diff --git a/docs/en/development/memory-layer.md b/docs/en/development/memory-layer.md
index 53d97a2ff..956a5ac0b 100644
--- a/docs/en/development/memory-layer.md
+++ b/docs/en/development/memory-layer.md
@@ -90,20 +90,14 @@ the same citation fields through list and exact-read operations.
 unavailable. Explicit `vector` and `hybrid` requests fail when the configured profile does not provide that
 capability.
 
-## Enable SQLite Vec1
+## Enable SQLite vector search
 
-SQLite vector search is enabled only when both a Vec1 0.7 or newer loadable extension and an embedding model are
-supplied. PowerContext does not install or build the native extension; provide a compatible library for the target
-operating system and architecture:
+SQLite vector search is enabled when an embedding model is supplied. The `powercontext[builtin]` extra bundles
+`sqlite-vec`, so no extension path or separate native-library installation is required:
 
 ```python
-from pathlib import Path
-
 config = BuiltinConfig(
-    database=SQLiteConfig(
-        url="sqlite+aiosqlite:///powercontext.db",
-        vec1_extension=Path("/opt/sqlite-extensions/vec1"),
-    )
+    database=SQLiteConfig(url="sqlite+aiosqlite:///powercontext.db")
 )
 async with open_builtin_runtime(
     config,
@@ -112,7 +106,7 @@ async with open_builtin_runtime(
     ...
 ```
 
-The SQLite profile composes FTS5 and Vec1 strategies. It reports `fts`, `vector`, and `hybrid` through Memory
+The SQLite profile composes FTS5 and sqlite-vec strategies. It reports `fts`, `vector`, and `hybrid` through Memory
 capabilities.
 Stored projections and query vectors must use the same `EmbeddingProfile`, including model name, dimension, distance,
 and normalization. Changing that profile requires rebuilding projections before vector search resumes.
@@ -145,7 +139,7 @@ async with open_builtin_runtime(
 
 The OceanBase profile uses the same index composition as SQLite. Its full-text strategy is always available. Supplying
 an embedding model adds a `VECTOR` projection and HNSW strategy, enabling `vector` and `hybrid` modes. SQLite FTS5 and
-OceanBase FULLTEXT therefore serve the same Runtime and Server search calls; Vec1 and HNSW do the same for vector
+OceanBase FULLTEXT therefore serve the same Runtime and Server search calls; sqlite-vec and HNSW do the same for vector
 search.
 
 ## Operational checks
@@ -155,7 +149,7 @@ Before serving requests, verify:
 - the selected profile opens and initializes successfully;
 - each tenant or project maps to the intended scope ID;
 - scheduled extraction has a candidate pipeline;
-- Vec1 configuration includes a matching embedding model;
+- SQLite vector search has a matching embedding model;
 - OceanBase vector search has a matching embedding model;
 - capability responses match the indexes actually initialized;
 - database and scheduler resources close with the process lifecycle.
diff --git a/docs/en/development/pydantic-ai-inference.md b/docs/en/development/pydantic-ai-inference.md
index b21bf668f..85d8ac914 100644
--- a/docs/en/development/pydantic-ai-inference.md
+++ b/docs/en/development/pydantic-ai-inference.md
@@ -33,15 +33,14 @@ Vector search needs the embedding model and its complete deployment profile:
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_MODEL="provider:embedding-model"
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_PROFILE_ID="project-embedding-v1"
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_DIMENSION="1536"
-export POWERCONTEXT_SERVER_DATABASE_VEC1_EXTENSION="/opt/sqlite-extensions/vec1"
 ```
 
 Provider credentials remain in the environment variables understood by the selected Pydantic AI provider. They are
 not fields on PowerContext models.
 
 The Server rejects a partial embedding profile. `embedding_model`, `embedding_profile_id`, and `embedding_dimension`
-must be configured together. Vec1 also requires that embedding configuration because the index dimension and stored
-vectors must agree.
+must be configured together. SQLite vector search uses that embedding configuration because the index dimension and
+stored vectors must agree.
 
 ## Compose generation directly
 
@@ -99,9 +98,9 @@ embedding_model = PydanticAIEmbeddingModel(
 )
 ```
 
-Pass this adapter to `open_builtin_contexts()` or `open_builtin_runtime()` with a `SQLiteConfig` that selects the Vec1
-extension. The adapter verifies output count, order, dimension, and finite numeric values, then applies the declared
-unit normalization before vectors reach persistence.
+Pass this adapter to `open_builtin_contexts()` or `open_builtin_runtime()` with a `SQLiteConfig`. The bundled
+sqlite-vec index is enabled automatically. The adapter verifies output count, order, dimension, and finite numeric
+values, then applies the declared unit normalization before vectors reach persistence.
 
