Adversary MCP Server

{ "pricing_models": { "gpt-5": { "prompt_tokens_per_1k": 0.00125, "completion_tokens_per_1k": 0.01, "currency": "USD", "description": "GPT-5 - OpenAI's latest flagship unified model with reasoning capabilities", "display_name": "GPT-5", "category": "latest", "capabilities": ["chat", "code", "analysis", "reasoning", "multimodal", "agentic"], "max_context_tokens": 272000, "max_output_tokens": 128000, "speed_rating": "medium", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["Advanced coding", "Complex reasoning", "Agentic tasks", "Production workflows"], "notes": ["Most advanced OpenAI model", "Unified reasoning + fast response", "Frontier performance", "272K context window"] }, "gpt-5-mini": { "prompt_tokens_per_1k": 0.00025, "completion_tokens_per_1k": 0.002, "currency": "USD", "description": "GPT-5 Mini - Smaller, efficient version of GPT-5", "display_name": "GPT-5 Mini", "category": "latest", "capabilities": ["chat", "code", "analysis", "reasoning", "fast"], "max_context_tokens": 272000, "max_output_tokens": 128000, "speed_rating": "fast", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["High-volume tasks", "Cost-effective reasoning", "Fast responses"], "notes": ["Balance of cost and capability", "Same context as GPT-5", "Efficient reasoning"] }, "gpt-5-nano": { "prompt_tokens_per_1k": 0.00005, "completion_tokens_per_1k": 0.0004, "currency": "USD", "description": "GPT-5 Nano - Ultra-efficient version for high-volume use", "display_name": "GPT-5 Nano", "category": "budget", "capabilities": ["chat", "code", "fast"], "max_context_tokens": 272000, "max_output_tokens": 128000, "speed_rating": "fast", "quality_rating": "good", "is_available": true, "is_deprecated": false, "best_for": ["Ultra high-volume", "Budget-conscious applications", "Simple tasks"], "notes": ["Most cost-effective GPT-5 variant", "Large context at low cost"] }, "gpt-5-chat-latest": { "prompt_tokens_per_1k": 0.00125, "completion_tokens_per_1k": 0.01, "currency": "USD", "description": "GPT-5 Chat - Non-reasoning version optimized for ChatGPT", "display_name": "GPT-5 Chat", "category": "latest", "capabilities": ["chat", "code", "analysis", "fast"], "max_context_tokens": 272000, "max_output_tokens": 128000, "speed_rating": "fast", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["Chat applications", "Fast responses", "Non-reasoning tasks"], "notes": ["Non-reasoning variant of GPT-5", "Optimized for speed"] }, "gpt-4o": { "prompt_tokens_per_1k": 0.0025, "completion_tokens_per_1k": 0.01, "currency": "USD", "description": "GPT-4o - Multimodal model with vision and audio capabilities", "display_name": "GPT-4o", "category": "specialized", "capabilities": ["chat", "code", "analysis", "reasoning", "multimodal", "vision", "audio"], "max_context_tokens": 128000, "speed_rating": "fast", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["Multimodal tasks", "Vision analysis", "Audio processing"], "notes": ["Strong multimodal capabilities", "Vision and audio support"] }, "gpt-4o-mini": { "prompt_tokens_per_1k": 0.00015, "completion_tokens_per_1k": 0.0006, "currency": "USD", "description": "GPT-4o Mini - Cost-effective multimodal model", "display_name": "GPT-4o Mini", "category": "budget", "capabilities": ["chat", "code", "analysis", "multimodal", "vision", "fast"], "max_context_tokens": 128000, "speed_rating": "fast", "quality_rating": "good", "is_available": true, "is_deprecated": false, "best_for": ["Budget multimodal", "High-volume vision tasks", "Cost-effective analysis"], "notes": ["Cheapest multimodal option", "Good vision capabilities"] }, "gpt-4-turbo": { "prompt_tokens_per_1k": 0.01, "completion_tokens_per_1k": 0.03, "currency": "USD", "description": "GPT-4 