{"data": [{"id": "route-llm-code", "name": "route-llm-code", "display_name": "RouteLLM (Code)", "description": "CodeLLM's agent router: classifies each turn by complexity and routes it to GPT-6 Luna, GLM, Sonnet or Opus. Built for coding agents, unlike route-llm. Non-Pro callers are served the low-cost ladder instead. Rates shown are the ladder's most common destination; you are billed for the model each turn actually resolves to.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.00001", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "abacus", "cached_input_token_rate": "0.0000001", "api_formats": ["openai", "anthropic"], "agentic": true, "code_agent": true, "tools": true}, {"id": "route-llm-code-low", "name": "route-llm-code-low", "display_name": "RouteLLM (Code, Low)", "description": "The same classifier on the low-cost ladder (Smaug Flash, or Qwen3.8 Flash Next for image turns). The default lane for non-Pro callers. Rates shown are that ladder's destination; you are billed for the model each turn actually resolves to.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.0000001", "output_token_rate": "0.0000004", "context_length": 1000000, "max_completion_tokens": 384000, "provider": "abacus", "cached_input_token_rate": "0.000000005", "api_formats": ["openai", "anthropic"], "agentic": true, "code_agent": true, "tools": true}, {"id": "route-llm", "name": "route-llm", "display_name": "RouteLLM", "description": "RouteLLM is a model that routes the user's message to the appropriate text-generation model. Routing is prompt-cache aware, so follow-up turns keep hitting the provider's prompt cache. We will route to one of Sonnet 5, GPT-5.6 Terra, Grok 4.6, or Gemini 3.7 Flash models. The price shown is for the most expensive model (Sonnet 5) that we use.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.00001", "context_length": null, "max_completion_tokens": 64000, "provider": "abacus", "cutoff_date": "2026-04-30", "cached_input_token_rate": "0.0000001", "api_formats": ["openai"], "featured": true, "top_model": true, "tools": true, "code_agent": true}, {"id": "gpt-6.1-sol", "name": "gpt-6.1-sol", "display_name": "GPT-6.1 Sol", "description": "GPT-6.1 Sol from OpenAI is a flagship model for complex reasoning, coding and agentic work, with a 1M context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.00001", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2026-04-30", "cached_input_token_rate": "0.0000001", "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "gpt-6.1-terra", "name": "gpt-6.1-terra", "display_name": "GPT-6.1 Terra", "description": "GPT-6.1 Terra from OpenAI is a cost-efficient mid-tier model for everyday reasoning, coding and agentic work, with a 1M context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000001", "output_token_rate": "0.000005", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2026-06-01", "cached_input_token_rate": "0.00000005", "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "gpt-6-astra", "name": "gpt-6-astra", "display_name": "GPT-6 Astra", "description": "GPT-6 Astra from OpenAI is its most capable model, built for the hardest end-to-end work: complex reasoning, coding, computer use, research, and document creation, with a 1M context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00001", "output_token_rate": "0.00005", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2026-04-30", "cached_input_token_rate": "0.000001", "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "gpt-6-sol", "name": "gpt-6-sol", "display_name": "GPT-6 Sol", "description": "GPT-6 Sol from OpenAI is a flagship model for complex reasoning, coding and agentic work, with a 1M context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.00001", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2026-04-20", "cached_input_token_rate": "0.0000002", "thinking": true, "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true}, {"id": "gpt-6-luna", "name": "gpt-6-luna", "display_name": "GPT-6 Luna", "description": "GPT-6 Luna from OpenAI is a fast, cost-efficient model for high-volume chat, summarization and lightweight agentic tasks, with a 1M context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.0000001", "output_token_rate": "0.0000005", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2026-05-18", "cached_input_token_rate": "0.00000001", "thinking": true, "api_formats": ["openai", "responses", "decisions"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true}, {"id": "gpt-5.5", "name": "gpt-5.5", "display_name": "GPT-5.5", "description": "GPT-5.5 