 An `EmbeddingProfile` is a deployment contract, not descriptive metadata. Stored projections and query embeddings
 must use the same profile. When the model, dimension, or normalization changes, rebuild Memory projections from the
diff --git a/docs/en/development/remote-access-implementation.md b/docs/en/development/remote-access-implementation.md
index 542a50f4f..71257f8bf 100644
--- a/docs/en/development/remote-access-implementation.md
+++ b/docs/en/development/remote-access-implementation.md
@@ -56,7 +56,7 @@ export POWERCONTEXT_SERVER_DATABASE_KIND="oceanbase"
 export POWERCONTEXT_SERVER_DATABASE_URL="mysql+aoceanbase://user:password@host:2881/powercontext?charset=utf8mb4"
 ```
 
-Both database choices expose full-text search through the same Server API. With an embedding model, SQLite uses Vec1
+Both database choices expose full-text search through the same Server API. With an embedding model, SQLite uses sqlite-vec
 and OceanBase uses HNSW for `vector` and `hybrid` searches.
 
 Inference configuration is documented in [Configure Pydantic AI inference](pydantic-ai-inference.md).
diff --git a/docs/en/docs/reference/configuration.md b/docs/en/docs/reference/configuration.md
index 4a193e379..ebeadfaf5 100644
--- a/docs/en/docs/reference/configuration.md
+++ b/docs/en/docs/reference/configuration.md
@@ -178,24 +178,21 @@ Optional settings are `POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_NORMALIZATION` an
 
 Embedding normalization defaults to `unit`.
 
-### SQLite Vec1
+### SQLite vector search
 
-SQLite vector and hybrid search additionally require a
-[SQLite Vec1](https://sqlite.org/vec1/doc/trunk/doc/vec1.md) 0.7 or newer loadable extension. PowerContext does not
-download, build, or update this native library. Obtain it for the Server's operating system and architecture, then
-set its path together with the complete embedding profile:
+SQLite vector and hybrid search use [sqlite-vec](https://alexgarcia.xyz/sqlite-vec/), which is bundled with the
+`powercontext[builtin]` dependency set. Configure the complete embedding profile; no extension path is needed:
 
 ```bash
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_MODEL=provider:embedding-model
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_PROFILE_ID=embedding-model-v1
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_DIMENSION=1024
 export POWERCONTEXT_SERVER_DATABASE_URL=sqlite+aiosqlite:////srv/powercontext/powercontext.db
-export POWERCONTEXT_SERVER_DATABASE_VEC1_EXTENSION=/opt/sqlite-extensions/vec1
 powercontext server run
 ```
 
-The extension path must identify a library that the SQLite loader can open. PowerContext loads and probes the
-extension when the Server opens the database; startup fails if the library is incompatible or older than 0.7.
+PowerContext loads and probes the bundled extension when the Server opens the database. Startup fails if the package
+does not contain a library compatible with the current platform or SQLite build.
 
 In another terminal, confirm that the initialized runtime reports vector and hybrid search:
 
@@ -203,8 +200,7 @@ In another terminal, confirm that the initialized runtime reports vector and hyb
 powercontext capabilities
 ```
 
-If Vec1 is unavailable, leave `POWERCONTEXT_SERVER_DATABASE_VEC1_EXTENSION` unset. SQLite full-text search remains
-available without an embedding model or native extension.
+SQLite full-text search remains available when no embedding model is configured.
 
 ## CLI Server connection
 
diff --git a/docs/zh/development/memory-layer.md b/docs/zh/development/memory-layer.md
index 64b2673c3..0bb11b2fb 100644
--- a/docs/zh/development/memory-layer.md
+++ b/docs/zh/development/memory-layer.md
@@ -88,19 +88,14 @@ operation 返回相同的 citation 字段。
 `mode="auto"` 会选择当前可用的最强模式，并可在 query embedding 暂时不可用时回退到 FTS。显式请求 `vector`
 或 `hybrid` 时，如果 profile 没有提供相应能力，操作会失败。
 
-## 启用 SQLite Vec1
+## 启用 SQLite 向量检索
 
-只有同时提供 0.7 或更高版本的 Vec1 loadable extension 和 embedding model，SQLite 才会启用向量检索。
-PowerContext 不负责安装或构建这个 native extension；请提供适用于目标操作系统和架构的 library：
+提供 embedding model 后，SQLite 会启用向量检索。`powercontext[builtin]` 已捆绑 `sqlite-vec`，无需配置 extension
+路径或单独安装 native library：
 
 ```python
-from pathlib import Path
-
 config = BuiltinConfig(
-    database=SQLiteConfig(
-        url="sqlite+aiosqlite:///powercontext.db",
-        vec1_extension=Path("/opt/sqlite-extensions/vec1"),
-    )
+    database=SQLiteConfig(url="sqlite+aiosqlite:///powercontext.db")
 )
 async with open_builtin_runtime(
     config,
@@ -109,7 +104,7 @@ async with open_builtin_runtime(
     ...
 ```
 
-SQLite profile 会组合 FTS5 和 Vec1 strategy，并通过 Memory capabilities 报告 `fts`、`vector` 和 `hybrid`。持久化
+SQLite profile 会组合 FTS5 和 sqlite-vec strategy，并通过 Memory capabilities 报告 `fts`、`vector` 和 `hybrid`。持久化
 projection 与 query vector 必须使用同一个 `EmbeddingProfile`，包括 model name、dimension、distance 和
 normalization。更换 profile 后，应先重建 projection，再恢复 vector search。
 
@@ -141,7 +136,7 @@ async with open_builtin_runtime(
 
 OceanBase profile 与 SQLite 使用相同的 index 组合方式。全文 strategy 始终可用；提供 embedding model 后，会增加
 `VECTOR` projection 和 HNSW strategy，并启用 `vector` 与 `hybrid` mode。SQLite FTS5 与 OceanBase FULLTEXT
-服务于同一组 Runtime 和 Server search 调用，Vec1 与 HNSW 也通过同一接口提供向量检索。
+服务于同一组 Runtime 和 Server search 调用，sqlite-vec 与 HNSW 也通过同一接口提供向量检索。
 
