diff --git a/packages/gooddata-eval/src/gooddata_eval/core/reporting/html_report.py b/packages/gooddata-eval/src/gooddata_eval/core/reporting/html_report.py
index 10db5cec8..b943cb369 100644
--- a/packages/gooddata-eval/src/gooddata_eval/core/reporting/html_report.py
+++ b/packages/gooddata-eval/src/gooddata_eval/core/reporting/html_report.py
@@ -74,6 +74,20 @@ def _model_of(label: str, run: dict) -> str:
     return run.get("model") or (label.split(" · ", 1)[1] if " · " in label else label)
 
 
+def _alias_name(n: int) -> str:
+    """``Model A`` .. ``Model Z``, then ``Model AA``, ``Model AB``, ...
+
+    Spelled out rather than ``chr(65 + n)``, which walks straight off the end of the alphabet: the
+    27th model came out as ``Model [``, then ``Model \\``, ``Model ]``.
+    """
+    letters = ""
+    n += 1
+    while n:
+        n, remainder = divmod(n - 1, 26)
+        letters = chr(65 + remainder) + letters
+    return f"Model {letters}"
+
+
 def _aliases(doc: dict) -> dict[str, tuple[str, str]]:
     """Map each run label to its redacted ``(run label, model name)``.
 
@@ -92,7 +106,7 @@ def _aliases(doc: dict) -> dict[str, tuple[str, str]]:
     }
     letters: dict[tuple[str, str], str] = {}
     for model in identity.values():
-        letters.setdefault(model, f"Model {chr(65 + len(letters))}")
+        letters.setdefault(model, _alias_name(len(letters)))
     per_model = Counter(identity.values())
     seen: Counter[tuple[str, str]] = Counter()
     aliases: dict[str, tuple[str, str]] = {}
diff --git a/packages/gooddata-eval/tests/test_html_report.py b/packages/gooddata-eval/tests/test_html_report.py
index f78b99303..d90413cdd 100644
--- a/packages/gooddata-eval/tests/test_html_report.py
+++ b/packages/gooddata-eval/tests/test_html_report.py
@@ -7,7 +7,7 @@
 import orjson
 import pytest
 from gooddata_eval.cli.main import main
-from gooddata_eval.core.reporting.html_report import build_html, load_report_files, merge_docs
+from gooddata_eval.core.reporting.html_report import _aliases, build_html, load_report_files, merge_docs
 
 
 def _doc(model: str, passed: bool) -> dict:
@@ -129,6 +129,17 @@ def test_redact_gives_one_model_one_alias_across_its_runs() -> None:
     assert {entry["provider_name"] for entry in data["comparison"].values()} == {""}
 
 
+def test_redact_alias_survives_past_twenty_six_models() -> None:
+    """chr(65 + n) walked off the alphabet: the 27th model was `Model [`."""
+    doc = {"runs": {f"m{i}": {"model": f"model-{i}", "items": {}} for i in range(30)}, "comparison": {}}
+
+    names = [name.removeprefix("Model ") for _, name in _aliases(doc).values()]
+
+    assert len(set(names)) == 30
+    assert all(name.isalpha() for name in names), names
+    assert names[25:28] == ["Z", "AA", "AB"]
+
+
 def test_redact_gives_each_model_its_own_letter_and_numbers_only_a_repeated_model() -> None:
     merged = merge_docs([("a", _doc("gpt-5", True)), ("b", _doc("claude", True)), ("c", _doc("gpt-5", False))])