diff --git a/.github/scripts/auto_update_price_and_context_window_file.py b/.github/scripts/auto_update_price_and_context_window_file.py index 461d8d347d9..4c9a05bd4f0 100644 --- a/.github/scripts/auto_update_price_and_context_window_file.py +++ b/.github/scripts/auto_update_price_and_context_window_file.py @@ -1,6 +1,8 @@ import asyncio import aiohttp import json +import math +from typing import Any # Asynchronously fetch data from a given URL async def fetch_data(url): @@ -21,11 +23,157 @@ async def fetch_data(url): print("Error fetching data from URL:", e) return None + +FRIENDLI_API_URL = "https://api.friendli.ai/serverless/v1/models" +FRIENDLI_PROVIDER = "friendliai" + +INHERITABLE_BASE_KEYS = ( + "supports_pdf_input", + "supports_assistant_prefill", + "supports_adaptive_thinking", + "supports_output_config", +) + +REASONING_EFFORT_LEVEL_ORDER = ("none", "minimal", "low", "medium", "high", "xhigh", "max") + + +def _find_base_model_entry(base_model: str, local_data: dict) -> str | None: + if not base_model: + return None + bm_tail = base_model.split("/")[-1].lower() + if base_model in local_data: + return base_model + for key in local_data: + if key.startswith("sample_spec") or key == "fallback_generalizations": + continue + if key.split("/")[-1].lower() == bm_tail: + return key + return None + + +def _reasoning_effort_levels(reasoning_options: list) -> list: + offered = { + val + for opt in reasoning_options or [] + if opt.get("type") == "effort" + for val in opt.get("values", []) + } + return [level for level in REASONING_EFFORT_LEVEL_ORDER if level in offered] + + +def _valid_token_price(value: object) -> bool: + try: + price = float(value) # pyright: ignore[reportArgumentType] # non-numeric values are rejected via the except + except (TypeError, ValueError): + return False + return math.isfinite(price) and price >= 0 + + +def _has_valid_token_prices(pricing: dict | None) -> bool: + prices = pricing or {} + return _valid_token_price(prices.get("input")) and _valid_token_price(prices.get("output")) + + +def _pricing(pricing: dict) -> dict: + out: dict[str, Any] = {} + if not pricing: + return out + if "input" in pricing: + out["input_cost_per_token"] = float(pricing["input"]) + if "output" in pricing: + out["output_cost_per_token"] = float(pricing["output"]) + if "input_cache_read" in pricing and pricing["input_cache_read"] is not None: + out["cache_read_input_token_cost"] = float(pricing["input_cache_read"]) + return out + + +def _modality_flags(input_mods: list) -> dict: + mods = input_mods or [] + has_image = "image" in mods + return { + "supports_vision": has_image, + "supports_image_input": has_image, + "supports_video_input": "video" in mods, + } + + +def transform_friendli_data(data: list, local_data: dict) -> dict: + transformed: dict[str, dict] = {} + if not data: + return transformed + for model in data: + # An unpriced row must never wholesale-replace an already priced local entry: + # missing prices cost-calculate as zero, silently zeroing tracked spend + if not _has_valid_token_prices(model.get("pricing")): + continue + model_id = model["id"] + base_model = model.get("base_model") or "" + entry: dict[str, Any] = { + "litellm_provider": FRIENDLI_PROVIDER, + } + + base_key = _find_base_model_entry(base_model, local_data) + if base_key: + base_entry = local_data[base_key] + for k in INHERITABLE_BASE_KEYS: + if k in base_entry: + entry[k] = base_entry[k] + + ctx = model.get("context_length") + if ctx is not None: + entry["max_input_tokens"] = int(ctx) + max_out = model.get("max_completion_tokens") + if max_out is not None: + entry["max_output_tokens"] = int(max_out) + entry["max_tokens"] = int(max_out) + + pricing = _pricing(model.get("pricing", {})) + entry.update(pricing) + entry["supports_prompt_caching"] = "cache_read_input_token_cost" in pricing + + reasoning = model.get("reasoning") is True + entry["supports_reasoning"] = reasoning + if reasoning: + entry["reasoning_effort_levels"] = _reasoning_effort_levels( + model.get("reasoning_options", []) + ) + + func = model.get("functionality", {}) + entry["supports_function_calling"] = func.get("tool_call") is True + entry["supports_parallel_function_calling"] = func.get("parallel_tool_call") is True + is_struct = func.get("structured_output") is True + entry["supports_response_schema"] = is_struct + entry["supports_native_structured_output"] = is_struct + entry["supports_system_messages"] = func.get("system_messages") is True + entry["supports_tool_choice"] = func.get("tool_choice") is