 ## 运行检查
 
@@ -150,7 +145,7 @@ OceanBase profile 与 SQLite 使用相同的 index 组合方式。全文 strateg
 - 所选 profile 能够成功打开并完成初始化；
 - 每个 tenant 或 project 映射到预期的 scope ID；
 - 定时 extraction 已经配置 candidate pipeline；
-- Vec1 配置包含匹配的 embedding model；
+- SQLite vector search 配置了匹配的 embedding model；
 - OceanBase vector search 配置了匹配的 embedding model；
 - capability response 与实际初始化的 index 一致；
 - database 和 scheduler 资源会随进程生命周期关闭。
diff --git a/docs/zh/development/pydantic-ai-inference.md b/docs/zh/development/pydantic-ai-inference.md
index ce438799f..70b65463e 100644
--- a/docs/zh/development/pydantic-ai-inference.md
+++ b/docs/zh/development/pydantic-ai-inference.md
@@ -32,13 +32,12 @@ vector search 需要 embedding model 和完整的 deployment profile：
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_MODEL="provider:embedding-model"
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_PROFILE_ID="project-embedding-v1"
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_DIMENSION="1536"
-export POWERCONTEXT_SERVER_DATABASE_VEC1_EXTENSION="/opt/sqlite-extensions/vec1"
 ```
 
 provider credential 仍使用所选 Pydantic AI provider 支持的环境变量，不属于 PowerContext model 字段。
 
 Server 会拒绝不完整的 embedding profile。`embedding_model`、`embedding_profile_id` 和
-`embedding_dimension` 必须一起配置。Vec1 也依赖这组配置，因为 index dimension 必须与持久化向量一致。
+`embedding_dimension` 必须一起配置。SQLite vector search 使用这组配置，因为 index dimension 必须与持久化向量一致。
 
 ## 直接组合 generation
 
@@ -95,8 +94,8 @@ embedding_model = PydanticAIEmbeddingModel(
 )
 ```
 
-将这个 adapter 传给 `open_builtin_contexts()` 或 `open_builtin_runtime()`，并通过 `SQLiteConfig` 选择 Vec1
-extension。向量进入持久化之前，adapter 会校验输出数量、顺序、dimension 和数值有效性，并执行 profile 声明的单位归一化。
+将这个 adapter 与 `SQLiteConfig` 一起传给 `open_builtin_contexts()` 或 `open_builtin_runtime()` 后，会自动启用捆绑的
+sqlite-vec index。向量进入持久化之前，adapter 会校验输出数量、顺序、dimension 和数值有效性，并执行 profile 声明的单位归一化。
 
 `EmbeddingProfile` 是 deployment contract，不是描述性 metadata。持久化 projection 和 query embedding 必须使用
 同一个 profile。model、dimension 或 normalization 发生变化后，应从权威 Memory revision 重建 projection。
diff --git a/docs/zh/development/remote-access-implementation.md b/docs/zh/development/remote-access-implementation.md
index ed59a46b0..4ca94f3eb 100644
--- a/docs/zh/development/remote-access-implementation.md
+++ b/docs/zh/development/remote-access-implementation.md
@@ -54,7 +54,7 @@ export POWERCONTEXT_SERVER_DATABASE_KIND="oceanbase"
 export POWERCONTEXT_SERVER_DATABASE_URL="mysql+aoceanbase://user:password@host:2881/powercontext?charset=utf8mb4"
 ```
 
-两种 database 都通过同一组 Server API 提供全文检索。配置 embedding model 后，SQLite 使用 Vec1，OceanBase
+两种 database 都通过同一组 Server API 提供全文检索。配置 embedding model 后，SQLite 使用 sqlite-vec，OceanBase
 使用 HNSW 提供 `vector` 和 `hybrid` 检索。
 
 inference 配置见[配置 Pydantic AI 推理](pydantic-ai-inference.md)。
diff --git a/docs/zh/docs/reference/configuration.md b/docs/zh/docs/reference/configuration.md
index ef3ce3044..5a4e328c9 100644
--- a/docs/zh/docs/reference/configuration.md
+++ b/docs/zh/docs/reference/configuration.md
@@ -172,24 +172,21 @@ export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_DIMENSION=1024
 
 Embedding normalization 默认为 `unit`。
 
-### SQLite Vec1
+### SQLite 向量检索
 
-SQLite vector 和 hybrid search 还需要 0.7 或更高版本的
-[SQLite Vec1](https://sqlite.org/vec1/doc/trunk/doc/vec1.md) loadable extension。PowerContext 不负责下载、构建或更新
-这个 native library。请先获取适用于 Server 操作系统和架构的构建产物，再同时配置 extension 路径和完整的
-embedding profile：
+SQLite vector 和 hybrid search 使用 [sqlite-vec](https://alexgarcia.xyz/sqlite-vec/)，它已包含在
+`powercontext[builtin]` 依赖中。只需配置完整的 embedding profile，无需配置 extension 路径：
 
 ```bash
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_MODEL=provider:embedding-model
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_PROFILE_ID=embedding-model-v1
 export POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_DIMENSION=1024
 export POWERCONTEXT_SERVER_DATABASE_URL=sqlite+aiosqlite:////srv/powercontext/powercontext.db
-export POWERCONTEXT_SERVER_DATABASE_VEC1_EXTENSION=/opt/sqlite-extensions/vec1
 powercontext server run
 ```
 
-extension 路径必须指向 SQLite loader 可以打开的 library。Server 打开数据库时，PowerContext 会加载并探测该
-extension；如果 library 不兼容或版本低于 0.7，启动会失败。
+Server 打开数据库时，PowerContext 会加载并探测捆绑的 extension；如果当前 platform 或 SQLite build 与 package
+中的 library 不兼容，启动会失败。
 
 在另一个终端确认初始化后的 Runtime 已报告 vector 和 hybrid search：
 
@@ -197,8 +194,7 @@ extension；如果 library 不兼容或版本低于 0.7，启动会失败。
 powercontext capabilities
 ```
 
-如果没有可用的 Vec1，请不要设置 `POWERCONTEXT_SERVER_DATABASE_VEC1_EXTENSION`。即使没有 embedding model 或
-native extension，SQLite full-text search 仍然可用。
+没有配置 embedding model 时，SQLite full-text search 仍然可用。
 
