egemma2 0.3.1__py3-none-win_amd64.whl

egemma2/__init__.py

@@ -0,0 +1,22 @@
+"""EmbeddingGemma 2 in Rust: text, image, audio and video embeddings in one 768-d space, on CPU or CUDA.
+
+`pip install egemma2` installs the `egemma2` command-line binary and this Python client for it:
+
+    from egemma2 import Egemma2
+
+    with Egemma2(precision="bf16") as model:
+        vectors = model.embed(["a query", "a document"], prompt_name="Document")
+
+See https://github.com/Dalailalama/embeddinggemma2-rust for the documentation.
+"""
+
+from .client import Egemma2, Egemma2Error, find_egemma2_bin, find_model_dir
+
+try:
+    from importlib.metadata import PackageNotFoundError, version
+
+    __version__ = version("egemma2")
+except (ImportError, PackageNotFoundError):  # used from a source checkout
+    __version__ = "0+unknown"
+
+__all__ = ["Egemma2", "Egemma2Error", "find_egemma2_bin", "find_model_dir", "__version__"]

egemma2/__main__.py

@@ -0,0 +1,17 @@
+"""`python -m egemma2 ...` runs the egemma2 binary installed with this package (same as `egemma2 ...`)."""
+
+import subprocess
+import sys
+
+from .client import find_egemma2_bin
+
+
+def main():
+    try:
+        sys.exit(subprocess.call([find_egemma2_bin(), *sys.argv[1:]]))
+    except KeyboardInterrupt:
+        sys.exit(130)
+
+
+if __name__ == "__main__":
+    main()

