sync from b7516

This commit is contained in:
2026-01-16 11:16:14 +08:00
parent f4ae4cc7da
commit 6ee41dd9e3
380 changed files with 18435 additions and 38806 deletions

View File

@@ -393,12 +393,12 @@ def test_completion_unified(n_ctx, n_slots, n_predict_vals, expected_success):
for res, n_predict, expect_ok in zip(results, n_predict_vals, expected_success):
if expect_ok:
assert res.status_code == 200
# note: https://github.com/ggml-org/llama.cpp/pull/18700#issuecomment-3728695581
if res.status_code == 200:
assert "content" in res.body
if "timings" in res.body:
assert res.body["timings"]["predicted_n"] == n_predict
else:
assert res.status_code == 500
assert "content" not in res.body
@pytest.mark.parametrize(