from OpenAI is a flagship model with advanced reasoning capabilities and a 1M context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000005", "output_token_rate": "0.00003", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2025-08-31", "cached_input_token_rate": "0.0000005", "thinking": true, "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "agentic": true, "tools": true}, {"id": "gpt-5.4", "name": "gpt-5.4", "display_name": "GPT-5.4", "description": "GPT-5.4", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.0000025", "output_token_rate": "0.000015", "context_length": 400000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2025-08-31", "cached_input_token_rate": "0.00000025", "thinking": true, "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "agentic": true, "tools": true, "code_agent": true}, {"id": "gpt-5.4-mini", "name": "gpt-5.4-mini", "display_name": "GPT-5.4 Mini", "description": "GPT-5.4 Mini", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000075", "output_token_rate": "0.0000045", "context_length": 400000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2025-08-31", "cached_input_token_rate": "0.000000075", "thinking": true, "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "tools": true, "code_agent": true}, {"id": "gpt-5.4-nano", "name": "gpt-5.4-nano", "display_name": "GPT-5.4 Nano", "description": "GPT-5.4 Nano", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.0000002", "output_token_rate": "0.00000125", "context_length": 400000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2025-08-31", "cached_input_token_rate": "0.00000002", "api_formats": ["openai"], "featured": true, "top_model": true, "tools": true}, {"id": "claude-sonnet-5-5", "name": "claude-sonnet-5-5", "display_name": "Claude Sonnet 5.5", "description": "Claude Sonnet 5.5 from Anthropic is a step-change improvement over Sonnet 5, performing great on well-scoped everyday tasks like building features, fixing bugs, and creating polished documents, slides, and spreadsheets, and writing more clearly than Sonnet 5.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.00001", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2026-06-01", "cached_input_token_rate": "0.0000001", "api_formats": ["openai", "anthropic"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "claude-sonnet-5", "name": "claude-sonnet-5", "display_name": "Claude Sonnet 5", "description": "Claude Sonnet 5 from Anthropic is a balanced AI model, offering excellent performance in coding, reasoning, and general tasks with improved efficiency. It provides a strong balance between capability and cost, making it ideal for most use cases.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.00001", "context_length": 1000000, "max_completion_tokens": 64000, "provider": "anthropic", "cutoff_date": "2025-08-01", "thinking": true, "api_formats": ["openai", "anthropic"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true}, {"id": "gemini-3.8-flash", "name": "gemini-3.8-flash", "display_name": "Gemini 3.8 Flash", "description": "Gemini 3.8 Flash from Google is its most capable Flash model yet, with significant gains over Gemini 3.7 Flash in software engineering, long-horizon agentic tasks, and multi-step reasoning in specialized domains at Flash speed and cost.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio", "video"], "output_modalities": ["text"], "input_token_rate": "0.00000075", "output_token_rate": "0.00000375", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2026-03-01", "cached_input_token_rate": "0.000000075", "thinking": true, "api_formats": ["openai"], "featured": true, "top_model": true, "tools": true}, {"id": "grok-4.7", "name": "grok-4.7", "display_name": "Grok 4.7", "description": "Grok 4.7, the latest frontier multimodal reasoning model from xAI with a 500K context window built for coding, agentic tasks, and knowledge work. Reasoning is always on at the default effort.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.000006", "context_length": 500000, "max_completion_tokens": 32768, "provider": "xai", "cutoff_date": "2026-05-01", "cached_input_token_rate": "0.0000005", "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "agentic": true, "tools": true, "code_agent": true}, {"id": "claude-fable-5-2", "name": "claude-fable-5-2", "display_name": "Claude Fable 5.2", "description": "Claude Fable 5.2 from Anthropic is the most capable Claude model for technical work, setting a new bar in coding, finance and the sciences. It plans, orchestrates subagents and runs autonomously on long multi-step problems, and is faster than Fable 5.1.