local-llm-server 0.1.0__tar.gz

local_llm_server-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

local_llm_server-0.1.0/PKG-INFO

@@ -0,0 +1,68 @@
+Metadata-Version: 2.4
+Name: local-llm-server
+Version: 0.1.0
+Summary: ローカルLLM（mlx / mlx-vlm / llama.cpp / router）を OpenAI 互換 API として起動・管理する軽量サーバー
+Author: ToPo-ToPo-ToPo
+License-Expression: Apache-2.0
+Project-URL: Repository, https://github.com/ToPo-ToPo-ToPo/local-llm-server
+Keywords: llm,local-llm,mlx,mlx-vlm,llama.cpp,openai,server,inference
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Software Development
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: mlx
+Requires-Dist: mlx-lm>=0.31.3; (sys_platform == "darwin" and platform_machine == "arm64") and extra == "mlx"
+Requires-Dist: mlx-vlm>=0.5.0; (sys_platform == "darwin" and platform_machine == "arm64") and extra == "mlx"
+Dynamic: license-file
+
+# local-llm-server
+
+ローカルLLM（**mlx** / **mlx-vlm** / **llama.cpp**）を **OpenAI 互換 API** として
+起動・管理する軽量サーバー。テキストと画像（vision）を自動で振り分ける
+**router** モードを備え、任意の OpenAI 互換クライアントからそのまま利用できる。
+
+- コア機能（プロセスの起動・監視・graceful shutdown・router プロキシ）は
+  **標準ライブラリのみ**で動作。
+- 実際の推論バックエンドは extras で導入（Apple Silicon では `mlx` を自動選択）。
+
+## インストール
+
+```bash
+pip install local-llm-server          # コア（バックエンドは別途用意）
+pip install "local-llm-server[mlx]"   # Apple Silicon 向け mlx / mlx-vlm 同梱
+```
+
+## 使い方
+
+```bash
+# テキストLLMを起動（既定バックエンドは環境に応じて自動選択）
+local-llm-server --backend mlx
+
+# 画像入力対応（vision）
+local-llm-server --backend mlx-vlm
+
+# テキストLLMとVLMを同時起動し、リクエスト内容で自動振り分け
+local-llm-server --backend router
+```
+
+起動後、表示される `base_url`（例 `http://127.0.0.1:8080/v1`）を
+OpenAI 互換クライアントに設定する。
+
+```python
+import local_llm_server as srv
+
+config = srv.ServerConfig(model="mlx-community/Qwen3.6-27B-4bit", backend="mlx")
+with srv.LocalServer(config) as server:
+    server.wait_until_ready()
+    print(srv.list_models(server.base_url))
+```
+
+## ライセンス
+
+Apache-2.0

local_llm_server-0.1.0/README.md

@@ -0,0 +1,45 @@
+# local-llm-server
+
+ローカルLLM（**mlx** / **mlx-vlm** / **llama.cpp**）を **OpenAI 互換 API** として
+起動・管理する軽量サーバー。テキストと画像（vision）を自動で振り分ける
+**router** モードを備え、任意の OpenAI 互換クライアントからそのまま利用できる。
+
+- コア機能（プロセスの起動・監視・graceful shutdown・router プロキシ）は
+  **標準ライブラリのみ**で動作。
+- 実際の推論バックエンドは extras で導入（Apple Silicon では `mlx` を自動選択）。
+
+## インストール
+
+```bash
+pip install local-llm-server          # コア（バックエンドは別途用意）
+pip install "local-llm-server[mlx]"   # Apple Silicon 向け mlx / mlx-vlm 同梱
+```
+
+## 使い方
+
+```bash
+# テキストLLMを起動（既定バックエンドは環境に応じて自動選択）
+local-llm-server --backend mlx
+
+# 画像入力対応（vision）
+local-llm-server --backend mlx-vlm
+
+# テキストLLMとVLMを同時起動し、リクエスト内容で自動振り分け
+local-llm-server --backend router
+```
+
+起動後、表示される `base_url`（例 `http://127.0.0.1:8080/v1`）を
+OpenAI 互換クライアントに設定する。
+
+```python
+import local_llm_server as srv
+
+config = srv.ServerConfig(model="mlx-community/Qwen3.6-27B-4bit", backend="mlx")
+with srv.LocalServer(config) as server:
+    server.wait_until_ready()
+    print(srv.list_models(server.base_url))
+```
+
+## ライセンス
+
+Apache-2.0