 ## CLI Server 连接
 
diff --git a/e2e/bub/uv.lock b/e2e/bub/uv.lock
index 3e803e6c5..60cc5df6d 100644
--- a/e2e/bub/uv.lock
+++ b/e2e/bub/uv.lock
@@ -1643,6 +1643,8 @@ requires-dist = [
     { name = "rfc8785", specifier = ">=0.1.4,<1" },
     { name = "sqlalchemy", extras = ["asyncio"], marker = "extra == 'builtin'", specifier = ">=2,<3" },
     { name = "sqlalchemy", extras = ["asyncio"], marker = "extra == 'server'", specifier = ">=2,<3" },
+    { name = "sqlite-vec", marker = "extra == 'builtin'", specifier = ">=0.1.9,<0.2" },
+    { name = "sqlite-vec", marker = "extra == 'server'", specifier = ">=0.1.9,<0.2" },
     { name = "typer", marker = "extra == 'cli'", specifier = ">=0.16,<1" },
     { name = "uvicorn", marker = "extra == 'server'", specifier = ">=0.34,<1" },
 ]
diff --git a/pyproject.toml b/pyproject.toml
index 8cf2989e9..7f898291d 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -43,6 +43,7 @@ builtin = [
     "pydantic-settings>=2.7,<3",
     "pyobvector>=0.2.28,<0.3",
     "sqlalchemy[asyncio]>=2,<3",
+    "sqlite-vec>=0.1.9,<0.2",
 ]
 client = [
     "httpx[socks]>=0.28,<1",
diff --git a/src/powercontext/builtin/persistence/sqlite/memory_index.py b/src/powercontext/builtin/persistence/sqlite/memory_index.py
index 86d8347e0..36c22e162 100644
--- a/src/powercontext/builtin/persistence/sqlite/memory_index.py
+++ b/src/powercontext/builtin/persistence/sqlite/memory_index.py
@@ -12,15 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-"""SQLite Memory search indexes using FTS5 and Vec1."""
+"""SQLite Memory search indexes using FTS5 and sqlite-vec."""
 
 from __future__ import annotations
 
 import json
-import re
 import struct
 from collections.abc import Mapping
-from pathlib import Path
 from typing import Any
 
 from sqlalchemy import (
@@ -36,6 +34,7 @@
     select,
     text,
 )
+from sqlalchemy.exc import SQLAlchemyError
 from sqlalchemy.ext.asyncio import AsyncConnection
 
 from powercontext.artifacts import ArtifactRef
@@ -105,11 +104,10 @@
 )
 
 
-SQLITE_MEMORY_VEC1_TABLES = (SQLITE_MEMORY_VECTOR_ENTRIES_TABLE,)
+SQLITE_MEMORY_VECTOR_TABLES = (SQLITE_MEMORY_VECTOR_ENTRIES_TABLE,)
 
 
 _FTS_TABLE_NAME = "pc_memory_entry_fts"
-_MINIMUM_VEC1_VERSION = (0, 7)
 _CREATE_FTS_SQL = """
 CREATE VIRTUAL TABLE IF NOT EXISTS pc_memory_entry_fts USING fts5(
     scope_id UNINDEXED,
@@ -152,27 +150,28 @@
     )
     """
 )
-_CREATE_VECTOR_SQL = "CREATE VIRTUAL TABLE IF NOT EXISTS pc_memory_entry_vec USING vec1(embedding)"
 _DELETE_VECTOR_SQL = text("DELETE FROM pc_memory_entry_vec WHERE rowid = :vector_id")
 _INSERT_VECTOR_SQL = text("INSERT INTO pc_memory_entry_vec (rowid, embedding) VALUES (:vector_id, :embedding)")
 _SELECT_VECTOR_SQL = text("SELECT embedding FROM pc_memory_entry_vec WHERE rowid = :vector_id")
 _VECTOR_SEARCH_SQL = text(
     """
-    WITH candidates AS (
-        SELECT rowid, embedding
-        FROM pc_memory_entry_vec(:query_vector, :parameters)
+    WITH nearest AS (
+        SELECT rowid, distance
+        FROM pc_memory_entry_vec
+        WHERE embedding MATCH :query_vector
+          AND k = :neighbor_limit
     )
     SELECT m.memory_artifact_id, m.head_revision, m.entry_id, m.entry_version_id, v.text,
-           vec1_l2_distance(:query_vector, c.embedding) AS distance
-    FROM candidates AS c
-    JOIN pc_memory_vector_entries AS m ON m.vector_id = c.rowid
+           nearest.distance
+    FROM nearest
+    JOIN pc_memory_vector_entries AS m ON m.vector_id = nearest.rowid
     JOIN pc_memory_entry_versions AS v
       ON v.scope_id = m.scope_id
      AND v.memory_artifact_id = m.memory_artifact_id
      AND v.entry_version_id = m.entry_version_id
     WHERE m.scope_id = :scope_id
       AND m.memory_artifact_id IN (SELECT value FROM json_each(:memory_artifact_ids))
-    ORDER BY vec1_l2_distance(:query_vector, c.embedding),
+    ORDER BY nearest.distance,
              m.memory_artifact_id, m.entry_id, m.entry_version_id
     LIMIT :candidate_limit
     """
@@ -304,31 +303,29 @@ async def _insert_row(
         await connection.execute(_INSERT_FTS_SQL, values)
 
 
-class SQLiteMemoryVec1Index:
-    """SQLite Vec1 strategy over rebuildable active-head embeddings."""
+class SQLiteMemoryVectorIndex:
+    """sqlite-vec strategy over rebuildable active-head embeddings."""
 