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000008", "output_token_rate": "0.00004", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2026-01-01", "cached_input_token_rate": "0.0000002", "api_formats": ["openai", "anthropic"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "claude-fable-5-1", "name": "claude-fable-5-1", "display_name": "Claude Fable 5.1", "description": "Claude Fable 5.1 from Anthropic is the next-generation Claude model, delivering major advances in coding, reasoning, and agentic tasks. It excels in tasks requiring long, complex reasoning chains, but is also expensive.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00001", "output_token_rate": "0.00005", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2026-01-01", "cached_input_token_rate": "0.00000025", "api_formats": ["openai", "anthropic"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "gemini-4-carbon", "name": "gemini-4-carbon", "display_name": "Gemini 4 Carbon", "description": "Gemini 4 Carbon", "model_type": "text_generation", "input_modalities": ["text", "image", "audio"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.00001", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.0000001", "api_formats": ["openai"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "gemini-4-argon", "name": "gemini-4-argon", "display_name": "Gemini 4 Argon", "description": "Gemini 4 Argon", "model_type": "text_generation", "input_modalities": ["text", "image", "audio"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.00001", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.0000001", "api_formats": ["openai"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "gemini-3.1-pro-preview", "name": "gemini-3.1-pro-preview", "display_name": "Gemini 3.1 Pro", "description": "Gemini 3.1 Pro", "model_type": "text_generation", "input_modalities": ["text", "image", "audio"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.000012", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.0000002", "api_formats": ["openai"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "claude-opus-5-5", "name": "claude-opus-5-5", "display_name": "Claude Opus 5.5", "description": "Claude Opus 5.5 from Anthropic is a step-change improvement over Opus 5, with the biggest gains in agentic coding, long-running agentic tasks, and knowledge work. It excels in tasks requiring long, complex reasoning chains, but is also expensive.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000004", "output_token_rate": "0.00002", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2026-06-01", "cached_input_token_rate": "0.0000002", "api_formats": ["openai", "anthropic"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "claude-opus-5", "name": "claude-opus-5", "display_name": "Claude Opus 5", "description": "Claude Opus 5 from Anthropic is a highly capable AI model, focused on advanced coding, reasoning, and agentic tasks, with improved performance in multi-file code refactoring, debugging, and detail-oriented reasoning. It excels in tasks requiring long, complex reasoning chains, but is also expensive.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000005", "output_token_rate": "0.000025", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2026-01-01", "thinking": true, "api_formats": ["openai", "anthropic"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true}, {"id": "muse-wonder-1.0", "name": "muse-wonder-1.0", "display_name": "Muse Wonder 1.0", "description": "Muse Wonder is Meta's frontier model for coding, long-running agents, and advanced computer use: natively multimodal with a 1M token context window and text, image, video, audio, and PDF inputs, it plans long tasks, delegates work to sub-agents, and recovers from errors, with native tool use and parallel tool calling.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio", "video"], "output_modalities": ["text"], "input_token_rate": "0.000004", "output_token_rate": "0.00002", "context_length": 1048576, "max_completion_tokens": 131072, "provider": "meta", "cutoff_date": "2026-01-01", "cached_input_token_rate": "0.0000002", "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "tools": true, "code_agent": true}, {"id": "muse-spark-1.3", "name": "muse-spark-1.3", "display_name": "Muse Spark 1.3", "description": "Muse Spark 1.3 is Meta's natively multimodal reasoning model with a 1M token context window and text, image, video, audio, and PDF inputs, tuned for long-running agentic, multi-agent, and coding workflows; it needs roughly 20% fewer tool calls and 25% fewer tokens than Muse Spark 1.2, with native tool use and parallel tool calling.