local_llm_server-0.1.0/local_llm_server/__init__.py

@@ -0,0 +1,22 @@
+"""local-llm-server — ローカルLLMを OpenAI 互換 API として起動・管理する。
+
+mlx / mlx-vlm / llama.cpp、および画像とテキストを自動振り分けする router を
+サブプロセスとして起動・監視する。依存は標準ライブラリのみで、任意の
+OpenAI 互換クライアントから利用できる。
+
+公開 API:
+  - サーバー管理: LocalServer / ServerConfig / ServerPool / build_command など
+  - 起動状態の確認: is_ready / list_models / models_match / parse_host_port
+  - 既定値・定数: DEFAULT_MODEL / DEFAULT_VISION_MODEL / DEFAULT_BACKEND / BACKENDS
+  - router: RouterServer
+"""
+from __future__ import annotations
+
+from .constants import (  # noqa: F401
+    BACKENDS,
+    DEFAULT_MODEL,
+    DEFAULT_VISION_MODEL,
+    _env_bool,
+)
+from .server import *  # noqa: F401,F403
+from .router import *  # noqa: F401,F403

local_llm_server-0.1.0/local_llm_server/cli.py

@@ -0,0 +1,282 @@
+from __future__ import annotations
+
+import argparse
+import os
+import sys
+import threading
+
+from .constants import DEFAULT_MODEL, DEFAULT_VISION_MODEL, _env_bool
+from .router import RouterServer
+from .server import (
+    BACKENDS,
+    DEFAULT_BACKEND,
+    ServerConfig,
+    ServerPool,
+    build_command,
+    build_pool_configs,
+    find_pids_on_port,
+    install_shutdown_handlers,
+    parallel_supported,
+    stop_pid,
+)
+
+# CLI で選べるバックエンド。router は内部でテキスト LLM と vision VLM を
+# 同時起動し、リクエスト内容で自動振り分けする特別モード。
+ROUTER = "router"
+CLI_BACKENDS = (*BACKENDS, ROUTER)
+
+
+def _run_router(args, extra: list[str], enable_thinking: bool) -> int:
+    """router モード: テキスト LLM と vision VLM を同時起動し、プロキシで振り分ける。
+
+    公開ポート(args.port)にルーターを立て、テキストを port+1、vision を port+2 で
+    起動する。クライアントは公開ポートの 1 つだけを base_url に指定すればよい。
+    vision バックエンドは mlx-vlm（Apple Silicon 前提）、テキストは mlx を使う。
+    """
+    text_port = args.port + 1
+    vision_port = args.port + 2
+    text_cfg = ServerConfig(
+        backend="mlx",
+        model=args.model,
+        host=args.host,
+        port=text_port,
+        disable_thinking=not enable_thinking,
+        extra_args=extra,
+    )
+    vision_cfg = ServerConfig(
+        backend="mlx-vlm",
+        model=args.vision_model,
+        host=args.host,
+        port=vision_port,
+    )
+    pool = ServerPool([text_cfg, vision_cfg])
+    print(f"Starting (text):   {' '.join(build_command(text_cfg))}", file=sys.stderr)
+    print(f"Starting (vision): {' '.join(build_command(vision_cfg))}", file=sys.stderr)
+    try:
+        pool.start()
+        pool.wait_until_ready()
+    except (RuntimeError, TimeoutError) as exc:
+        print(f"Error: {exc}", file=sys.stderr)
+        pool.stop()
+        return 1
+
+    router = RouterServer(
+        (args.host, args.port),
+        (args.host, text_port),
+        (args.host, vision_port),
+    )
+    public_url = f"http://{args.host}:{args.port}/v1"
+    print("Ready (automatic routing):", file=sys.stderr)
+    print(f"  - public: {public_url}", file=sys.stderr)
+    print(f"      text -> http://{args.host}:{text_port}/v1", file=sys.stderr)
+    print(f"      image/media -> http://{args.host}:{vision_port}/v1", file=sys.stderr)
+    print(
+        f'In agent.toml, set only the single base_url = "{public_url}". '
+        "Text-only requests are routed to the LLM, requests with images to the VLM, automatically.",
+        file=sys.stderr,
+    )
+    print(
+        "Note: if the VLM does not support function calling (tools), "
+        'set tool_mode = "prompt" when using images.',
+        file=sys.stderr,
+    )
+
+    thread = threading.Thread(target=router.serve_forever, daemon=True)
+    thread.start()
+    try:
+        pool.wait()  # 子サーバーが終了するまでブロック
+    except KeyboardInterrupt:
+        pass
+    finally:
+        router.shutdown()
+        router.server_close()
+        pool.stop()
+    return 0
+
+
+def _stop_servers(ports: list[int]) -> int:
+    """指定ポートで動いているローカルサーバーを探して停止する（--stop 用）。
+
+    各ポートを LISTEN しているプロセスを lsof で特定し、プロセスグループごと
+    SIGTERM→SIGKILL で止める。1つでも止めれば 0、見つからなければ 1 を返す。
+    """
+    if os.name != "posix":
+        print(
+            "--stop is supported on macOS / Linux only. On Windows, stop the server "
+            "from its own window (Ctrl+C) or via Task Manager.",
+            file=sys.stderr,
+        )
+        return 1
+    stopped = False
+    for port in ports:
+        pids = find_pids_on_port(port)
+        for pid in pids:
+            print(f"Stopping the server on port {port} (pid {pid})...", file=sys.stderr)
+            if stop_pid(pid):
+                stopped = True
+    if not stopped:
+        ports_str = ", ".join(str(p) for p in ports)
+        print(f"No running server found on port(s): {ports_str}.", file=sys.stderr)
+        return 1
+    print("Stopped.", file=sys.stderr)
+    return 0
+
+
+def main(argv: list[str] | None = None) -> int:
+    parser = argparse.ArgumentParser(
+        prog="local-llm-server",
+        description=(
+            "Start a local LLM server (mlx / mlx-vlm / llama.cpp / router). "
+            "Use the vision-capable mlx-vlm for image input (--backend mlx-vlm). "
+            "With --backend router, start a text LLM and a VLM together and route requests automatically based on their content."
+        ),
+        epilog="Backend-specific extra arguments can be passed after --.",
+    )
+    parser.add_argument(
+        "--backend",
+        choices=CLI_BACKENDS,