-    tables: tuple[Table, ...] = SQLITE_MEMORY_VEC1_TABLES
+    tables: tuple[Table, ...] = SQLITE_MEMORY_VECTOR_TABLES
 
-    def __init__(self, extension: str | Path, profile: EmbeddingProfile) -> None:
+    def __init__(self, profile: EmbeddingProfile) -> None:
         if profile.dimension < 1 or profile.distance != "l2" or profile.normalization != "unit":
             raise CapabilityNotSupportedError(
                 "vector",
-                "Vec1 requires a positive unit-normalized L2 embedding profile",
+                "sqlite-vec requires a positive unit-normalized L2 embedding profile",
             )
-        self.extension = str(extension)
         self.profile = profile
         self.capabilities = MemoryCapabilities(vector=True, embedding_profile=profile, fts=False)
 
     async def initialize(self, connection: AsyncConnection, /) -> None:
         if connection.dialect.name != "sqlite":
-            raise CapabilityNotSupportedError("sqlite-vec1")
+            raise CapabilityNotSupportedError("sqlite-vec")
         try:
-            info = (await connection.exec_driver_sql("SELECT vec1_info()")).scalar_one_or_none()
-        except Exception as error:
-            raise CapabilityNotSupportedError("vector", "SQLite Vec1 probe failed") from error
-        _validate_vec1_info(info)
-        try:
-            await connection.exec_driver_sql(_CREATE_VECTOR_SQL)
+            await connection.exec_driver_sql("SELECT vec_version()")
+            await connection.exec_driver_sql(
+                "CREATE VIRTUAL TABLE IF NOT EXISTS pc_memory_entry_vec "
+                f"USING vec0(embedding float[{self.profile.dimension}])"
+            )
             probe = _pack_vector((0.0,) * self.profile.dimension)
             await connection.execute(
                 _INSERT_VECTOR_SQL,
@@ -336,15 +333,15 @@ async def initialize(self, connection: AsyncConnection, /) -> None:
             )
             row = (
                 await connection.exec_driver_sql(
-                    "SELECT rowid FROM pc_memory_entry_vec(?, ?)",
-                    (probe, json.dumps({"k": 1}, separators=(",", ":"))),
+                    "SELECT rowid FROM pc_memory_entry_vec WHERE embedding MATCH ? AND k = 1",
+                    (probe,),
                 )
             ).one_or_none()
             await connection.execute(_DELETE_VECTOR_SQL, {"vector_id": -1})
-        except Exception as error:
-            raise CapabilityNotSupportedError("vector", "SQLite Vec1 probe failed") from error
+        except SQLAlchemyError as error:
+            raise CapabilityNotSupportedError("vector", "sqlite-vec probe failed") from error
         if row is None or int(row[0]) != -1:
-            raise CapabilityNotSupportedError("vector", "SQLite Vec1 probe returned an invalid row")
+            raise CapabilityNotSupportedError("vector", "sqlite-vec probe returned an invalid row")
 
     async def replace(
         self,
@@ -417,7 +414,7 @@ async def search(
                 _VECTOR_SEARCH_SQL,
                 {
                     "query_vector": query_vector,
-                    "parameters": json.dumps({"k": total}, separators=(",", ":")),
+                    "neighbor_limit": total,
                     "scope_id": scope_id,
                     "memory_artifact_ids": json.dumps(
                         tuple(ref.artifact_id for ref in request.memories),
@@ -534,10 +531,10 @@ def _pack_vector(vector: tuple[float, ...]) -> bytes:
 
 def _unpack_vector(value: object, dimension: int) -> tuple[float, ...]:
     if not isinstance(value, bytes | bytearray | memoryview):
-        raise CapabilityNotSupportedError("vector", "SQLite Vec1 returned an invalid vector")
+        raise CapabilityNotSupportedError("vector", "sqlite-vec returned an invalid vector")
     packed = bytes(value)
     if len(packed) != struct.calcsize(f"={dimension}f"):
-        raise CapabilityNotSupportedError("vector", "SQLite Vec1 returned the wrong vector dimension")
+        raise CapabilityNotSupportedError("vector", "sqlite-vec returned the wrong vector dimension")
     return tuple(struct.unpack(f"={dimension}f", packed))
 
 
@@ -550,9 +547,3 @@ def _embedding_hash(profile: EmbeddingProfile, entry_hash: str) -> str:
         normalization=profile.normalization,
         entry_content_hash=entry_hash,
     )
-
-
-def _validate_vec1_info(info: object) -> None:
-    match = re.search(r"\bversion\s+(\d+)\.(\d+)\b", "" if info is None else str(info))
-    if match is None or tuple(int(part) for part in match.groups()) < _MINIMUM_VEC1_VERSION:
-        raise CapabilityNotSupportedError("vector", "SQLite Vec1 0.7 or newer is required")
diff --git a/src/powercontext/builtin/persistence/sqlite/profile.py b/src/powercontext/builtin/persistence/sqlite/profile.py
index c9cdc6be8..79c64e6ca 100644
--- a/src/powercontext/builtin/persistence/sqlite/profile.py
+++ b/src/powercontext/builtin/persistence/sqlite/profile.py
@@ -23,6 +23,7 @@
 from typing import Literal
 from weakref import WeakKeyDictionary
 
+import sqlite_vec
 from aiosqlite import Connection
 from pydantic import BaseModel, ConfigDict, Field, field_validator
 from sqlalchemy import Table, event
@@ -50,7 +51,6 @@ class SQLiteConfig(BaseModel):
     journal_mode: Literal["WAL", "DELETE", "MEMORY"] = "WAL"
     foreign_keys: bool = True
     echo: bool = False
-    vec1_extension: Path | None = None
 