Turbo - Large context window model", "display_name": "GPT-4 Turbo", "category": "legacy", "capabilities": ["chat", "code", "analysis", "reasoning", "high_context"], "max_context_tokens": 128000, "speed_rating": "medium", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["Large documents", "Legacy applications"], "notes": ["Being superseded by GPT-5", "Still reliable for large context"] }, "o3": { "prompt_tokens_per_1k": 0.015, "completion_tokens_per_1k": 0.06, "currency": "USD", "description": "o3 - Advanced reasoning model for complex problems", "display_name": "o3", "category": "specialized", "capabilities": ["reasoning", "analysis", "code", "math", "science"], "max_context_tokens": 200000, "speed_rating": "slow", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["Complex reasoning", "Scientific problems", "Advanced math"], "notes": ["Dedicated reasoning model", "High-quality but slower"] }, "o4-mini": { "prompt_tokens_per_1k": 0.003, "completion_tokens_per_1k": 0.012, "currency": "USD", "description": "o4-mini - Efficient reasoning model for coding and math", "display_name": "o4-mini", "category": "specialized", "capabilities": ["reasoning", "code", "math", "analysis"], "max_context_tokens": 200000, "speed_rating": "medium", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["Coding problems", "Math tasks", "Cost-effective reasoning"], "notes": ["Efficient reasoning", "Good for STEM use cases"] }, "claude-opus-4.1": { "prompt_tokens_per_1k": 0.015, "completion_tokens_per_1k": 0.075, "currency": "USD", "description": "Claude Opus 4.1 - Anthropic's most intelligent hybrid reasoning model", "display_name": "Claude Opus 4.1", "category": "latest", "capabilities": ["chat", "code", "analysis", "reasoning", "agentic", "hybrid_reasoning"], "max_context_tokens": 200000, "speed_rating": "medium", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["Frontier intelligence", "Complex AI agents", "Advanced coding", "Long-horizon tasks"], "notes": ["Most intelligent Claude model", "Hybrid reasoning capability", "Premium performance", "Extended thinking mode available"] }, "claude-opus-4": { "prompt_tokens_per_1k": 0.015, "completion_tokens_per_1k": 0.075, "currency": "USD", "description": "Claude Opus 4 - Previous generation flagship model", "display_name": "Claude Opus 4", "category": "latest", "capabilities": ["chat", "code", "analysis", "reasoning", "agentic"], "max_context_tokens": 200000, "speed_rating": "medium", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["Complex coding", "AI agents", "Creative writing"], "notes": ["Previous flagship", "Strong coding capabilities", "Background task support"] }, "claude-sonnet-4-20250514": { "prompt_tokens_per_1k": 0.003, "completion_tokens_per_1k": 0.015, "currency": "USD", "description": "Claude Sonnet 4 - Superior intelligence with optimal efficiency", "display_name": "Claude Sonnet 4", "category": "latest", "capabilities": ["chat", "code", "analysis", "reasoning", "hybrid_reasoning", "computer_use"], "max_context_tokens": 1000000, "max_output_tokens": 64000, "speed_rating": "medium", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["Production coding", "Customer agents", "Large codebases", "Document synthesis"], "notes": ["1M token context window", "Hybrid reasoning", "Computer use capability", "Long context premium pricing above 200K tokens"] }, "claude-sonnet-3.7": { "prompt_tokens_per_1k": 0.003, "completion_tokens_per_1k": 0.015, "currency": "USD", "description": "Claude Sonnet 3.7 - First hybrid reasoning model", "display_name": "Claude Sonnet 3.7", "category": "latest", "capabilities": ["chat", "code", "analysis", "reasoning", "hybrid_reasoning"], "max_context_tokens": 