egemma2/client.py

@@ -0,0 +1,277 @@
+"""Python client for `egemma2 serve`: the model stays loaded in one subprocess for all your requests.
+
+Standard library only; numpy and Pillow are used when you pass their objects (or call `embed_numpy`), and
+huggingface_hub, if installed, to download the model. `pip install egemma2` puts this module and the egemma2
+binary it drives into the environment; the file also works on its own next to any egemma2 binary.
+
+    from egemma2 import Egemma2
+
+    with Egemma2(precision="bf16") as model:      # CUDA if available, else the CPU
+        docs = model.embed(["first document", "second document"], prompt_name="Document")
+        query = model.embed("what is in the documents?", prompt_name="SearchQuery")[0]
+        pages = model.embed([Path("page1.png"), Path("page2.png")])    # files, by extension
+        mixed = model.embed([{"text": "a photo of", "image": pil_image}])  # one interleaved input
+
+Inputs: a `str` is text. A `pathlib.Path` is a file chosen by its extension (image, .wav audio, or a video
+file / frame directory). Images can also be Pillow images, HxWx3 uint8 numpy arrays or encoded bytes. A dict
+such as `{"text": ..., "image": ...}` is ONE input whose parts follow the dict's order (like
+sentence-transformers); `{"parts": [...]}` takes any sequence of parts. Audio given as data is
+`(samples, sample_rate)`; a video can be a list of frames, with `{"video": frames, "fps": 30}`.
+
+The server is started with `egemma2 [options] serve` and talks JSON lines over its stdin/stdout; see the
+README's "Server mode" section. For an HTTP server instead, run `egemma2 serve --http 127.0.0.1:8080` and use
+any OpenAI-compatible client.
+"""
+
+import base64
+import json
+import os
+import shutil
+import subprocess
+import sys
+import sysconfig
+import threading
+from array import array
+from pathlib import Path
+
+__all__ = ["Egemma2", "Egemma2Error", "find_egemma2_bin", "find_model_dir"]
+
+MODEL_REPO = "google/embeddinggemma-2"
+
+IMAGE_EXTS = {".png", ".jpg", ".jpeg", ".webp", ".gif", ".bmp", ".tif", ".tiff"}
+AUDIO_EXTS = {".wav", ".wave"}
+VIDEO_EXTS = {".mp4", ".mov", ".mkv", ".webm", ".avi", ".m4v"}
+
+
+class Egemma2Error(RuntimeError):
+    """The server rejected a request, or stopped."""
+
+
+def _b64(data):
+    return base64.b64encode(data).decode("ascii")
+
+
+def _image(v):
+    if isinstance(v, (str, os.PathLike)):
+        s = os.fspath(v)
+        return s if s.startswith("data:") else str(Path(s).resolve())
+    if isinstance(v, (bytes, bytearray, memoryview)):
+        return {"base64": _b64(bytes(v))}
+    if hasattr(v, "convert") and hasattr(v, "tobytes"):  # Pillow image
+        rgb = v.convert("RGB")
+        return {"rgb": _b64(rgb.tobytes()), "width": rgb.width, "height": rgb.height}
+    shape = getattr(v, "shape", None)
+    if shape is not None and len(shape) == 3 and shape[2] == 3 and str(getattr(v, "dtype", "")) == "uint8":
+        import numpy as np
+
+        return {"rgb": _b64(np.ascontiguousarray(v).tobytes()), "width": int(shape[1]), "height": int(shape[0])}
+    raise TypeError(f"unsupported image value: {type(v).__name__}")
+
+
+def _audio(v):
+    if isinstance(v, (str, os.PathLike)):
+        return str(Path(v).resolve())
+    if isinstance(v, (bytes, bytearray, memoryview)):  # an encoded .wav file
+        return {"base64": _b64(bytes(v))}
+    if isinstance(v, tuple) and len(v) == 2:  # (samples, sample_rate)
+        samples, rate = v
+        if hasattr(samples, "astype"):
+            raw = samples.astype("<f4").tobytes()
+        else:
+            a = array("f", samples)
+            if sys.byteorder == "big":
+                a.byteswap()
+            raw = a.tobytes()
+        return {"pcm_f32": _b64(raw), "sample_rate": int(rate)}
+    raise TypeError(f"unsupported audio value: {type(v).__name__} (use a path, wav bytes or (samples, sample_rate))")
+
+
+def _video(v):
+    if isinstance(v, (str, os.PathLike)):
+        return str(Path(v).resolve())
+    if isinstance(v, (list, tuple)):