True + + entry.update(_modality_flags(model.get("input_modalities", []))) + + entry["mode"] = model.get("mode", "chat") + + desc = model.get("description") + if desc: + entry["comment"] = desc + + dep = model.get("deprecation_date") + if dep: + entry["deprecation_date"] = dep.split("T")[0] + + entry["source"] = FRIENDLI_API_URL + + transformed[f"{FRIENDLI_PROVIDER}/{model_id}"] = entry + return transformed + # Synchronize local data with remote data -def sync_local_data_with_remote(local_data, remote_data): +def sync_local_data_with_remote(local_data, remote_data, replace_keys=frozenset()): # Update existing keys in local_data with values from remote_data + # (replace_keys entries are swapped wholesale so a field the remote catalog + # dropped, e.g. cache pricing, cannot survive as a stale value) for key in (set(local_data) & set(remote_data)): - local_data[key].update(remote_data[key]) + if key in replace_keys: + local_data[key] = remote_data[key] + else: + local_data[key].update(remote_data[key]) # Add new keys from remote_data to local_data for key in (set(remote_data) - set(local_data)): @@ -46,6 +194,8 @@ def write_to_file(file_path, data): # Update the existing models and add the missing models for OpenRouter def transform_openrouter_data(data): transformed = {} + if not data: + return transformed for row in data: # Add the fields 'max_tokens' and 'input_cost_per_token' obj = { @@ -84,7 +234,14 @@ def transform_openrouter_data(data): # Update the existing models and add the missing models for Vercel AI Gateway def transform_vercel_ai_gateway_data(data): transformed = {} + if not data: + return transformed for row in data: + # Rows without token pricing or token limits (video/embedding models) previously KeyError'd the whole sync + if any(row.get(k) is None for k in ("context_window", "max_tokens")) or any( + row.get("pricing", {}).get(k) is None for k in ("input", "output") + ): + continue obj = { "max_tokens": row["context_window"], "input_cost_per_token": float(row["pricing"]["input"]), @@ -143,13 +300,16 @@ def main(): vercel_data = asyncio.run(fetch_data(vercel_ai_gateway_url)) # Transform the fetched Vercel AI Gateway data vercel_data = transform_vercel_ai_gateway_data(vercel_data) + + friendli_data = asyncio.run(fetch_data(FRIENDLI_API_URL)) + friendli_data = transform_friendli_data(friendli_data, local_data) # Combine both datasets - all_remote_data = {**openrouter_data, **vercel_data} + all_remote_data = {**openrouter_data, **vercel_data, **friendli_data} # If both local and openrouter data are available, synchronize and save if local_data and all_remote_data: - sync_local_data_with_remote(local_data, all_remote_data) + sync_local_data_with_remote(local_data, all_remote_data, replace_keys=frozenset(friendli_data)) write_to_file(local_file_path, local_data) else: print("Failed to fetch model data from either local file or URL.") diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 06c7a6aa46e..2220d0e1fe5 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -23083,58 +23083,206 @@ }, "friendliai/zai-org/GLM-5.3-Flash": { "litellm_provider": "friendliai", - "supports_reasoning": true, - "supports_function_calling": true, "max_input_tokens": 1048576, - "max_tokens": 1048576, "max_output_tokens": 1048576, + "max_tokens": 1048576, "input_cost_per_token": 1.5e-07, "output_cost_per_token": 5e-07, "cache_read_input_token_cost": 3e-08, "supports_prompt_caching": true, + "supports_reasoning": true, "reasoning_effort_levels": [ "low", "high", "max" ], + "supports_function_calling": 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": "Native multimodal GLM model for efficient coding and long-horizon agent tasks", - "source": "https://api.friendli.ai/serverless/v1/models", "supports_vision": true, "supports_image_input": true, - "supports_video_input": true + "supports_video_input": true, + "mode": "chat", + "comment": "Native multimodal GLM model for efficient coding and long-horizon agent tasks", + "source": "https://api.friendli.ai/serverless/v1/models" }, "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, + "max_tokens": 1048576, "input_cost_per_token": 1.26e-06, "output_cost_per_token": 3.96e-06, "cache_read_input_token_cost": 2.34e-07, "supports_prompt_caching": true, + "supports_reasoning": true, "reasoning_effort_levels": [ "low", "high", "max" ], + "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true, "supports_native_structured_output": true, "supports_system_messages": true, "supports_tool_choice": true, + "supports_vision": false, + "supports_image_input": false, + "supports_video_input": false, "mode": "chat", "comment": "Flagship GLM model for long-horizon coding, agents, and complex project delivery", - "source": "https://api.friendli.ai/serverless/v1/models", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/google/gemma-4-31B-it": { + "litellm_provider": "friendliai", + "max_input_tokens": 262144, + "max_output_tokens": 262144, + "max_tokens": 262144, + "input_cost_per_token": 1.4e-07, + "output_cost_per_token": 4e-07, + "supports_prompt_caching": false, + "supports_reasoning": true, + "reasoning_effort_levels": [], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": true, + "supports_image_input": true, + "supports_video_input": false, + "mode": "chat", + "comment": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/zai-org/GLM-5.2": { + "litellm_provider": "friendliai", + "max_input_tokens": 1048576, + "max_output_tokens": 1048576, + "max_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_reasoning": true, + "reasoning_effort_levels": [ + "high", + "max" + ], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, "supports_vision": false, - "supports_image_input": false + "supports_image_input": false, + "supports_video_input": false, + "mode": "chat", + "comment": "Open flagship GLM for long-horizon coding agents and million-token context work", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/LGAI-EXAONE/K-EXAONE-2.0-750B-A37B": { + "litellm_provider": "friendliai", + "max_input_tokens": 262144, + "max_output_tokens": 262144, + "max_tokens": 262144, + "input_cost_per_token": 6e-07, + "output_cost_per_token": 2.4e-06, + "cache_read_input_token_cost": 1.2e-07, + "supports_prompt_caching": true, + "supports_reasoning": true, + "reasoning_effort_levels": [], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_image_input": false, + "supports_video_input": false, + "mode": "chat", + "comment": "Frontier-scale multilingual language model developed by LG AI Research", + "deprecation_date": "2026-09-06", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/deepseek-ai/DeepSeek-V3.2": { + "litellm_provider": "friendliai", + "max_input_tokens": 163840, + "max_output_tokens": 163840, + "max_tokens": 163840, + "input_cost_per_token": 5e-07, + "output_cost_per_token": 1.5e-06, + "cache_read_input_token_cost": 2.5e-07, + "supports_prompt_caching": true, + "supports_reasoning": true, + "reasoning_effort_levels": [], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_image_input": false, + "supports_video_input": false, + "mode": "chat", + "comment": "DeepSeek chat model for instruction following, coding, and analysis", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/MiniMaxAI/MiniMax-M2.5": { + "litellm_provider": "friendliai", + "max_input_tokens": 196608, + "max_output_tokens": 196608, + "max_tokens": 196608, + "input_cost_per_token": 3e-07, + "output_cost_per_token": 1.2e-06, + "cache_read_input_token_cost": 6e-08, + "supports_prompt_caching": true, + "supports_reasoning": true, + "reasoning_effort_levels": [], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_image_input": false, + "supports_video_input": false, + "mode": "chat", + "comment": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/zai-org/GLM-5.1": { + "litellm_provider": "friendliai", + "max_input_tokens": 202752, + "max_output_tokens": 202752, + "max_tokens": 202752, + "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_reasoning": true, + "reasoning_effort_levels": [], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_image_input": false, + "supports_video_input": false, + "mode": "chat", + "comment": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "source": "https://api.friendli.ai/serverless/v1/models" }, "ft:babbage-002": { "deprecation_date": "2026-10-23", diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 06c7a6aa46e..2220d0e1fe5 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -23083,58 +23083,206 @@ }, "friendliai/zai-org/GLM-5.3-Flash": { "litellm_provider": "friendliai", - "supports_reasoning": true, - "supports_function_calling": true, "max_input_tokens": 1048576, - "max_tokens": 1048576, "max_output_tokens": 1048576, + "max_tokens": 1048576, "input_cost_per_token": 1.5e-07, "output_cost_per_token": 5e-07, "cache_read_input_token_cost": 3e-08, "supports_prompt_caching": true, + "supports_reasoning": true, "reasoning_effort_levels": [ "low", "high", "max" ], + "supports_function_calling": 