scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-11-11 08:38:24 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 359

Correct superbot's output limit to the router's published ceiling

Browse source 
The cost-map rows moved to the router's live ceiling, 131072, so the test
that asserts them moves with it: max_output_tokens and max_tokens are both
131072 now. The first push of this change left the test at 65_536, which is
what llms-providers caught.

Upstream tooling is not re-run here; this PR's CI runs it.
This commit is contained in:
superbot 2026-10-20 00:50:39 -06:00
parent 67a0bcd4bf
commit eb6dcb38f6
1 changed files with 2 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
tests/unit/llms/openai_like/test_superbot_provider.py
View file

@ -156,7 +156,8 @@ def test_superbot_models_are_priced_in_both_cost_maps(model_id: str, context: in
assert row["litellm_provider"] == "superbot-gg"
assert row["mode"] == "chat"
assert row["max_input_tokens"] == context
assert row["max_output_tokens"] == 65_536
assert row["max_output_tokens"] == 131_072
assert row["max_tokens"] == 131_072
assert row["input_cost_per_token"] == pytest.approx(1.0e-6)
assert row["output_cost_per_token"] == pytest.approx(5.0e-6)
assert row["cache_read_input_token_cost"] == pytest.approx(5.0e-8)