+        return [_image(f) for f in v]
+    raise TypeError(f"unsupported video value: {type(v).__name__} (use a path or a list of frames)")
+
+
+def _path_item(p):
+    p = Path(p)
+    ext = p.suffix.lower()
+    if p.is_dir() or ext in VIDEO_EXTS:
+        return {"video": _video(p)}
+    if ext in AUDIO_EXTS:
+        return {"audio": _audio(p)}
+    if ext in IMAGE_EXTS:
+        return {"image": _image(p)}
+    raise ValueError(f"cannot tell what {p} is from its extension; pass {{'image': ...}}, {{'audio': ...}} or {{'video': ...}}")
+
+
+_CONVERT = {"image": _image, "image_url": _image, "audio": _audio, "video": _video}
+
+
+def _item(x):
+    if isinstance(x, str):
+        return x
+    if isinstance(x, os.PathLike):
+        return _path_item(x)
+    if isinstance(x, dict):
+        out = {}
+        for k, v in x.items():  # dicts keep insertion order, which is the parts' order
+            if k in _CONVERT:
+                out[k] = _CONVERT[k](v)
+            elif k in ("parts", "content"):
+                out[k] = [_item(p) for p in v]
+            else:
+                out[k] = v
+        return out
+    try:
+        return {"image": _image(x)}
+    except TypeError:
+        raise TypeError(f"unsupported input: {type(x).__name__}") from None
+
+
+def _floats(v):
+    if isinstance(v, str):
+        a = array("f")
+        a.frombytes(base64.b64decode(v))
+        if sys.byteorder == "big":
+            a.byteswap()
+        return a
+    return v
+
+
+def find_egemma2_bin():
+    """The egemma2 binary: $EGEMMA2_EXE, the one a `pip install egemma2` put next to this Python, or the PATH."""
+    env = os.environ.get("EGEMMA2_EXE")
+    if env:
+        return env
+    name = "egemma2" + (sysconfig.get_config_var("EXE") or (".exe" if os.name == "nt" else ""))
+    dirs = [sysconfig.get_path("scripts")]
+    user_scheme = getattr(sysconfig, "get_preferred_scheme", lambda _: "nt_user" if os.name == "nt" else "posix_user")("user")
+    try:
+        dirs.append(sysconfig.get_path("scripts", scheme=user_scheme))  # pip install --user
+    except KeyError:
+        pass
+    dirs.append(os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), "bin"))  # pip install --target
+    for d in dirs:
+        path = os.path.join(d, name)
+        if os.path.isfile(path):
+            return path
+    found = shutil.which("egemma2")
+    if found:
+        return found
+    raise FileNotFoundError("egemma2 not found: `pip install egemma2`, pass exe=..., set EGEMMA2_EXE or put it on PATH")
+
+
+def find_model_dir():
+    """The model directory: $EGEMMA2_MODEL_DIR, ./models/embeddinggemma-2, or a download of google/embeddinggemma-2
+    into the Hugging Face cache (needs huggingface_hub: `pip install "egemma2[hub]"`; about 1.5 GB, once)."""
+    env = os.environ.get("EGEMMA2_MODEL_DIR")
+    if env:
+        return env
+    local = Path("models") / "embeddinggemma-2"
+    if (local / "model.safetensors").is_file():
+        return str(local.resolve())
+    try:
+        from huggingface_hub import snapshot_download
+    except ImportError:
+        raise FileNotFoundError(
+            "no model found: pass model_dir=..., set EGEMMA2_MODEL_DIR, install huggingface_hub to download it "
+            f'automatically (pip install "egemma2[hub]"), or run: hf download {MODEL_REPO} --local-dir models/embeddinggemma-2'
+        ) from None
+    return snapshot_download(MODEL_REPO, allow_patterns=["*.json", "*.jinja", "model.safetensors"])
+
+
+class Egemma2:
+    """One `egemma2 serve` subprocess. Thread-safe: concurrent calls are sent one at a time."""
+
+    def __init__(self, exe=None, model_dir=None, device="auto", precision=None, max_tokens=None, args=(), stderr=None, cwd=None):
+        """
+        exe: path to the egemma2 binary (default: see `find_egemma2_bin`).
+        model_dir: the model directory (default: see `find_model_dir`).
+        device: auto, cpu, cuda or cuda:N. precision: None (float32), "tf32" or "bf16" (GPU only).