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": "Native multimodal GLM model for efficient coding and long-horizon agent tasks", - "source": "https://api.friendli.ai/serverless/v1/models", "supports_vision": true, "supports_image_input": true, - "supports_video_input": true + "supports_video_input": true, + "mode": "chat", + "comment": "Native multimodal GLM model for efficient coding and long-horizon agent tasks", + "source": "https://api.friendli.ai/serverless/v1/models" }, "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, + "max_tokens": 1048576, "input_cost_per_token": 1.26e-06, "output_cost_per_token": 3.96e-06, "cache_read_input_token_cost": 2.34e-07, "supports_prompt_caching": true, + "supports_reasoning": true, "reasoning_effort_levels": [ "low", "high", "max" ], + "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true, "supports_native_structured_output": true, "supports_system_messages": true, "supports_tool_choice": true, + "supports_vision": false, + "supports_image_input": false, + "supports_video_input": false, "mode": "chat", "comment": "Flagship GLM model for long-horizon coding, agents, and complex project delivery", - "source": "https://api.friendli.ai/serverless/v1/models", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/google/gemma-4-31B-it": { + "litellm_provider": "friendliai", + "max_input_tokens": 262144, + "max_output_tokens": 262144, + "max_tokens": 262144, + "input_cost_per_token": 1.4e-07, + "output_cost_per_token": 4e-07, + "supports_prompt_caching": false, + "supports_reasoning": true, + "reasoning_effort_levels": [], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": true, + "supports_image_input": true, + "supports_video_input": false, + "mode": "chat", + "comment": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/zai-org/GLM-5.2": { + "litellm_provider": "friendliai", + "max_input_tokens": 1048576, + "max_output_tokens": 1048576, + "max_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_reasoning": true, + "reasoning_effort_levels": [ + "high", + "max" + ], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, "supports_vision": false, - "supports_image_input": false + "supports_image_input": false, + "supports_video_input": false, + "mode": "chat", + "comment": "Open flagship GLM for long-horizon coding agents and million-token context work", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/LGAI-EXAONE/K-EXAONE-2.0-750B-A37B": { + "litellm_provider": "friendliai", + "max_input_tokens": 262144, + "max_output_tokens": 262144, + "max_tokens": 262144, + "input_cost_per_token": 6e-07, + "output_cost_per_token": 2.4e-06, + "cache_read_input_token_cost": 1.2e-07, + "supports_prompt_caching": true, + "supports_reasoning": true, + "reasoning_effort_levels": [], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_image_input": false, + "supports_video_input": false, + "mode": "chat", + "comment": "Frontier-scale multilingual language model developed by LG AI Research", + "deprecation_date": "2026-09-06", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/deepseek-ai/DeepSeek-V3.2": { + "litellm_provider": "friendliai", + "max_input_tokens": 163840, + "max_output_tokens": 163840, + "max_tokens": 163840, + "input_cost_per_token": 5e-07, + "output_cost_per_token": 1.5e-06, + "cache_read_input_token_cost": 2.5e-07, + "supports_prompt_caching": true, + "supports_reasoning": true, + "reasoning_effort_levels": [], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_image_input": false, + "supports_video_input": false, + "mode": "chat", + "comment": "DeepSeek chat model for instruction following, coding, and analysis", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/MiniMaxAI/MiniMax-M2.5": { + "litellm_provider": "friendliai", + "max_input_tokens": 196608, + "max_output_tokens": 196608, + "max_tokens": 196608, + "input_cost_per_token": 3e-07, + "output_cost_per_token": 1.2e-06, + "cache_read_input_token_cost": 6e-08, + "supports_prompt_caching": true, + "supports_reasoning": true, + "reasoning_effort_levels": [], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_image_input": false, + "supports_video_input": false, + "mode": "chat", + "comment": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "source": "https://api.friendli.ai/serverless/v1/models" + }, + "friendliai/zai-org/GLM-5.1": { + "litellm_provider": "friendliai", + "max_input_tokens": 202752, + "max_output_tokens": 202752, + "max_tokens": 