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio", "video"], "output_modalities": ["text"], "input_token_rate": "0.00000125", "output_token_rate": "0.00000425", "context_length": 1048576, "max_completion_tokens": 131072, "provider": "meta", "cutoff_date": "2026-01-01", "cached_input_token_rate": "0.00000015", "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "tools": true, "code_agent": true}, {"id": "muse-spark-1.2", "name": "muse-spark-1.2", "display_name": "Muse Spark 1.2", "description": "Muse Spark 1.2 is Meta's natively multimodal reasoning model with a 1M token context window and text, image, video, audio, and PDF inputs, co-trained with the Muse Code agent for long-horizon coding, debugging, and whole-repository generation, with native tool use and parallel tool calling.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio", "video"], "output_modalities": ["text"], "input_token_rate": "0.00000125", "output_token_rate": "0.00000425", "context_length": 1048576, "max_completion_tokens": 131072, "provider": "meta", "cutoff_date": "2026-01-01", "cached_input_token_rate": "0.00000015", "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "tools": true, "code_agent": true}, {"id": "grok-4.3", "name": "grok-4.3", "display_name": "Grok 4.3", "description": "Grok 4.3, the latest frontier multimodal model from xAI with a 1M context window optimized for high-performance agentic tool calling. Reasoning is disabled.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000125", "output_token_rate": "0.0000025", "context_length": 1000000, "max_completion_tokens": 32768, "provider": "xai", "cutoff_date": "2025-12-01", "cached_input_token_rate": "0.0000002", "thinking": true, "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "tools": true, "code_agent": true}, {"id": "moonshotai/Kimi-K3", "name": "moonshotai/Kimi-K3", "display_name": "Kimi K3", "description": "Kimi K3 is Moonshot AI's flagship open-weight multimodal reasoning model with a 1M-token context window, built for long-horizon coding, agentic workflows, and end-to-end knowledge work.", "model_type": "text_generation", "input_modalities": ["text", "image", "video"], "output_modalities": ["text"], "input_token_rate": "0.000003", "output_token_rate": "0.000015", "context_length": 1048576, "max_completion_tokens": 131072, "provider": "moonshot", "cutoff_date": "2025-06-01", "cached_input_token_rate": "0.0000003", "api_formats": ["openai"], "featured": true, "top_model": true, "agentic": true, "tools": true, "code_agent": true}, {"id": "moonshotai/Kimi-K2.7-Code", "name": "moonshotai/Kimi-K2.7-Code", "display_name": "Kimi K2.7 Code", "description": "Kimi K2.7 Code is Kimi's most capable coding model to date. It follows instructions more reliably in long contexts and completes coding tasks with higher success rates.", "model_type": "text_generation", "input_modalities": ["text", "image", "video"], "output_modalities": ["text"], "input_token_rate": "0.00000095", "output_token_rate": "0.000004", "context_length": 262144, "max_completion_tokens": 262144, "provider": "moonshot", "cutoff_date": "2025-06-01", "cached_input_token_rate": "0.00000019", "api_formats": ["openai"], "featured": true, "top_model": true, "agentic": true, "tools": true, "code_agent": true}, {"id": "zai-org/GLM-5.3", "name": "zai-org/GLM-5.3", "display_name": "GLM 5.3", "description": "GLM-5.3 from Z.AI is a flagship model delivering major advances in complex software engineering and agent tasks over GLM-5.2, with a 1M token context window and strong coding, agentic, and cybersecurity capabilities.", "model_type": "text_generation", "input_modalities": ["text"], "output_modalities": ["text"], "input_token_rate": "0.0000014", "output_token_rate": "0.0000044", "context_length": 1048576, "max_completion_tokens": 131072, "provider": "zai", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.00000026", "api_formats": ["openai"], "featured": true, "top_model": true, "agentic": true, "tools": true, "code_agent": true}, {"id": "zai-org/GLM-5.3-Flash", "name": "zai-org/GLM-5.3-Flash", "display_name": "GLM 5.3 Flash", "description": "GLM-5.3 Flash from Z.AI is an efficient variant of GLM-5.3 optimized for coding and long-horizon agent tasks, with a 1M token context window and 131K max output tokens at significantly lower cost.