     @field_validator("url")
     @classmethod
@@ -118,10 +118,8 @@ def _create_database_directory(value: str) -> None:
 def _configure_sqlite(engine: AsyncEngine, config: SQLiteConfig) -> None:
     @event.listens_for(engine.sync_engine, "connect")
     def set_pragmas(dbapi_connection: DBAPIConnection, _connection_record: object) -> None:
-        extension = config.vec1_extension
-        if extension is not None:
-            run_async = dbapi_connection.run_async
-            run_async(lambda connection: _load_extension(connection, extension))
+        run_async = dbapi_connection.run_async
+        run_async(_load_sqlite_vec)
         cursor = dbapi_connection.cursor()
         try:
             cursor.execute(f"PRAGMA busy_timeout = {config.busy_timeout_ms}")
@@ -130,10 +128,10 @@ def set_pragmas(dbapi_connection: DBAPIConnection, _connection_record: object) -
             cursor.close()
 
 
-async def _load_extension(connection: Connection, extension: Path) -> None:
+async def _load_sqlite_vec(connection: Connection) -> None:
     await connection.enable_load_extension(True)
     try:
-        await connection.load_extension(str(extension))
+        await connection.load_extension(sqlite_vec.loadable_path())
     finally:
         await connection.enable_load_extension(False)
 
diff --git a/src/powercontext/builtin/runtime/composition.py b/src/powercontext/builtin/runtime/composition.py
index 14f3b3637..acf75d7b4 100644
--- a/src/powercontext/builtin/runtime/composition.py
+++ b/src/powercontext/builtin/runtime/composition.py
@@ -53,7 +53,7 @@
 )
 from powercontext.builtin.persistence.oceanbase.profile import OceanBaseConfig, OceanBaseProfile
 from powercontext.builtin.persistence.sqlite.experience_index import SQLiteExperienceFTSIndex
-from powercontext.builtin.persistence.sqlite.memory_index import SQLiteMemoryFTSIndex, SQLiteMemoryVec1Index
+from powercontext.builtin.persistence.sqlite.memory_index import SQLiteMemoryFTSIndex, SQLiteMemoryVectorIndex
 from powercontext.builtin.persistence.sqlite.profile import SQLiteConfig, SQLiteProfile
 from powercontext.builtin.persistence.tables import BUILTIN_TABLES
 from powercontext.builtin.runtime.application import BuiltinRuntime
@@ -317,10 +317,8 @@ async def open_builtin_contexts(
     if isinstance(database, SQLiteConfig):
         experience_index = SQLiteExperienceFTSIndex()
         indexes: list[MemoryIndex] = [SQLiteMemoryFTSIndex()]
-        if database.vec1_extension is not None:
-            if embedding_model is None:
-                raise ValueError("SQLite Vec1 requires an embedding model")  # noqa: TRY003
-            indexes.append(SQLiteMemoryVec1Index(database.vec1_extension, embedding_model.profile))
+        if embedding_model is not None:
+            indexes.append(SQLiteMemoryVectorIndex(embedding_model.profile))
         index = CompositeMemoryIndex(*indexes)
         async with SQLiteProfile.open(
             database,
diff --git a/tests/builtin/persistence/test_sqlite_profile.py b/tests/builtin/persistence/test_sqlite_profile.py
index ec5247767..5c80a07e8 100644
--- a/tests/builtin/persistence/test_sqlite_profile.py
+++ b/tests/builtin/persistence/test_sqlite_profile.py
@@ -51,6 +51,7 @@ async def scenario() -> None:
         async with SQLiteProfile.open(SQLiteConfig(), tables=SHARED_TABLES) as profile:
             async with profile.database.transaction() as connection:
                 assert await connection.scalar(select(func.sqlite_version())) is not None
+                assert (await connection.exec_driver_sql("SELECT vec_version()")).scalar_one() != ""
                 pragma = await connection.exec_driver_sql("PRAGMA foreign_keys")
                 assert int(pragma.scalar() or 0) == 1
                 await connection.execute(
diff --git a/tests/e2e/real_experience_skill/harness.py b/tests/e2e/real_experience_skill/harness.py
index 5784b6c9b..009a751a4 100644
--- a/tests/e2e/real_experience_skill/harness.py
+++ b/tests/e2e/real_experience_skill/harness.py
@@ -2012,12 +2012,6 @@ def _validate_configured_settings(settings: ServerSettings) -> None:
         _fail("configured E2E requires POWERCONTEXT_SERVER_INFERENCE_GENERATION_MODEL")
     if inference.embedding_model is None:
         _fail("configured E2E requires POWERCONTEXT_SERVER_INFERENCE_EMBEDDING_MODEL")
-    if isinstance(settings.database, SQLiteConfig):
-        extension = settings.database.vec1_extension
-        if extension is None or not extension.is_file():
-            _fail("configured SQLite E2E requires an installed Vec1 extension file")
-        if not hasattr(sqlite3.Connection, "enable_load_extension"):
-            _fail("configured SQLite E2E requires a Python build with SQLite extension loading enabled")
 
 
 def _new_configured_scopes() -> ConfiguredScopes:
@@ -2288,7 +2282,7 @@ def _arguments(argv: Sequence[str] | None) -> argparse.Namespace:
     parser.add_argument(
         "--configured",
         action="store_true",
-        help="Use real generation, embedding, database, Vec1, and External Skill settings from the environment.",
+        help="Use real generation, embedding, database, and External Skill settings from the environment.",
     )
     parser.add_argument(
         "--env-file",
diff --git a/tests/e2e/test_memory_search_concurrency.py b/tests/e2e/test_memory_search_concurrency.py
index ba8328f30..940d92c3d 100644
--- a/tests/e2e/test_memory_search_concurrency.py
+++ b/tests/e2e/test_memory_search_concurrency.py
@@ -238,10 +238,4 @@ def _database_config(
             pytest.skip("set POWERCONTEXT_TEST_OCEANBASE_URL to a dedicated OceanBase MySQL-mode test database")
         return OceanBaseConfig(url=SecretStr(url))
 