+        max_tokens: cut texts longer than this (default: no cut, like sentence-transformers).
+        args: extra options for `egemma2 serve` (e.g. ["--prompt-name", "Document"]).
+        stderr: where the server's log goes (default: this process's stderr).
+        """
+        exe = os.fspath(exe) if exe is not None else find_egemma2_bin()
+        # A relative path such as target/release/egemma2 is not searched for by Windows' CreateProcess.
+        exe = os.path.abspath(exe) if os.path.isfile(exe) else (shutil.which(exe) or exe)
+        model_dir = os.path.abspath(os.fspath(model_dir)) if model_dir is not None else find_model_dir()
+        cmd = [exe, "--device", device, "--model-dir", model_dir]
+        if precision in ("bf16", "tf32"):
+            cmd.append("--" + precision)
+        elif precision not in (None, "fp32", "f32", "float32"):
+            raise ValueError(f"precision must be None, 'tf32' or 'bf16' (got {precision!r})")
+        if max_tokens is not None:
+            cmd += ["--max-tokens", str(int(max_tokens))]
+        cmd += ["serve", *args]
+        self._proc = subprocess.Popen(cmd, stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=stderr, cwd=cwd)
+        self._lock = threading.Lock()
+        self._next_id = 0
+
+    def embed(self, inputs, *, prompt_name=None, prompt=None, dimensions=None):
+        """Embed one input or a list of inputs; returns one list of floats (L2-normalised) per input.
+
+        prompt_name: a sentence-transformers prompt such as "SearchQuery" or "Document" (or a literal
+        `prompt` string), placed before the whole input. dimensions: Matryoshka size (768, 512, 256, 128).
+        """
+        single = isinstance(inputs, (str, dict, os.PathLike)) or not isinstance(inputs, (list, tuple))
+        items = [_item(inputs)] if single else [_item(x) for x in inputs]
+        if not items:
+            return []
+        req = {"input": items, "encoding_format": "base64"}
+        if prompt_name is not None:
+            req["prompt_name"] = prompt_name
+        if prompt is not None:
+            req["prompt"] = prompt
+        if dimensions is not None:
+            req["dimensions"] = int(dimensions)
+        resp = self._call(req)
+        return [list(_floats(d["embedding"])) for d in sorted(resp["data"], key=lambda d: d["index"])]
+
+    def embed_numpy(self, inputs, **kwargs):
+        """Like `embed`, as a float32 numpy array of shape (n, dim)."""
+        import numpy as np
+
+        return np.asarray(self.embed(inputs, **kwargs), dtype=np.float32)
+
+    def _call(self, req):
+        with self._lock:
+            if self._proc.poll() is not None:
+                raise Egemma2Error(f"the egemma2 server has exited (code {self._proc.returncode})")
+            self._next_id += 1
+            req["id"] = self._next_id
+            try:
+                self._proc.stdin.write(json.dumps(req).encode("utf-8") + b"\n")
+                self._proc.stdin.flush()
+                line = self._proc.stdout.readline()
+            except (BrokenPipeError, OSError) as e:
+                raise Egemma2Error(f"the egemma2 server stopped: {e}") from None
+        if not line:
+            raise Egemma2Error(f"the egemma2 server stopped (exit code {self._proc.wait()}); see its log above")
+        resp = json.loads(line)
+        if "error" in resp:
+            raise Egemma2Error(resp["error"]["message"])
+        return resp
+
+    def close(self):
+        """Stop the server (it exits when its stdin closes)."""
+        if self._proc.poll() is None:
+            try:
+                self._proc.stdin.close()
+            except OSError:
+                pass
+            try:
+                self._proc.wait(timeout=30)
+            except subprocess.TimeoutExpired:
+                self._proc.kill()
+                self._proc.wait()
+        if self._proc.stdout is not None:
+            self._proc.stdout.close()
+
+    def __enter__(self):
+        return self
+
+    def __exit__(self, *exc):
+        self.close()

