fix(pricing): add undated azure aliases for gpt-audio-mini and gpt-realtime-mini

litellm/model_prices_and_context_window_backup.json

@@ -4867,6 +4867,38 @@
         "supports_tool_choice": true,
         "supports_vision": false
     },
+    "azure/gpt-audio-mini": {
+        "deprecation_date": "2027-04-06",
+        "input_cost_per_audio_token": 1e-05,
+        "input_cost_per_token": 6e-07,
+        "litellm_provider": "azure",
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "max_tokens": 16384,
+        "mode": "chat",
+        "output_cost_per_audio_token": 2e-05,
+        "output_cost_per_token": 2.4e-06,
+        "supported_endpoints": [
+            "/v1/chat/completions"
+        ],
+        "supported_modalities": [
+            "text",
+            "audio"
+        ],
+        "supported_output_modalities": [
+            "text",
+            "audio"
+        ],
+        "supports_function_calling": true,
+        "supports_native_streaming": true,
+        "supports_parallel_function_calling": true,
+        "supports_prompt_caching": false,
+        "supports_reasoning": false,
+        "supports_response_schema": false,
+        "supports_system_messages": true,
+        "supports_tool_choice": true,
+        "supports_vision": false
+    },
     "azure/gpt-audio-mini-2025-10-06": {
         "deprecation_date": "2027-04-06",
         "input_cost_per_audio_token": 1e-05,
@@ -5080,6 +5112,38 @@
         "supports_system_messages": true,
         "supports_tool_choice": true
     },
+    "azure/gpt-realtime-mini": {
+        "cache_creation_input_audio_token_cost": 3e-07,
+        "cache_read_input_token_cost": 6e-08,
+        "input_cost_per_audio_token": 1e-05,
+        "input_cost_per_image": 8e-07,
+        "input_cost_per_token": 6e-07,
+        "litellm_provider": "azure",
+        "max_input_tokens": 32000,
+        "max_output_tokens": 4096,
+        "max_tokens": 4096,
+        "mode": "realtime",
+        "output_cost_per_audio_token": 2e-05,
+        "output_cost_per_token": 2.4e-06,
+        "supported_endpoints": [
+            "/v1/realtime"
+        ],
+        "supported_modalities": [
+            "text",
+            "image",
+            "audio"
+        ],
+        "supported_output_modalities": [
+            "text",
+            "audio"
+        ],
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
     "azure/gpt-realtime-mini-2025-10-06": {
         "cache_creation_input_audio_token_cost": 3e-07,
         "cache_read_input_token_cost": 6e-08,

model_prices_and_context_window.json

@@ -4867,6 +4867,38 @@
         "supports_tool_choice": true,
         "supports_vision": false
     },
+    "azure/gpt-audio-mini": {
+        "deprecation_date": "2027-04-06",
+        "input_cost_per_audio_token": 1e-05,
+        "input_cost_per_token": 6e-07,
+        "litellm_provider": "azure",
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "max_tokens": 16384,
+        "mode": "chat",
+        "output_cost_per_audio_token": 2e-05,
+        "output_cost_per_token": 2.4e-06,
+        "supported_endpoints": [
+            "/v1/chat/completions"
+        ],
+        "supported_modalities": [
+            "text",
+            "audio"
+        ],
+        "supported_output_modalities": [
+            "text",
+            "audio"
+        ],
+        "supports_function_calling": true,
+        "supports_native_streaming": true,
+        "supports_parallel_function_calling": true,
+        "supports_prompt_caching": false,
+        "supports_reasoning": false,
+        "supports_response_schema": false,
+        "supports_system_messages": true,
+        "supports_tool_choice": true,
+        "supports_vision": false
+    },
     "azure/gpt-audio-mini-2025-10-06": {
         "deprecation_date": "2027-04-06",
         "input_cost_per_audio_token": 1e-05,
@@ -5080,6 +5112,38 @@
         "supports_system_messages": true,
         "supports_tool_choice": true
     },
+    "azure/gpt-realtime-mini": {
+        "cache_creation_input_audio_token_cost": 3e-07,
+        "cache_read_input_token_cost": 6e-08,
+        "input_cost_per_audio_token": 1e-05,
+        "input_cost_per_image": 8e-07,
+        "input_cost_per_token": 6e-07,
+        "litellm_provider": "azure",
+        "max_input_tokens": 32000,
+        "max_output_tokens": 4096,
+        "max_tokens": 4096,
+        "mode": "realtime",
+        "output_cost_per_audio_token": 2e-05,
+        "output_cost_per_token": 2.4e-06,
+        "supported_endpoints": [
+            "/v1/realtime"
+        ],
+        "supported_modalities": [
+            "text",
+            "image",
+            "audio"
+        ],
+        "supported_output_modalities": [
+            "text",
+            "audio"
+        ],
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
     "azure/gpt-realtime-mini-2025-10-06": {
         "cache_creation_input_audio_token_cost": 3e-07,
         "cache_read_input_token_cost": 6e-08,