202752, + "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_reasoning": true, + "reasoning_effort_levels": [], + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_response_schema": true, + "supports_native_structured_output": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_image_input": false, + "supports_video_input": false, + "mode": "chat", + "comment": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "source": "https://api.friendli.ai/serverless/v1/models" }, "ft:babbage-002": { "deprecation_date": "2026-10-23", diff --git a/tests/test_litellm/test_auto_update_price_and_context_window_file.py b/tests/test_litellm/test_auto_update_price_and_context_window_file.py new file mode 100644 index 00000000000..435747e9a09 --- /dev/null +++ b/tests/test_litellm/test_auto_update_price_and_context_window_file.py @@ -0,0 +1,222 @@ +"""Unit tests for the Friendli transform in +`.github/scripts/auto_update_price_and_context_window_file.py`.""" + +from __future__ import annotations + +import importlib.util +import sys +from pathlib import Path + +import pytest + +SCRIPT_PATH = ( + Path(__file__).resolve().parents[2] + / ".github" + / "scripts" + / "auto_update_price_and_context_window_file.py" +) + + +@pytest.fixture(scope="module") +def sync_module(): + spec = importlib.util.spec_from_file_location( + "auto_update_price_and_context_window_file", SCRIPT_PATH + ) + assert spec and spec.loader, f"Could not load spec for {SCRIPT_PATH}" + module = importlib.util.module_from_spec(spec) + sys.modules["auto_update_price_and_context_window_file"] = module + spec.loader.exec_module(module) + return module + + +def _reasoning_model(**overrides: object) -> dict: + model = { + "id": "zai-org/GLM-Test", + "base_model": "zhipuai/glm-test", + "context_length": 1048576, + "max_completion_tokens": 131072, + "pricing": {"input": "0.00000015", "output": "0.0000005", "input_cache_read": "0.00000003"}, + "reasoning": True, + "reasoning_options": [{"type": "effort", "values": ["max", "high", "low"]}], + "functionality": { + "tool_call": True, + "parallel_tool_call": True, + "structured_output": True, + "system_messages": True, + "tool_choice": True, + }, + "input_modalities": ["text", "image", "video"], + "mode": "chat", + } + model.update(overrides) + return model + + +def test_transform_emits_declared_effort_levels_in_canonical_order(sync_module): + entry = sync_module.transform_friendli_data([_reasoning_model()], {})[ + "friendliai/zai-org/GLM-Test" + ] + assert entry["supports_reasoning"] is True + assert entry["reasoning_effort_levels"] == ["low", "high", "max"] + assert not any(k.endswith("_reasoning_effort") for k in entry) + + +def test_transform_reasoning_model_without_effort_options_declares_empty_levels(sync_module): + model = _reasoning_model(reasoning_options=[{"type": "budget_tokens", "values": []}]) + entry = sync_module.transform_friendli_data([model], {})["friendliai/zai-org/GLM-Test"] + assert entry["reasoning_effort_levels"] == [] + + +def test_transform_non_reasoning_model_declares_no_levels(sync_module): + model = _reasoning_model(reasoning=False, reasoning_options=[]) + entry = sync_module.transform_friendli_data([model], {})["friendliai/zai-org/GLM-Test"] + assert entry["supports_reasoning"] is False + assert "reasoning_effort_levels" not in entry + + +def test_transform_max_tokens_mirrors_output_cap_not_context(sync_module): + entry = sync_module.transform_friendli_data([_reasoning_model()], {})[ + "friendliai/zai-org/GLM-Test" + ] + assert entry["max_input_tokens"] == 1048576 + assert entry["max_output_tokens"] == 131072 + assert entry["max_tokens"] == entry["max_output_tokens"] + + +def test_transform_prompt_caching_follows_cache_pricing(sync_module): + cached = sync_module.transform_friendli_data([_reasoning_model()], {})[ + "friendliai/zai-org/GLM-Test" + ] + assert cached["supports_prompt_caching"] is True + assert cached["cache_read_input_token_cost"] == 3e-08 + + uncached_model = _reasoning_model(pricing={"input": "0.00000014", "output": "0.0000004"}) + uncached = sync_module.transform_friendli_data([uncached_model], {})[ + "friendliai/zai-org/GLM-Test" + ] + assert uncached["supports_prompt_caching"] is False + assert "cache_read_input_token_cost" not in uncached + + +def test_transform_modalities_set_vision_image_and_video_flags(sync_module): + entry = sync_module.transform_friendli_data([_reasoning_model()], {})[ + "friendliai/zai-org/GLM-Test" + ] + assert entry["supports_vision"] is True + assert entry["supports_image_input"] is True + assert