200000, "speed_rating": "medium", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["Coding", "Content generation", "Data analysis", "Planning"], "notes": ["First hybrid reasoning model", "State-of-the-art coding", "Extended thinking capability"] }, "claude-3-5-sonnet-20241022": { "prompt_tokens_per_1k": 0.003, "completion_tokens_per_1k": 0.015, "currency": "USD", "description": "Claude 3.5 Sonnet - Previous generation balanced model", "display_name": "Claude 3.5 Sonnet", "category": "legacy", "capabilities": ["chat", "code", "analysis", "reasoning"], "max_context_tokens": 200000, "speed_rating": "medium", "quality_rating": "excellent", "is_available": true, "is_deprecated": false, "best_for": ["General purpose", "Balanced performance"], "notes": ["Superseded by Sonnet 3.7", "Still reliable for most tasks"] }, "claude-3-5-haiku-20241022": { "prompt_tokens_per_1k": 0.0008, "completion_tokens_per_1k": 0.004, "currency": "USD", "description": "Claude 3.5 Haiku - Fast and affordable model", "display_name": "Claude 3.5 Haiku", "category": "budget", "capabilities": ["chat", "code", "analysis", "fast"], "max_context_tokens": 200000, "speed_rating": "fast", "quality_rating": "good", "is_available": true, "is_deprecated": false, "best_for": ["Fast scanning", "Budget analysis", "High-volume tasks"], "notes": ["Fast and affordable", "Good for routine tasks"] }, "claude-3-haiku-20240307": { "prompt_tokens_per_1k": 0.00025, "completion_tokens_per_1k": 0.00125, "currency": "USD", "description": "Claude 3 Haiku - Ultra-affordable legacy model", "display_name": "Claude 3 Haiku", "category": "budget", "capabilities": ["chat", "code", "fast"], "max_context_tokens": 200000, "speed_rating": "fast", "quality_rating": "basic", "is_available": true, "is_deprecated": false, "best_for": ["Budget applications", "Simple tasks", "High-volume processing"], "notes": ["Most affordable Claude", "Legacy but still functional"] }, "claude-3-opus-20240229": { "prompt_tokens_per_1k": 0.015, "completion_tokens_per_1k": 0.075, "currency": "USD", "description": "Claude 3 Opus - Legacy maximum capability model (DEPRECATED)", "display_name": "Claude 3 Opus (Deprecated)", "category": "legacy", "capabilities": ["chat", "code", "analysis", "reasoning"], "max_context_tokens": 200000, "speed_rating": "slow", "quality_rating": "excellent", "is_available": false, "is_deprecated": true, "best_for": [], "notes": ["DEPRECATED: No longer available", "Use Claude Opus 4.1 or Claude Sonnet 4 instead"] }, "claude-3-sonnet-20240229": { "prompt_tokens_per_1k": 0.003, "completion_tokens_per_1k": 0.015, "currency": "USD", "description": "Claude 3 Sonnet - Legacy balanced model (DEPRECATED)", "display_name": "Claude 3 Sonnet (Deprecated)", "category": "legacy", "capabilities": ["chat", "code", "analysis"], "max_context_tokens": 200000, "speed_rating": "medium", "quality_rating": "good", "is_available": false, "is_deprecated": true, "best_for": [], "notes": ["DEPRECATED: No longer available", "Use Claude 3.5 Sonnet or Claude Sonnet 4 instead"] } }, "batch_pricing": { "claude-opus-4.1": { "prompt_tokens_per_1k": 0.0075, "completion_tokens_per_1k": 0.0375, "description": "50% discount for batch processing" }, "claude-opus-4": { "prompt_tokens_per_1k": 0.0075, "completion_tokens_per_1k": 0.0375, "description": "50% discount for batch processing" }, "claude-sonnet-4-20250514": { "prompt_tokens_per_1k": 0.0015, "completion_tokens_per_1k": 0.0075, "description": "50% discount for batch processing" }, "claude-sonnet-3.7": { "prompt_tokens_per_1k": 0.0015, "completion_tokens_per_1k": 0.0075, "description": "50% discount for batch processing" }, "claude-3-5-sonnet-20241022": { "prompt_tokens_per_1k": 