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000015", "output_token_rate": "0.0000005", "context_length": 1048576, "max_completion_tokens": 131072, "provider": "zai", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.00000003", "api_formats": ["openai"], "featured": true, "top_model": true, "agentic": true, "tools": true, "code_agent": true}, {"id": "deepseek-ai/DeepSeek-V4.1-Flash-1017", "name": "deepseek-ai/DeepSeek-V4.1-Flash-1017", "display_name": "Deepseek V4.1 Flash", "description": "DeepSeek V4.1 Flash is the multimodal successor to DeepSeek V4 Flash developed by DeepSeek: a fast, low-cost model for text, agents and reasoning that also understands images (JPEG/PNG/GIF/WebP), with a 1M-token context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000022", "output_token_rate": "0.00000066", "context_length": 1000000, "max_completion_tokens": 384000, "provider": "deepseek", "cutoff_date": "2026-04-01", "cached_input_token_rate": "0.000000007", "thinking": true, "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "agentic": true, "tools": true, "code_agent": true}, {"id": "deepseek-ai/DeepSeek-V4-Pro-0813", "name": "deepseek-ai/DeepSeek-V4-Pro-0813", "display_name": "Deepseek V4 Pro", "description": "DeepSeek V4 Pro is an advanced, high-performance large language model (LLM) developed by DeepSeek.", "model_type": "text_generation", "input_modalities": ["text"], "output_modalities": ["text"], "input_token_rate": "0.00000132", "output_token_rate": "0.00000396", "context_length": 1000000, "max_completion_tokens": 384000, "provider": "deepseek", "cutoff_date": "2026-04-01", "cached_input_token_rate": "0.000000044", "thinking": true, "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "agentic": true, "tools": true}, {"id": "deepseek-ai/DeepSeek-V4-Flash-0731", "name": "deepseek-ai/DeepSeek-V4-Flash-0731", "display_name": "Deepseek V4 Flash 0731", "description": "DeepSeek V4 Flash is an advanced, high-performance large language model (LLM) developed by DeepSeek.", "model_type": "text_generation", "input_modalities": ["text"], "output_modalities": ["text"], "input_token_rate": "0.00000044", "output_token_rate": "0.00000132", "context_length": 1000000, "max_completion_tokens": 384000, "provider": "deepseek", "cutoff_date": "2026-04-01", "cached_input_token_rate": "0.000000028", "thinking": true, "api_formats": ["openai", "responses"], "featured": true, "top_model": true, "agentic": true, "tools": true, "code_agent": true}, {"id": "claude-haiku-5-5", "name": "claude-haiku-5-5", "display_name": "Claude Haiku 5.5", "description": "Claude Haiku 5.5 from Anthropic is the fastest and most affordable Claude model, built for high-volume, latency-sensitive tasks such as classification, extraction, routing and subagent work, with adaptive thinking and a 1M token context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.0000001", "output_token_rate": "0.0000005", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2026-06-01", "cached_input_token_rate": "0.00000001", "api_formats": ["openai", "anthropic"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "gemini-3.5-flash-lite", "name": "gemini-3.5-flash-lite", "display_name": "Gemini 3.5 Flash Lite", "description": "Gemini 3.5 Flash Lite from Google is the fastest and most cost-effective Gemini 3.5 series model, built for high-throughput, low-latency workloads like agentic search and document processing.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio"], "output_modalities": ["text"], "input_token_rate": "0.0000003", "output_token_rate": "0.0000025", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.00000003", "thinking": true, "api_formats": ["openai"], "featured": true, "top_model": true, "tools": true}, {"id": "gemini-3.1-flash-lite", "name": "gemini-3.1-flash-lite", "display_name": "Gemini 3.1 Flash Lite", "description": "Gemini 3.1 Flash Lite from Google is a fast and cost-efficient Gemini 3 series model, built for high-volume developer workloads at scale.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio"], "output_modalities": ["text"], "input_token_rate": "0.00000025", "output_token_rate": "0.0000015", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.000000025", "thinking": true, "api_formats": ["openai"], "featured": true, "top_model": true, "tools": true}, {"id": "gemini-3-flash-preview", "name": "gemini-3-flash-preview", "display_name": "Gemini 3 Flash", "description": "Gemini 3 Flash from Google is an advanced large language model. It is designed for efficiency while balancing reasoning capabilities.