entry["supports_video_input"] is True + + text_only = _reasoning_model(input_modalities=["text"]) + entry_text = sync_module.transform_friendli_data([text_only], {})[ + "friendliai/zai-org/GLM-Test" + ] + assert entry_text["supports_vision"] is False + assert entry_text["supports_image_input"] is False + assert entry_text["supports_video_input"] is False + + +def test_transform_skips_rows_without_valid_token_prices_so_priced_local_entries_survive(sync_module): + local = { + "friendliai/zai-org/GLM-Test": { + "litellm_provider": "friendliai", + "input_cost_per_token": 1.5e-07, + "output_cost_per_token": 5e-07, + } + } + unpriced_rows = [ + _reasoning_model(pricing={}), + _reasoning_model(pricing=None), + _reasoning_model(pricing={"input": "0.00000015"}), + _reasoning_model(pricing={"output": "0.0000005"}), + _reasoning_model(pricing={"input": "not-a-number", "output": "0.0000005"}), + _reasoning_model(pricing={"input": "-0.00000015", "output": "0.0000005"}), + _reasoning_model(pricing={"input": "inf", "output": "0.0000005"}), + _reasoning_model(pricing={"input": "nan", "output": "0.0000005"}), + ] + remote = sync_module.transform_friendli_data(unpriced_rows, local) + assert remote == {} + sync_module.sync_local_data_with_remote(local, remote, replace_keys=frozenset(remote)) + assert local["friendliai/zai-org/GLM-Test"]["input_cost_per_token"] == 1.5e-07 + assert local["friendliai/zai-org/GLM-Test"]["output_cost_per_token"] == 5e-07 + + +def test_transform_keeps_zero_priced_rows(sync_module): + free_model = _reasoning_model(pricing={"input": "0", "output": "0"}) + entry = sync_module.transform_friendli_data([free_model], {})["friendliai/zai-org/GLM-Test"] + assert entry["input_cost_per_token"] == 0.0 + assert entry["output_cost_per_token"] == 0.0 + + +def test_transforms_survive_failed_fetch(sync_module): + assert sync_module.transform_friendli_data(None, {}) == {} + assert sync_module.transform_friendli_data([], {}) == {} + assert sync_module.transform_openrouter_data(None) == {} + assert sync_module.transform_vercel_ai_gateway_data(None) == {} + + +def test_vercel_transform_skips_rows_without_token_pricing_or_limits(sync_module): + rows = [ + { + "id": "wan-video", + "pricing": {"video_duration_pricing": [{"resolution": "720p", "cost_per_second": "0.1"}]}, + }, + { + "id": "qwen3-embedding", + "context_window": 32768, + "max_tokens": 32768, + "pricing": {"input": "0.00000001"}, + }, + { + "id": "no-limits-chat", + "pricing": {"input": "0.000001", "output": "0.000002"}, + }, + { + "id": "good-chat", + "context_window": 128000, + "max_tokens": 8192, + "pricing": {"input": "0.000001", "output": "0.000002"}, + }, + ] + transformed = sync_module.transform_vercel_ai_gateway_data(rows) + assert list(transformed) == ["vercel_ai_gateway/good-chat"] + assert transformed["vercel_ai_gateway/good-chat"]["input_cost_per_token"] == 1e-06 + assert transformed["vercel_ai_gateway/good-chat"]["output_cost_per_token"] == 2e-06 + + +def test_sync_replaces_friendli_entries_so_dropped_cache_pricing_does_not_survive(sync_module): + local = { + "friendliai/zai-org/GLM-Test": { + "litellm_provider": "friendliai", + "cache_read_input_token_cost": 3e-08, + "supports_prompt_caching": True, + } + } + uncached_model = _reasoning_model(pricing={"input": "0.00000014", "output": "0.0000004"}) + remote = sync_module.transform_friendli_data([uncached_model], local) + sync_module.sync_local_data_with_remote(local, remote, replace_keys=frozenset(remote)) + synced = local["friendliai/zai-org/GLM-Test"] + assert "cache_read_input_token_cost" not in synced + assert synced["supports_prompt_caching"] is False + + +def test_sync_still_merges_entries_outside_replace_keys(sync_module): + local = {"openrouter/some-model": {"input_cost_per_token": 1e-06, "supports_vision": True}} + remote = {"openrouter/some-model": {"input_cost_per_token": 2e-06}} + sync_module.sync_local_data_with_remote(local, remote) + assert local["openrouter/some-model"] == {"input_cost_per_token": 2e-06, "supports_vision": True} + + +def test_transform_inherits_allowlisted_keys_from_base_model_entry(sync_module): + local = { + "zhipuai/glm-test": { + "supports_pdf_input": True, + "supports_assistant_prefill": True, + "input_cost_per_token": 9e-06, + } + } + entry = sync_module.transform_friendli_data([_reasoning_model()], local)[ + "friendliai/zai-org/GLM-Test" + ] + assert entry["supports_pdf_input"] is True + assert entry["supports_assistant_prefill"] is True + assert entry["input_cost_per_token"] == 1.5e-07