0.0015, "completion_tokens_per_1k": 0.0075, "description": "50% discount for batch processing" }, "claude-3-5-haiku-20241022": { "prompt_tokens_per_1k": 0.0004, "completion_tokens_per_1k": 0.002, "description": "50% discount for batch processing" }, "claude-3-haiku-20240307": { "prompt_tokens_per_1k": 0.000125, "completion_tokens_per_1k": 0.000625, "description": "50% discount for batch processing" } }, "prompt_caching": { "description": "Anthropic offers prompt caching to reduce costs for repeated context", "cache_write_multiplier_5min": 1.25, "cache_write_multiplier_1hour": 2.0, "cache_read_multiplier": 0.1, "notes": ["5-minute cache: 1.25x base price for writes, 0.1x for reads", "1-hour cache: 2x base price for writes, 0.1x for reads"] }, "special_features": { "web_search": { "anthropic": { "price_per_1k_searches": 10, "description": "Web search tool for Claude models", "notes": ["Plus standard token costs for search content"] } }, "computer_use": { "anthropic": { "description": "Computer use capability for Claude Sonnet 4", "system_prompt_overhead": "466-499 tokens", "tool_definition_tokens": "735 tokens (Claude 4/Sonnet 3.7), 683 tokens (Claude 3.5 Sonnet)" } }, "code_execution": { "anthropic": { "price_per_session_hour": 0.05, "minimum_session": "5 minutes", "description": "Python code execution in sandboxed environment" } }, "long_context": { "claude-sonnet-4": { "standard_context": 200000, "extended_context": 1000000, "premium_threshold": 200000, "description": "Premium pricing for inputs over 200K tokens", "availability": "Beta for Tier 4 customers" } } }, "default_pricing": { "prompt_tokens_per_1k": 0.01, "completion_tokens_per_1k": 0.03, "currency": "USD", "description": "Default pricing for unknown models", "category": "legacy", "capabilities": ["chat"], "speed_rating": "medium", "quality_rating": "good", "is_available": true, "is_deprecated": false }, "metadata": { "last_updated": "2025-08-16", "version": "3.0", "source": "Official provider pricing as of August 2025", "notes": [ "Major updates: Added GPT-5 family, Claude Opus 4.1, Claude Sonnet 4 with 1M context", "GPT-5 is OpenAI's latest unified model combining reasoning and fast response", "Claude Opus 4.1 is Anthropic's most intelligent model with hybrid reasoning", "Claude Sonnet 4 supports up to 1M token context (beta, premium pricing >200K)", "Prices are subject to change by LLM providers", "Categories: latest (newest models), budget (cost-effective), specialized (specific use cases), legacy (older models)", "Capabilities: chat, code, analysis, reasoning, fast, multimodal, vision, audio, agentic, hybrid_reasoning, computer_use", "Speed ratings: fast, medium, slow", "Quality ratings: excellent, good, basic", "Both providers now offer significant batch processing discounts (50% for Anthropic)", "Anthropic offers prompt caching for cost optimization", "Some Claude 3 models (opus-20240229, sonnet-20240229) are fully deprecated", "OpenAI has largely moved to GPT-5 family as primary offering" ], "category_descriptions": { "latest": "Newest models with cutting-edge capabilities", "budget": "Cost-effective options for high-volume use", "specialized": "Models optimized for specific use cases (reasoning, multimodal)", "legacy": "Older models, some deprecated" }, "provider_highlights": { "openai": { "flagship": "GPT-5 family", "key_features": ["Unified reasoning + fast response", "272K context", "Aggressive pricing"], "strengths": ["Coding", "Agentic tasks", "Cost-effectiveness"] }, "anthropic": { "flagship": "Claude Opus 4.1 & Claude Sonnet 4", "key_features": ["Hybrid reasoning", "1M context (Sonnet 4)", "Computer use"], "strengths": ["Advanced reasoning", "Long context", "Safety focus"] } } } }