egemma2-0.3.1.dist-info/METADATA

@@ -0,0 +1,113 @@
+Metadata-Version: 2.4
+Name: egemma2
+Version: 0.3.1
+Classifier: Development Status :: 4 - Beta
+Classifier: Environment :: GPU :: NVIDIA CUDA
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: Operating System :: Microsoft :: Windows
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Rust
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Text Processing :: Indexing
+Requires-Dist: huggingface-hub>=0.23 ; extra == 'hub'
+Requires-Dist: numpy ; extra == 'numpy'
+Provides-Extra: hub
+Provides-Extra: numpy
+License-File: LICENSE
+License-File: NOTICE
+Summary: EmbeddingGemma 2 in Rust: text, image, audio and video embeddings on CPU or CUDA, with a Python client and an OpenAI-compatible server
+Keywords: embeddings,embeddinggemma,gemma,multimodal,semantic-search,rag,cuda,rust
+Author: DawgOG
+License-Expression: Apache-2.0
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
+Project-URL: Documentation, https://github.com/Dalailalama/embeddinggemma2-rust#server-mode-keep-the-model-loaded
+Project-URL: Funding, https://buymeacoffee.com/dawgog
+Project-URL: Homepage, https://github.com/Dalailalama/embeddinggemma2-rust
+Project-URL: Hugging Face, https://huggingface.co/DawgOG/embeddinggemma2-rust
+
+# egemma2: EmbeddingGemma 2 in Rust, for Python
+
+An unofficial, from-scratch Rust implementation of Google's
+[EmbeddingGemma 2](https://huggingface.co/google/embeddinggemma-2): text, images, audio and video
+embedded into one 768-d space (Matryoshka cuts to 512 / 256 / 128), on the CPU or an NVIDIA GPU.
+This package installs the `egemma2` binary plus a small Python client that keeps it loaded, so
+Python code gets the speed without PyTorch:
+
+| Workload (RTX A3000 laptop GPU, bf16) | PyTorch + sentence-transformers | egemma2 |
+|---|---|---|
+| short query | 52.3 ms | **2.5 ms** |
+| 2,208-token document | 136 ms | **47.6 ms** |
+| 16 images per call, per image | 99.9 ms | **51.4 ms** |
+| 9.3 s audio clip | 91.0 ms | **18.9 ms** |
+
+Embeddings match the PyTorch reference: up to 1e-6 apart in float32, and within cosine 0.99996 in
+bf16 mode. Texts are not truncated, as in sentence-transformers.
+
+## Install
+
+```bash
+pip install "egemma2[hub]"     # with huggingface_hub, to download the model on first use (~1.5 GB, once)
+pip install egemma2            # or just the binary and the client, for a model you already have
+```
+
+Wheels: Windows and Linux x64, any Python 3.8+. The GPU path needs an NVIDIA driver and the CUDA 13
+runtime libraries (cuBLAS, NVRTC) on `PATH`, for example from the CUDA Toolkit; bf16 needs compute
+capability 8.0+ (Ampere or newer). Without them egemma2 runs on the CPU.
+
+## Use
+
+```python
+from pathlib import Path
+from egemma2 import Egemma2
+
+with Egemma2(precision="bf16") as model:            # device="auto": CUDA if available, else the CPU
+    docs = model.embed(["Mars is red.", "Oceans are blue."], prompt_name="Document")
+    query = model.embed("which planet is red?", prompt_name="SearchQuery")[0]
+    scores = [sum(q * d for q, d in zip(query, doc)) for doc in docs]   # cosine similarity
+
+    pages = model.embed([Path("page1.png"), Path("page2.png")])          # files, by extension
+    mixed = model.embed([{"text": "Figure 3:", "image": pil_image}])     # one interleaved input
+    matrix = model.embed_numpy(texts, dimensions=256)                    # (n, 256) float32
+```
+
+- **Inputs:** a `str` is text, a `Path` is a file (image, `.wav` audio, video file or frame
+  directory). Images can also be Pillow images, HxWx3 uint8 numpy arrays or encoded bytes; audio can
+  be `(samples, sample_rate)`. A dict such as `{"text": ..., "image": ...}` is **one** input whose
+  parts follow the dict's order, like sentence-transformers' dict inputs.
+- **Prompts:** `prompt_name` takes the model's sentence-transformers prompts (`SearchQuery`,
+  `Document`, `QuestionAnswering`, `CodeRetrieval`, ...).
+- **Model directory:** `Egemma2(model_dir=...)`, else `$EGEMMA2_MODEL_DIR`, else
+  `./models/embeddinggemma-2`, else a download of `google/embeddinggemma-2` via huggingface_hub.
+- **Other options:** `device="cpu" | "cuda" | "cuda:N"`, `precision=None | "tf32" | "bf16"`,
+  `max_tokens=N` to cut long texts.
+
+The client starts `egemma2 serve` once and sends it JSON lines, so the model loads once; a short
+query takes 2.6 ms round trip. Requests from several threads are queued safely.
+
+## Command line and HTTP server
+
+The same install gives you the `egemma2` command (also `python -m egemma2`):
+
+```bash
+egemma2 --model-dir <dir> embed --prompt-name SearchQuery --text "which planet is red?"
+egemma2 --model-dir <dir> --bf16 serve --http 127.0.0.1:8080     # OpenAI-compatible /v1/embeddings
+```
+
+Any OpenAI-style client (the `openai` SDK, LangChain, LlamaIndex) can use the HTTP server:
+
+```python
+from openai import OpenAI
+client = OpenAI(base_url="http://127.0.0.1:8080/v1", api_key="unused")
+r = client.embeddings.create(model="embeddinggemma-2", input=["a document"], extra_body={"prompt_name": "Document"})
+```
+
+## More
+
+Documentation, benchmarks, the parity tests and the source:
+<https://github.com/Dalailalama/embeddinggemma2-rust>. Not affiliated with or endorsed by Google;
+the model weights are Google's (Apache-2.0). Code: Apache-2.0. If it saves you time, you can
+[buy me a coffee](https://buymeacoffee.com/dawgog).
+