-    extension = os.environ.get("POWERCONTEXT_VEC1_EXTENSION")
-    if mode in {"vector", "hybrid"} and (extension is None or not Path(extension).is_file()):
-        pytest.skip("set POWERCONTEXT_VEC1_EXTENSION to a Vec1 extension file")
-    return SQLiteConfig(
-        url=f"sqlite+aiosqlite:///{tmp_path / f'memory-search-{mode}.db'}",
-        vec1_extension=None if extension is None else Path(extension),
-    )
+    return SQLiteConfig(url=f"sqlite+aiosqlite:///{tmp_path / f'memory-search-{mode}.db'}")
diff --git a/tests/e2e/test_observability.py b/tests/e2e/test_observability.py
index 6f4dad4c3..08a8dbb9e 100644
--- a/tests/e2e/test_observability.py
+++ b/tests/e2e/test_observability.py
@@ -593,8 +593,8 @@ def test_vector_search_exports_embedding_under_memory_search_without_recording_t
         lambda _model, **_kwargs: TestEmbeddingModel(dimensions=3),
     )
     monkeypatch.setattr(
-        "powercontext.builtin.runtime.composition.SQLiteMemoryFTSIndex",
-        _VectorMemoryIndex,
+        "powercontext.builtin.runtime.composition.SQLiteMemoryVectorIndex",
+        lambda _profile: _VectorMemoryIndex(),
     )
 
     exporter = InMemorySpanExporter()
diff --git a/tests/e2e/test_runtime_server.py b/tests/e2e/test_runtime_server.py
index f0a80121f..61a6de486 100644
--- a/tests/e2e/test_runtime_server.py
+++ b/tests/e2e/test_runtime_server.py
@@ -597,13 +597,7 @@ def test_server_databases_share_vector_and_hybrid_search_behavior(
             pytest.skip("set POWERCONTEXT_TEST_OCEANBASE_URL to a dedicated OceanBase MySQL-mode test database")
         database = OceanBaseConfig(url=SecretStr(OCEANBASE_URL))
     else:
-        configured = os.environ.get("POWERCONTEXT_VEC1_EXTENSION")
-        if configured is None or not Path(configured).is_file():
-            pytest.skip("set POWERCONTEXT_VEC1_EXTENSION to a Vec1 extension file")
-        database = SQLiteConfig(
-            url=f"sqlite+aiosqlite:///{tmp_path / 'vector-runtime.db'}",
-            vec1_extension=Path(configured),
-        )
+        database = SQLiteConfig(url=f"sqlite+aiosqlite:///{tmp_path / 'vector-runtime.db'}")
     scope_id = f"vector-e2e-{uuid4()}"
     app = create_server_app(
         settings=ServerSettings(
diff --git a/tests/e2e/test_sqlite_vec1.py b/tests/e2e/test_sqlite_vec.py
similarity index 82%
rename from tests/e2e/test_sqlite_vec1.py
rename to tests/e2e/test_sqlite_vec.py
index 9e5c9ef57..2e63da4c9 100644
--- a/tests/e2e/test_sqlite_vec1.py
+++ b/tests/e2e/test_sqlite_vec.py
@@ -15,10 +15,6 @@
 from __future__ import annotations
 
 import asyncio
-import os
-from pathlib import Path
-
-import pytest
 
 from powercontext.builtin.artifacts.memory import (
     EmbeddingProfile,
@@ -53,23 +49,10 @@ def vector(text: str) -> tuple[float, float, float]:
         return EmbeddingResult(vectors=vectors)
 
 
-def _vec1_extension() -> Path:
-    configured = os.environ.get("POWERCONTEXT_VEC1_EXTENSION")
-    if configured is None:
-        pytest.skip("POWERCONTEXT_VEC1_EXTENSION is not configured")
-    extension = Path(configured)
-    if not extension.is_file():
-        pytest.skip("POWERCONTEXT_VEC1_EXTENSION does not point to a file")
-    return extension
-
-
-def test_sqlite_vec1_supports_vector_and_hybrid_search(tmp_path) -> None:
+def test_sqlite_vec_supports_vector_and_hybrid_search(tmp_path) -> None:
     async def scenario() -> None:
         model = _KeywordEmbeddingModel()
-        config = SQLiteConfig(
-            url=f"sqlite+aiosqlite:///{tmp_path / 'memory.db'}",
-            vec1_extension=_vec1_extension(),
-        )
+        config = SQLiteConfig(url=f"sqlite+aiosqlite:///{tmp_path / 'memory.db'}")
         async with open_builtin_contexts(
             BuiltinConfig(database=config),
             embedding_model=model,
diff --git a/tests/test_server.py b/tests/test_server.py
index 4f49fba1a..57bbc7396 100644
--- a/tests/test_server.py
+++ b/tests/test_server.py
@@ -55,7 +55,7 @@ class _FailingEmbeddingModel:
         model="test:embedding",
         dimension=3,
         distance="l2",
-        normalization="none",
+        normalization="unit",
     )
 
     def __init__(self, error: Exception) -> None:
@@ -131,16 +131,6 @@ def test_settings_load_server_environment(monkeypatch) -> None:
     assert settings.external_skills.codex_roots[0].path.as_posix() == "/srv/project/.agents/skills"
 
