From 2e0c6da454a92bd6ff3297a0473269dad7a8d2e9 Mon Sep 17 00:00:00 2001 From: Amaan Date: Wed, 28 Oct 2026 17:34:08 +0530 Subject: [PATCH] fix(zerogpu): drop the retired max_tokens field from the zerogpu rows Upstream #48471 retired max_tokens from the cost map in favor of max_input_tokens and max_output_tokens, and test_no_row_carries_the_retired_max_tokens_field failed on all 13 zerogpu rows. Every row already carries the same value in max_output_tokens, so nothing is lost. The zerogpu limits test now checks that each row's output limit fits its context window --- litellm/model_prices_and_context_window_backup.json | 13 ------------- model_prices_and_context_window.json | 13 ------------- tests/unit/test_model_prices_and_context_window.py | 6 +----- 3 files changed, 1 insertion(+), 31 deletions(-) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 74f3cfc42df..f1bdb8261a3 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -65283,7 +65283,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 1048576, "max_output_tokens": 65536, - "max_tokens": 65536, "mode": "chat", "output_cost_per_token": 1.2e-06, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65299,7 +65298,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 1048576, "max_output_tokens": 32768, - "max_tokens": 32768, "mode": "chat", "output_cost_per_token": 3.5e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65315,7 +65313,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 1047576, "max_output_tokens": 32768, - "max_tokens": 32768, "mode": "chat", "output_cost_per_token": 1.6e-06, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65330,7 +65327,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 272000, "max_output_tokens": 128000, - "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 1.25e-06, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65346,7 +65342,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 272000, "max_output_tokens": 128000, - "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 1.2e-06, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65362,7 +65357,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 272000, "max_output_tokens": 128000, - "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 5e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65374,7 +65368,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 128000, "max_output_tokens": 32768, - "max_tokens": 32768, "mode": "chat", "output_cost_per_token": 6e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65388,7 +65381,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 32768, "max_output_tokens": 8192, - "max_tokens": 8192, "mode": "chat", "output_cost_per_token": 5e-08, "source": "https://docs.zerogpu.ai/docs/model-catalog" @@ -65398,7 +65390,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 32768, "max_output_tokens": 8192, - "max_tokens": 8192, "mode": "chat", "output_cost_per_token": 5e-08, "source": "https://docs.zerogpu.ai/docs/model-catalog" @@ -65409,7 +65400,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 131072, "max_output_tokens": 8192, - "max_tokens": 8192, "mode": "chat", "output_cost_per_token": 2.8e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65423,7 +65413,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 131072, "max_output_tokens": 8192, - "max_tokens": 8192, "mode": "chat", "output_cost_per_token": 3e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog" @@ -65434,7 +65423,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 32768, "max_output_tokens": 16384, - "max_tokens": 16384, "mode": "chat", "output_cost_per_token": 4.5e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65449,7 +65437,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 16384, "max_output_tokens": 16384, - "max_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.5e-06, "source": "https://docs.zerogpu.ai/docs/model-catalog" diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 74f3cfc42df..f1bdb8261a3 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -65283,7 +65283,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 1048576, "max_output_tokens": 65536, - "max_tokens": 65536, "mode": "chat", "output_cost_per_token": 1.2e-06, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65299,7 +65298,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 1048576, "max_output_tokens": 32768, - "max_tokens": 32768, "mode": "chat", "output_cost_per_token": 3.5e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65315,7 +65313,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 1047576, "max_output_tokens": 32768, - "max_tokens": 32768, "mode": "chat", "output_cost_per_token": 1.6e-06, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65330,7 +65327,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 272000, "max_output_tokens": 128000, - "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 1.25e-06, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65346,7 +65342,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 272000, "max_output_tokens": 128000, - "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 1.2e-06, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65362,7 +65357,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 272000, "max_output_tokens": 128000, - "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 5e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65374,7 +65368,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 128000, "max_output_tokens": 32768, - "max_tokens": 32768, "mode": "chat", "output_cost_per_token": 6e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65388,7 +65381,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 32768, "max_output_tokens": 8192, - "max_tokens": 8192, "mode": "chat", "output_cost_per_token": 5e-08, "source": "https://docs.zerogpu.ai/docs/model-catalog" @@ -65398,7 +65390,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 32768, "max_output_tokens": 8192, - "max_tokens": 8192, "mode": "chat", "output_cost_per_token": 5e-08, "source": "https://docs.zerogpu.ai/docs/model-catalog" @@ -65409,7 +65400,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 131072, "max_output_tokens": 8192, - "max_tokens": 8192, "mode": "chat", "output_cost_per_token": 2.8e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65423,7 +65413,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 131072, "max_output_tokens": 8192, - "max_tokens": 8192, "mode": "chat", "output_cost_per_token": 3e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog" @@ -65434,7 +65423,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 32768, "max_output_tokens": 16384, - "max_tokens": 16384, "mode": "chat", "output_cost_per_token": 4.5e-07, "source": "https://docs.zerogpu.ai/docs/model-catalog", @@ -65449,7 +65437,6 @@ "litellm_provider": "zerogpu", "max_input_tokens": 16384, "max_output_tokens": 16384, - "max_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.5e-06, "source": "https://docs.zerogpu.ai/docs/model-catalog" diff --git a/tests/unit/test_model_prices_and_context_window.py b/tests/unit/test_model_prices_and_context_window.py index 568f7d1c14a..a85dda778d6 100644 --- a/tests/unit/test_model_prices_and_context_window.py +++ b/tests/unit/test_model_prices_and_context_window.py @@ -120,19 +120,15 @@ class _TokenLimits(BaseModel): max_input_tokens: int max_output_tokens: int - max_tokens: int @pytest.mark.parametrize("path", (PRICES_PATH, BACKUP_PRICES_PATH), ids=("main", "backup")) -def test_zerogpu_rows_cap_max_tokens_at_an_output_limit_that_fits_the_context_window(path: Path) -> None: +def test_zerogpu_rows_have_an_output_limit_that_fits_the_context_window(path: Path) -> None: prices: Final[dict[str, JsonValue]] = TypeAdapter(dict[str, JsonValue]).validate_json(path.read_bytes()) rows: Final = MappingProxyType( {key: _TokenLimits.model_validate(value) for key, value in prices.items() if key.startswith("zerogpu/")} ) assert rows, f"No ZeroGPU rows found in {path}" - assert {key: row.max_tokens for key, row in rows.items()} == { - key: row.max_output_tokens for key, row in rows.items() - } assert tuple(key for key, row in rows.items() if row.max_output_tokens > row.max_input_tokens) == ()