egemma2-0.3.1.dist-info/RECORD

@@ -0,0 +1,10 @@
+egemma2/__init__.py,sha256=-d8VtBroazNGKD3zkjQcqw-6QZwr2DOFdKcPkuL7Ycs,829
+egemma2/__main__.py,sha256=5Nm17EBUcRCaAHKEAHS5lzpa94fPpg3UGpoPEoMBLFc,361
+egemma2/client.py,sha256=5FxCehhMDQ1ZNR0c_Oa2WRO0BG7_NcMyBvrFJvry2hc,11818
+egemma2-0.3.1.data/scripts/egemma2.exe,sha256=1G3NgzZb4z52P_O8tYYXu_EnutYUs4Dafdl3hg22Gk8,8970752
+egemma2-0.3.1.dist-info/METADATA,sha256=hLVipXWfaDfptV6AJWJBrxBkgU1IJ1hs80JJQfB-cB4,5475
+egemma2-0.3.1.dist-info/WHEEL,sha256=8Aej0W0a6Cz6apA3IzJrTnxLRVLAt-w0Oh8SA3Con_c,94
+egemma2-0.3.1.dist-info/licenses/LICENSE,sha256=Pd-b5cKP4n2tFDpdx27qJSIq0d1ok0oEcGTlbtL6QMU,11560
+egemma2-0.3.1.dist-info/licenses/NOTICE,sha256=DGVNMScQwtnm-xdyZISnvm4Tac1Mx81n3Pk1o6V3Rmk,734
+egemma2-0.3.1.dist-info/sboms/embeddinggemma2.cyclonedx.json,sha256=QTAjgn4KN3SwjXZ21bo-qXhWtKLDDLjdL_wdMEGFHWU,222338
+egemma2-0.3.1.dist-info/RECORD,,

egemma2-0.3.1.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: maturin (1.15.0)
+Root-Is-Purelib: false
+Tag: py3-none-win_amd64

egemma2-0.3.1.dist-info/licenses/LICENSE

@@ -0,0 +1,202 @@
+
+                                 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.

egemma2-0.3.1.dist-info/licenses/NOTICE

@@ -0,0 +1,12 @@
+egemma2: EmbeddingGemma 2 in Rust
+Copyright 2026 DawgOG (https://huggingface.co/DawgOG, https://github.com/Dalailalama)
+
+This product is licensed under the Apache License, Version 2.0 (see LICENSE).
+
+It is an unofficial, independent implementation of Google's EmbeddingGemma 2 model
+(https://huggingface.co/google/embeddinggemma-2) and is not affiliated with, sponsored by or
+endorsed by Google. The model weights are not part of this repository; Google distributes them
+under the Apache License, Version 2.0. "Gemma" and "EmbeddingGemma" are trademarks of Google LLC.
+
+The reference behaviour it reproduces is that of the Hugging Face transformers and
+sentence-transformers implementations (Apache License, Version 2.0).