TimeCopilot · AzulGarza · Aug 17, 2026 · Aug 14, 2026 · Aug 14, 2026 · Aug 14, 2026
diff --git a/docs/api/gift-eval/gift-eval.md b/docs/api/gift-eval/gift-eval.md
@@ -1,11 +1,34 @@
-# `timecopilot.gift_eval`
+# `timecopilot_gift_eval`
 
-::: timecopilot.gift_eval.eval
+GIFT-Eval integration lives in the standalone [`timecopilot-gift-eval`](https://pypi.org/project/timecopilot-gift-eval/) package on PyPI. Install it with TimeCopilot via:
+
+```bash
+pip install "timecopilot[gift-eval]"
+```
+
+## Migration from `timecopilot.gift_eval`
+
+The in-repo module `timecopilot.gift_eval` was removed in favor of the standalone package. Update imports as follows:
+
+| Before | After |
+|--------|-------|
+| `from timecopilot.gift_eval.eval import GIFTEval` | `from timecopilot_gift_eval import GIFTEval` |
+| `from timecopilot.gift_eval.gluonts_predictor import GluonTSPredictor` | `from timecopilot_gift_eval import GluonTSPredictor` |
+| `from timecopilot.gift_eval.utils import DATASETS_WITH_TERMS` | `from timecopilot_gift_eval.utils import DATASETS_WITH_TERMS` |
+
+No compatibility shim is provided — install `timecopilot-gift-eval` (or `timecopilot[gift-eval]`) and switch to the `timecopilot_gift_eval` import path.
+
+::: timecopilot_gift_eval.eval
     options:
         members:
             - GIFTEval
 
-::: timecopilot.gift_eval.gluonts_predictor
+::: timecopilot_gift_eval.gluonts_predictor
+    options:
+        members:
+            - GluonTSPredictor
+
+::: timecopilot_gift_eval.protocol
     options:
         members:
-            - GluonTSPredictor
+            - ForecasterProtocol
diff --git a/docs/changelogs/index.md b/docs/changelogs/index.md
@@ -2,6 +2,7 @@
 
 Welcome to the TimeCopilot Changelog. Here, you will find a comprehensive list of all the changes, updates, and improvements made to the TimeCopilot project. This section is designed to keep you informed about the latest features, bug fixes, and enhancements as we continue to develop and refine the TimeCopilot experience. Stay tuned for regular updates and feel free to explore the details of each release below.
 
+- [v0.0.31](v0.0.31.md)
 - [v0.0.30](v0.0.30.md)
 - [v0.0.29](v0.0.29.md)
 - [v0.0.28](v0.0.28.md)