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio"], "output_modalities": ["text"], "input_token_rate": "0.0000005", "output_token_rate": "0.000003", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.00000005", "thinking": true, "api_formats": ["openai"], "featured": true, "top_model": true, "tools": true}, {"id": "claude-sonnet-4-6", "name": "claude-sonnet-4-6", "display_name": "Claude Sonnet 4.6", "description": "Claude Sonnet 4.6 from Anthropic is a balanced AI model, offering excellent performance in coding, reasoning, and general tasks with improved efficiency. It provides a strong balance between capability and cost, making it ideal for most use cases.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000003", "output_token_rate": "0.000015", "context_length": 1000000, "max_completion_tokens": 64000, "provider": "anthropic", "cutoff_date": "2025-08-01", "thinking": true, "api_formats": ["openai", "anthropic"], "featured": true, "top_model": true, "agentic": true, "computer_use": true, "tools": true}, {"id": "gemini-3.7-flash", "name": "gemini-3.7-flash", "display_name": "Gemini 3.7 Flash", "description": "Gemini 3.7 Flash from Google is an advanced multimodal reasoning model built for coding and agentic workflows that delivers better debugging, knowledge work, and multimodal performance with tunable thinking levels than Gemini 3.6 Flash.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio", "video"], "output_modalities": ["text"], "input_token_rate": "0.00000075", "output_token_rate": "0.00000375", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2026-03-01", "cached_input_token_rate": "0.000000075", "thinking": true, "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "thinkingmachines/Inkling", "name": "thinkingmachines/Inkling", "display_name": "Tinker Inkling", "description": "Inkling is Thinking Machines Lab's first open-weights model, a 975B-parameter mixture-of-experts (41B active) that reasons across text and images, served with a 262,144 token context window. Supports tools and streaming.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000374", "output_token_rate": "0.00000936", "context_length": 262144, "max_completion_tokens": 131072, "provider": "thinkingmachines", "cutoff_date": "2026-01-01", "cached_input_token_rate": "0.000000748", "thinking": true, "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "muse-spark-1.3-contributor", "name": "muse-spark-1.3-contributor", "display_name": "Muse Spark 1.3 Contributor", "description": "Muse Spark 1.3 Contributor is the contributor tier of Muse Spark 1.3, Meta's natively multimodal reasoning model with a 1M token context window and text, image, video, audio, and PDF inputs, with native tool use and parallel tool calling. It is the same model at heavily discounted rates in exchange for Meta using prompts and completions to train future Meta models.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio", "video"], "output_modalities": ["text"], "input_token_rate": "0.0000001", "output_token_rate": "0.0000002", "context_length": 1048576, "max_completion_tokens": 131072, "provider": "meta", "cutoff_date": "2026-01-01", "cached_input_token_rate": "0.000000002", "api_formats": ["openai", "responses"], "featured": false, "tools": true, "code_agent": true}, {"id": "mimo-v2.6-pro", "name": "mimo-v2.6-pro", "display_name": "MiMo V2.6 Pro", "description": "Xiaomi MiMo V2.6 Pro is Xiaomi's trillion-parameter open-weight flagship reasoning model with a 1M token context window, 128K max output, and native image input, tuned for agentic coding, research and visual tasks with tool use and structured output.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000000435", "output_token_rate": "0.00000087", "context_length": 1048576, "max_completion_tokens": 131072, "provider": "xiaomi", "cutoff_date": "2024-12-01", "cached_input_token_rate": "0.0000000036", "api_formats": ["openai"], "featured": false, "tools": true, "code_agent": true}, {"id": "elorian-labs/glimpse-1.0", "name": "elorian-labs/glimpse-1.0", "display_name": "Glimpse 1.0", "description": "Glimpse 1.0 is Elorian's visual thinking model, which reasons directly over images rather than first translating them to text, with a 262,144 token context window. Reasoning runs at the highest effort by default. Supports tools and streaming.