tests/test_litellm/test_azure_audio_price_aliases.py

@@ -0,0 +1,67 @@
+"""Undated azure aliases for the audio models must exist and match their dated
+variants — Azure deployments are commonly created against the undated model
+name, and `base_model: azure/gpt-audio-mini` previously resolved to nothing
+(text tokens billed at $0). Issue #33170."""
+
+import pytest
+
+import litellm
+
+
+@pytest.fixture(autouse=True)
+def _use_local_model_cost_map(monkeypatch):
+    original_model_cost = litellm.model_cost
+    monkeypatch.setenv("LITELLM_LOCAL_MODEL_COST_MAP", "True")
+    litellm.model_cost = litellm.get_model_cost_map(url="")
+    try:
+        yield
+    finally:
+        litellm.model_cost = original_model_cost
+
+
+COST_FIELDS = (
+    "input_cost_per_token",
+    "output_cost_per_token",
+    "input_cost_per_audio_token",
+    "output_cost_per_audio_token",
+)
+
+
+@pytest.mark.parametrize(
+    "undated, dated",
+    [
+        ("azure/gpt-audio-mini", "azure/gpt-audio-mini-2025-10-06"),
+        ("azure/gpt-realtime-mini", "azure/gpt-realtime-mini-2025-10-06"),
+    ],
+)
+def test_undated_azure_audio_alias_matches_dated_entry(undated, dated):
+    undated_info = litellm.get_model_info(undated)
+    dated_info = litellm.get_model_info(dated)
+
+    for field in COST_FIELDS:
+        assert undated_info.get(field) == dated_info.get(field), field
+        # the production symptom was text tokens billed at $0 — make sure the
+        # alias carries real, non-zero prices
+        assert (undated_info.get(field) or 0) > 0, f"{undated}.{field} must be non-zero"
+
+    assert undated_info.get("litellm_provider") == "azure"
+    assert undated_info.get("mode") == dated_info.get("mode")
+
+
+@pytest.mark.parametrize(
+    "undated, dated",
+    [
+        ("azure/gpt-audio-mini", "azure/gpt-audio-mini-2025-10-06"),
+        ("azure/gpt-realtime-mini", "azure/gpt-realtime-mini-2025-10-06"),
+    ],
+)
+def test_undated_azure_audio_alias_is_exact_mirror(undated, dated):
+    """The undated alias must be a byte-for-byte mirror of its dated entry —
+    covers every field (incl. realtime-specific cache/audio cost keys) so any
+    future drift between the pair is caught, not just the core COST_FIELDS."""
+    model_map = litellm.model_cost
+    assert undated in model_map, f"{undated} missing from model cost map"
+    assert model_map[undated] == model_map[dated], (
+        f"{undated} must exactly mirror {dated}; "
+        f"diff keys: {[k for k in set(model_map[undated]) | set(model_map[dated]) if model_map[undated].get(k) != model_map[dated].get(k)]}"
+    )