diff --git a/docs/changelogs/v0.0.31.md b/docs/changelogs/v0.0.31.md
@@ -0,0 +1,22 @@
+### Breaking Changes
+
+* **GIFT-Eval moved to standalone package**: The in-repo `timecopilot.gift_eval` module was removed. GIFT-Eval integration now lives in the [`timecopilot-gift-eval`](https://pypi.org/project/timecopilot-gift-eval/) PyPI package (`timecopilot_gift_eval` import path). Install via `pip install "timecopilot[gift-eval]"` or `pip install timecopilot-gift-eval>=0.3.0`.
+
+    **Migration:**
+
+    ```text
+    from timecopilot.gift_eval.eval import GIFTEval
+    from timecopilot.gift_eval.gluonts_predictor import GluonTSPredictor
+    from timecopilot.gift_eval.utils import DATASETS_WITH_TERMS
+    ```
+
+    ```python
+    from timecopilot_gift_eval import GIFTEval, GluonTSPredictor
+    from timecopilot_gift_eval.utils import DATASETS_WITH_TERMS
+    ```
+
+    See the [GIFT-Eval API docs](../api/gift-eval/gift-eval.md) for details.
+
+---
+
+**Full Changelog**: https://github.com/TimeCopilot/timecopilot/compare/v0.0.30...v0.0.31
diff --git a/docs/examples/gift-eval.ipynb b/docs/examples/gift-eval.ipynb
@@ -53,7 +53,7 @@
    },
    "outputs": [],
    "source": [
-    "%pip install \"timecopilot>=0.0.22\""
+    "%pip install \"timecopilot[gift-eval]>=0.0.22\""
    ]
   },
   {
@@ -64,7 +64,7 @@
    "source": [
     "## Dataset Setup\n",
     "\n",
-    "TimeCopilot includes built-in [GIFT-Eval integration](https://timecopilot.dev/api/gift-eval/gift-eval/#timecopilot.gift_eval.eval.GIFTEval) for dataset handling:"
+    "TimeCopilot includes built-in [GIFT-Eval integration](https://timecopilot.dev/api/gift-eval/gift-eval/#timecopilot_gift_eval.eval.GIFTEval) for dataset handling:"
    ]
   },
   {
@@ -76,7 +76,7 @@
    },
    "outputs": [],
    "source": [
-    "from timecopilot.gift_eval.eval import GIFTEval\n",
+    "from timecopilot_gift_eval import GIFTEval\n",
     "\n",
     "# TimeCopilot's built-in GIFT-Eval dataset downloader\n",
     "# Handles the complete benchmark dataset with all 97 configurations\n",
@@ -156,7 +156,7 @@
     "id": "M2YcjoDF5NH7"
    },
    "source": [
-    "With TimeCopilot you can evaluate any [Forecaster](https://timecopilot.dev/api/models/utils/forecaster/#timecopilot.models.utils.forecaster.Forecaster) in a standardized way using its [GIFT-Eval](https://timecopilot.dev/api/gift-eval/gift-eval/#timecopilot.gift_eval.eval.GIFTEval) integration."
+    "With TimeCopilot you can evaluate any [Forecaster](https://timecopilot.dev/api/models/utils/forecaster/#timecopilot.models.utils.forecaster.Forecaster) in a standardized way using its [GIFT-Eval](https://timecopilot.dev/api/gift-eval/gift-eval/#timecopilot_gift_eval.eval.GIFTEval) integration."
    ]
   },
   {
@@ -168,8 +168,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from timecopilot.gift_eval.eval import GIFTEval\n",
-    "from timecopilot.gift_eval.gluonts_predictor import GluonTSPredictor\n",
+    "from timecopilot_gift_eval import GIFTEval, GluonTSPredictor\n",
     "\n",
     "\n",
     "def evaluate_forecaster(\n",
@@ -710,7 +709,7 @@
    },
    "outputs": [],
    "source": [
-    "from timecopilot.gift_eval.utils import DATASETS_WITH_TERMS"
+    "from timecopilot_gift_eval.utils import DATASETS_WITH_TERMS"
    ]
   },
   {
@@ -789,7 +788,7 @@
     "id": "g0-oVisu6XX3"
    },
    "source": [
-    "The TimeCopilot's [GIFT-Eval integration](https://timecopilot.dev/api/gift-eval/gift-eval/#timecopilot.gift_eval.eval.GIFTEval) was designed considering reproducibility as one of its main features. The library can replicate the official results provided by the mantainers of the benchmark for the [`SeasonalNaive`](https://huggingface.co/spaces/Salesforce/GIFT-Eval/tree/main/results/seasonal_naive) method. The following code replicates the Seasonal Naive performance for the datasets evaluated in this notebook. The reproducibility of the results for the rest of the datasets are tested continuously in the [library's repo](https://github.com/TimeCopilot/timecopilot/blob/main/tests/gift_eval/test_evaluation.py)."
+    "The TimeCopilot's [GIFT-Eval integration](https://timecopilot.dev/api/gift-eval/gift-eval/#timecopilot_gift_eval.eval.GIFTEval) was designed considering reproducibility as one of its main features. The library can replicate the official results provided by the mantainers of the benchmark for the [`SeasonalNaive`](https://huggingface.co/spaces/Salesforce/GIFT-Eval/tree/main/results/seasonal_naive) method. The following code replicates the Seasonal Naive performance for the datasets evaluated in this notebook. The reproducibility of the results for the rest of the datasets are tested continuously in the [library's repo](https://github.com/TimeCopilot/timecopilot/blob/main/tests/gift_eval/test_evaluation.py)."
    ]
   },
   {

diff --git a/experiments/gift-eval/pyproject.toml b/experiments/gift-eval/pyproject.toml
@@ -2,13 +2,14 @@
 dependencies = [
   "modal>=1.0.5",
   "s3fs>=2023.12.1",
+  "timecopilot-gift-eval>=0.3.0",
   "timecopilot>=0.0.21",
   "transformers<4.54",
   "transformers==4.40.1 ; python_full_version < '3.12'",
   "typer>=0.16.0",
 ]
 description = "TimeCopilot experiments for GIFT-Eval"
-name = "timecopilot-gift-eval"
+name = "timecopilot-gift-eval-experiments"
 readme = "README.md"
 requires-python = ">=3.11"
 version = "0.2.0"
diff --git a/experiments/gift-eval/src/download_results.py b/experiments/gift-eval/src/download_results.py
@@ -2,8 +2,7 @@
 from pathlib import Path
 
 import pandas as pd
-
-from timecopilot.gift_eval.utils import DATASETS_WITH_TERMS
+from timecopilot_gift_eval.utils import DATASETS_WITH_TERMS
 
 logging.basicConfig(level=logging.INFO)
 

diff --git a/experiments/gift-eval/src/run_modal.py b/experiments/gift-eval/src/run_modal.py
@@ -59,8 +59,7 @@ def main():
     import logging
 
     import fsspec
-
-    from timecopilot.gift_eval.utils import DATASETS_WITH_TERMS
+    from timecopilot_gift_eval.utils import DATASETS_WITH_TERMS
 
     logging.basicConfig(level=logging.INFO)
 

diff --git a/experiments/gift-eval/src/run_timecopilot.py b/experiments/gift-eval/src/run_timecopilot.py
@@ -2,9 +2,8 @@
 from typing import Annotated
 
 import typer
+from timecopilot_gift_eval import GIFTEval, GluonTSPredictor
 
-from timecopilot.gift_eval.eval import GIFTEval
-from timecopilot.gift_eval.gluonts_predictor import GluonTSPredictor
 from timecopilot.models.ensembles.median import MedianEnsemble
 from timecopilot.models.foundation.chronos import Chronos
 from timecopilot.models.foundation.timesfm import TimesFM