 
-def test_server_settings_vec1_preserves_file_database(tmp_path, monkeypatch) -> None:
-    data_dir = tmp_path / "powercontext-data"
-    monkeypatch.setenv("POWERCONTEXT_HOME", str(data_dir))
-    monkeypatch.setenv("POWERCONTEXT_SERVER_DATABASE_VEC1_EXTENSION", str(tmp_path / "vec1"))
-
-    settings = ServerSettings()
-
-    assert settings.database.url == f"sqlite+aiosqlite:///{data_dir / 'powercontext.db'}"
-
-
 def test_server_settings_select_oceanbase(monkeypatch) -> None:
     url = "mysql+aoceanbase://root:test@127.0.0.1:2881/powercontext?charset=utf8mb4"
     monkeypatch.setenv("POWERCONTEXT_SERVER_DATABASE_KIND", "oceanbase")
diff --git a/tests/test_server_embedding.py b/tests/test_server_embedding.py
index f57c90319..ea08a7bfc 100644
--- a/tests/test_server_embedding.py
+++ b/tests/test_server_embedding.py
@@ -65,6 +65,6 @@ def test_embedding_settings_reject_partial_profiles(values: dict[str, object]) -
         InferenceConfig.model_validate(values)
 
 
-def test_component_config_rejects_unknown_values() -> None:
-    with pytest.raises(ValidationError):
-        SQLiteConfig.model_validate({"legacy_path": "powercontext.db"})
+def test_sqlite_config_rejects_the_removed_vec1_extension_path() -> None:
+    with pytest.raises(ValidationError, match="vec1_extension"):
+        SQLiteConfig.model_validate({"vec1_extension": "/opt/sqlite-extensions/vec1"})
diff --git a/uv.lock b/uv.lock
index 70ab422b4..7a7f9bf5b 100644
--- a/uv.lock
+++ b/uv.lock
@@ -1807,6 +1807,7 @@ builtin = [
     { name = "pydantic-settings" },
     { name = "pyobvector" },
     { name = "sqlalchemy", extra = ["asyncio"] },
+    { name = "sqlite-vec" },
 ]
 cli = [
     { name = "httpx", extra = ["socks"] },
@@ -1834,6 +1835,7 @@ server = [
     { name = "pydantic-settings" },
     { name = "pyobvector" },
     { name = "sqlalchemy", extra = ["asyncio"] },
+    { name = "sqlite-vec" },
     { name = "uvicorn" },
 ]
 tracing-otlp = [
@@ -1881,6 +1883,7 @@ requires-dist = [
     { name = "pyobvector", marker = "extra == 'builtin'", specifier = ">=0.2.28,<0.3" },
     { name = "rfc8785", specifier = ">=0.1.4,<1" },
     { name = "sqlalchemy", extras = ["asyncio"], marker = "extra == 'builtin'", specifier = ">=2,<3" },
+    { name = "sqlite-vec", marker = "extra == 'builtin'", specifier = ">=0.1.9,<0.2" },
     { name = "typer", marker = "extra == 'cli'", specifier = ">=0.16,<1" },
     { name = "uvicorn", marker = "extra == 'server'", specifier = ">=0.34,<1" },
 ]
@@ -2880,6 +2883,18 @@ wheels = [
     { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f2/2f/2076eca54f6a8ed1c86301bdb4bb2ae4181b0c1c4dbc041062ca997dc1b2/sqlglot-30.13.0-py3-none-any.whl", hash = "sha256:08f87ff7b052246d61b731628c8c2db0bc91f2c9e69f5ba68a1d160a9f5b49b1", size = 719120, upload-time = "2026-07-20T20:16:53.248Z" },
 ]
 
+[[package]]
+name = "sqlite-vec"
+version = "0.1.9"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/68/85/9fad0045d8e7c8df3e0fa5a56c630e8e15ad6e5ca2e6106fceb666aa6638/sqlite_vec-0.1.9-py3-none-macosx_10_6_x86_64.whl", hash = "sha256:1b62a7f0a060d9475575d4e599bbf94a13d85af896bc1ce86ee80d1b5b48e5fb", size = 131171, upload-time = "2026-03-31T08:02:31.717Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/a4/3d/3677e0cd2f92e5ebc43cd29fbf565b75582bff1ccfa0b8327c7508e1084f/sqlite_vec-0.1.9-py3-none-macosx_11_0_arm64.whl", hash = "sha256:1d52e30513bae4cc9778ddbf6145610434081be4c3afe57cd877893bad9f6b6c", size = 165434, upload-time = "2026-03-31T08:02:32.712Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/00/d4/f2b936d3bdc38eadcbd2a87875815db36430fab0363182ba5d12cd8e0b51/sqlite_vec-0.1.9-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4e921e592f24a5f9a18f590b6ddd530eb637e2d474e3b1972f9bbeb773aa3cb9", size = 160076, upload-time = "2026-03-31T08:02:33.796Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/6f/ad/6afd073b0f817b3e03f9e37ad626ae341805891f23c74b5292818f49ac63/sqlite_vec-0.1.9-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux1_x86_64.whl", hash = "sha256:1515727990b49e79bcaf75fdee2ffc7d461f8b66905013231251f1c8938e7786", size = 163388, upload-time = "2026-03-31T08:02:34.888Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/42/89/81b2907cda14e566b9bf215e2ad82fc9b349edf07d2010756ffdb902f328/sqlite_vec-0.1.9-py3-none-win_amd64.whl", hash = "sha256:4a28dc12fa4b53d7b1dced22da2488fade444e96b5d16fd2d698cd670675cf32", size = 292804, upload-time = "2026-03-31T08:02:36.035Z" },
+]
+
 [[package]]
 name = "sse-starlette"
 version = "3.4.5"