCollection(pluginConfig.collection).catch((e) => {
-    if (e instanceof EmbeddingDimensionMismatchError) {
+    if (e instanceof EmbeddingDimensionMismatchError || e instanceof EmbeddingModelMismatchError) {
       logger.error(`openclaw-amem: memory is UNUSABLE \u2014 ${e.message}`);
     } else {
       logger.warn(`openclaw-amem: ensureCollection failed \u2014 ${e.message}`);

package/openclaw.plugin.json

@@ -2,7 +2,7 @@
   "id": "openclaw-amem",
   "name": "amem",
   "description": "Catches memories that contradict each other. Notes rewrite themselves as new ones arrive, link into a graph, and stay separated per agent and per person. Runs on a small model. Requires a local Qdrant. 中文走 jieba 分词，提示词有中文版。",
-  "version": "1.4.2",
+  "version": "1.4.3",
   "kind": "memory",
   "openclaw": {
     "compat": {
@@ -40,6 +40,8 @@
           "AMEM_CRUD_UPDATE_MIN_SIM",
           "AMEM_EMBED_MODEL",
           "AMEM_EMBED_POOLING",
+          "AMEM_EMBED_DEVICE",
+          "AMEM_EMBED_DTYPE",
           "AMEM_COLLECTION",
           "AMEM_DATA_DIR",
           "AMEM_EVO_COUNTER_PATH",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "openclaw-amem",
-  "version": "1.4.2",
+  "version": "1.4.3",
   "description": "Catches memories that contradict each other. Notes rewrite themselves as new ones arrive, link into a graph, and stay separated per agent and per person. Runs on a small model. Requires a local Qdrant. 中文走 jieba 分词，提示词有中文版。",
   "license": "MIT",
   "main": "dist/index.js",