+        default=os.environ.get("CODER_BACKEND", DEFAULT_BACKEND),
+        help="Backend to use (env: CODER_BACKEND, default: %(default)s)",
+    )
+    parser.add_argument(
+        "--model",
+        default=os.environ.get("CODER_MODEL", DEFAULT_MODEL),
+        help="Model (mlx: name/path, llama-cpp: .gguf path; used as the text model in router mode). env: CODER_MODEL, default: %(default)s",
+    )
+    parser.add_argument(
+        "--vision-model",
+        default=os.environ.get("CODER_VISION_MODEL", DEFAULT_VISION_MODEL),
+        help="Vision model that handles images/media in router mode. env: CODER_VISION_MODEL, default: %(default)s",
+    )
+    parser.add_argument("--host", default="127.0.0.1", help="Host to bind to")
+    parser.add_argument("--port", type=int, default=8080, help="Port to bind to (the first port when using a pool)")
+    parser.add_argument(
+        "--stop",
+        action="store_true",
+        help="Stop a server already running on --port instead of starting one "
+        "(use the same --port / --instances / --backend router as when you started it). macOS / Linux only.",
+    )
+    parser.add_argument(
+        "--parallel",
+        type=int,
+        help="Number of concurrent processing slots (llama.cpp only; mlx processes sequentially)",
+    )
+    parser.add_argument(
+        "--instances",
+        type=int,
+        default=1,
+        help="Number of servers to start on consecutive ports (to gain parallelism with mlx; each loads its own model)",
+    )
+    parser.add_argument(
+        "--thinking",
+        action=argparse.BooleanOptionalAction,
+        default=None,
+        help="Thinking mode. Use --no-thinking to disable (env: CODER_ENABLE_THINKING)",
+    )
+    parser.add_argument(
+        "--draft-model",
+        default=os.environ.get("CODER_DRAFT_MODEL"),
+        help="MTP drafter for Gemma 4 (HF id/path, e.g. "
+        "mlx-community/gemma-4-E4B-it-qat-assistant-bf16). Use 'auto' to select automatically from the main model name. "
+        "Speeds up inference without changing the main model's output. mlx-vlm backend only. env: CODER_DRAFT_MODEL",
+    )
+    args, extra = parser.parse_known_args(argv)
+
+    if args.instances < 1:
+        parser.error("--instances must be 1 or greater")
+
+    # --stop: 起動の代わりに、起動時と同じポート割り当てを再現して停止する。
+    # router は public(port) / text(port+1) / vision(port+2)、通常は連番ポート。
+    if args.stop:
+        if args.backend == ROUTER:
+            ports = [args.port, args.port + 1, args.port + 2]
+        else:
+            ports = [args.port + i for i in range(args.instances)]
+        return _stop_servers(ports)
+
+    # kill / ターミナルを閉じる（SIGTERM・SIGHUP）でも下流の finally（pool.stop）を
+    # 必ず通し、起動したバックエンドを孫プロセスとして残さない。
+    install_shutdown_handlers()
+
+    if args.parallel is not None:
+        if args.parallel < 1:
+            parser.error("--parallel must be 1 or greater")
+        if not parallel_supported(args.backend):
+            print(
+                f"Warning: {args.backend} does not support parallel slots "
+                "(sequential processing). --parallel is ignored.",
+                file=sys.stderr,
+            )
+
+    if args.backend == "mlx-vlm":
+        print(
+            "Info: starting with the vision backend (mlx-vlm). It supports image input (image_url), "
+            "but some models do not support function calling (tools)/streaming. "
+            'In that case, set tool_mode = "prompt" in agent.toml.',
+            file=sys.stderr,
+        )
+
+    enable_thinking = (
+        args.thinking if args.thinking is not None
+        else _env_bool("CODER_ENABLE_THINKING", False)
+    )
+
+    # argparse が拾った先頭の "--" を除去してバックエンドへ素通し
+    if extra and extra[0] == "--":
+        extra = extra[1:]
+
+    if args.backend == ROUTER:
+        if args.instances != 1:
+            parser.error("--instances cannot be used with --backend router")
+        return _run_router(args, extra, enable_thinking)
+
+    base = ServerConfig(
+        backend=args.backend,
+        model=args.model,
+        host=args.host,
+        port=args.port,
+        parallel=args.parallel,
+        disable_thinking=not enable_thinking,
+        draft_model=args.draft_model,
+        extra_args=extra,
+    )
+    configs = build_pool_configs(base, args.instances)
+    pool = ServerPool(configs)
+
+    for config in configs:
+        print(f"Starting: {' '.join(build_command(config))}", file=sys.stderr)
+    try:
+        pool.start()
+        pool.wait_until_ready()
+    except (RuntimeError, TimeoutError) as exc:
+        print(f"Error: {exc}", file=sys.stderr)
+        pool.stop()
+        return 1
+
+    print("Ready:", file=sys.stderr)
+    for url in pool.base_urls:
+        print(f"  - {url}", file=sys.stderr)
+    if args.instances > 1:
+        print(
+            "Point each OpenAI-compatible client at one of the base URLs above. "
+            f'Example: base_url = "{pool.base_urls[0]}"',
+            file=sys.stderr,
+        )
+    else:
+        print(
+            "Point your OpenAI-compatible client at "
+            f'base_url = "{pool.base_urls[0]}".',
+            file=sys.stderr,
+        )
+
+    try:
+        pool.wait()
+    except KeyboardInterrupt:
+        pass
+    finally:
+        pool.stop()
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

