feat(friendli): add zai-org/GLM-5.3 model pricing

Per https://api.friendli.ai/serverless/v1/models:
- $1.40 input / $4.40 output / $0.26 cached input per MTok
- 1M context, 1M max output, reasoning with effort low/high/max
  (per HF chat_template.jinja: low/high honored, anything else -> max)
- tool calling, parallel tool calls, structured output, prompt caching
- text-only (no vision), flagship GLM model

model_prices_and_context_window.json

@@ -19564,6 +19564,30 @@
         "supports_system_messages": true,
         "supports_tool_choice": true
     },
+    "friendliai/zai-org/GLM-5.3": {
+        "litellm_provider": "friendliai",
+        "supports_reasoning": true,
+        "supports_function_calling": true,
+        "max_input_tokens": 1048576,
+        "max_tokens": 1048576,
+        "max_output_tokens": 1048576,
+        "input_cost_per_token": 1.4e-06,
+        "output_cost_per_token": 4.4e-06,
+        "cache_read_input_token_cost": 2.6e-07,
+        "supports_prompt_caching": true,
+        "supports_max_reasoning_effort": true,
+        "supports_low_reasoning_effort": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_native_structured_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true,
+        "mode": "chat",
+        "comment": "Flagship GLM model for long-horizon coding, agents, and complex project delivery",
+        "source": "https://api.friendli.ai/serverless/v1/models",
+        "supports_vision": false,
+        "supports_image_input": false
+    },
     "ft:babbage-002": {
         "deprecation_date": "2026-10-23",
         "input_cost_per_token": 1.6e-06,

tests/test_litellm/test_friendli_glm_5_3_model_metadata.py

@@ -0,0 +1,36 @@
+import json
+from pathlib import Path
+
+from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
+
+
+def test_friendli_glm_5_3_model_info():
+    model = "friendliai/zai-org/GLM-5.3"
+    json_path = Path(__file__).parents[2] / "model_prices_and_context_window.json"
+    with open(json_path) as f:
+        model_cost = json.load(f)
+
+    info = model_cost.get(model)
+    assert (
+        info is not None
+    ), f"{model} not found in model_prices_and_context_window.json"
+    assert info["litellm_provider"] == "friendliai"
+    assert info["mode"] == "chat"
+    assert info["input_cost_per_token"] == 1.4e-06
+    assert info["output_cost_per_token"] == 4.4e-06
+    assert info["cache_read_input_token_cost"] == 2.6e-07
+    assert info["max_input_tokens"] == 1048576
+    assert info["max_output_tokens"] == 1048576
+    assert info["supports_function_calling"] is True
+    assert info["supports_reasoning"] is True
+    assert info["supports_low_reasoning_effort"] is True
+    assert info["supports_max_reasoning_effort"] is True
+    assert info["supports_tool_choice"] is True
+    assert info["supports_prompt_caching"] is True
+    # GLM-5.3 (non-flash) is text-only on Friendli's catalog
+    assert info["supports_vision"] is False
+    assert info["supports_image_input"] is False
+
+    routed_model, provider, _, _ = get_llm_provider(model=model)
+    assert routed_model == "zai-org/GLM-5.3"
+    assert provider == "friendliai"