local_llm_server-0.1.0/local_llm_server/constants.py

@@ -0,0 +1,26 @@
+"""エージェント側（agent/）とサーバー側（server/）で共有する中立な定数・ヘルパー。
+
+どちらのサブパッケージにも依存しないため、ここを基点に server/ と agent/ を
+完全に分離できる（server/ が agent/ を import する逆向き依存を作らないための層）。
+"""
+from __future__ import annotations
+
+import os
+
+# サーバー・モデルが明示されないときに使う既定モデル
+DEFAULT_MODEL = "mlx-community/Qwen3.6-27B-4bit"
+
+# 画像・メディアを処理する vision モデルの既定。
+# 既定モデル(Qwen3.6)はマルチモーダルなので、テキストと共通のものを使う。
+DEFAULT_VISION_MODEL = DEFAULT_MODEL
+
+# 起動可能なローカルLLMサーバーのバックエンド一覧。
+# server/ が実装し、agent/ は agent.toml の backend 検証に使う（共有値）。
+BACKENDS = ("mlx", "mlx-vlm", "llama-cpp")
+
+
+def _env_bool(name: str, default: bool) -> bool:
+    value = os.environ.get(name)
+    if value is None:
+        return default
+    return value.strip().lower() in ("1", "true", "yes", "on")