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.0000006", "output_token_rate": "0.000003", "context_length": 262144, "max_completion_tokens": 131072, "provider": "elorian", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.00000006", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "MiniMaxAI/MiniMax-M3.1-Flash", "name": "MiniMaxAI/MiniMax-M3.1-Flash", "display_name": "MiniMax M3.1 Flash", "description": "MiniMax M3.1 Flash is MiniMax's fast, low-cost reasoning model with a 1,000,000 token context window, taking text and image input. Reasoning is always on and runs at MiniMax's max effort by default. Supports tools and streaming; for schema-bound output use a tool call, as MiniMax ignores response_format.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000015", "output_token_rate": "0.0000006", "context_length": 1000000, "max_completion_tokens": 131072, "provider": "minimax", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.000000015", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "MiniMaxAI/MiniMax-M3", "name": "MiniMaxAI/MiniMax-M3", "display_name": "MiniMax M3", "description": "MiniMax M3 is MiniMax's latest flagship reasoning model with a 1,000,000 token context window and 131,072 max output tokens. Supports tools, structured output, and reasoning.", "model_type": "text_generation", "input_modalities": ["text", "image", "video"], "output_modalities": ["text"], "input_token_rate": "0.0000003", "output_token_rate": "0.0000012", "context_length": 1000000, "max_completion_tokens": 131072, "provider": "minimax", "cutoff_date": "2025-01-01", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "google/gemma-4-31b-it", "name": "google/gemma-4-31b-it", "display_name": "Gemma 4 31B IT", "description": "Gemma 4 31B IT is Google's open-weight vision-language model with a 256K context window. It supports interleaved text and image inputs, structured output, function calling, and reasoning tasks.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000014", "output_token_rate": "0.0000004", "context_length": 262144, "max_completion_tokens": 131072, "provider": "google", "cutoff_date": "2025-01-01", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "moonshotai/Kimi-K3.1-Pro", "name": "moonshotai/Kimi-K3.1-Pro", "display_name": "Kimi K3.1 Pro", "description": "Kimi K3.1 Pro is Moonshot AI's most capable open-weight multimodal reasoning model, a post-trained Kimi K3 with a 1M-token context window, built for coding, 3D creation, financial research, and long-horizon agentic work.", "model_type": "text_generation", "input_modalities": ["text", "image", "video"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.00001", "context_length": 1048576, "max_completion_tokens": 131072, "provider": "moonshot", "cutoff_date": "2025-06-01", "cached_input_token_rate": "0.0000002", "api_formats": ["openai", "responses"], "featured": false, "agentic": true, "tools": true}, {"id": "moonshotai/Kimi-K2.6", "name": "moonshotai/Kimi-K2.6", "display_name": "Kimi K2.6", "description": "Kimi K2.6 is a natively multimodal model with powerful coding capabilities and enhanced agent performance. It builds on the K2.5 foundation with improved reasoning depth, cleaner agent planning, and faster multi-step tool call reliability.", "model_type": "text_generation", "input_modalities": ["text", "image", "video"], "output_modalities": ["text"], "input_token_rate": "0.00000095", "output_token_rate": "0.000004", "context_length": 262144, "max_completion_tokens": 262144, "provider": "moonshot", "cutoff_date": "2025-06-01", "cached_input_token_rate": "0.00000016", "thinking": true, "api_formats": ["openai"], "featured": false, "tools": true, "code_agent": true}, {"id": "grok-4.6", "name": "grok-4.6", "display_name": "Grok 4.6", "description": "Grok 4.6, the latest frontier multimodal model from xAI with a 500K context window built for coding, agentic tasks, and knowledge work. Reasoning is disabled.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.000006", "context_length": 500000, "max_completion_tokens": 32768, "provider": "xai", "cutoff_date": "2026-02-01", "cached_input_token_rate": "0.0000005", "api_formats": ["openai", "responses"], "featured": false, "agentic": true, "tools": true, "code_agent": true}, {"id": "grok-4.5", "name": "grok-4.5", "display_name": "Grok 4.5", "description": "Grok 4.5, the latest frontier multimodal model from xAI with a 500K context window optimized for high-performance agentic tool calling. Reasoning is disabled.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.000006", "context_length": 500000, "max_completion_tokens": 32768, "provider": "xai", "cutoff_date": "2026-02-01", "cached_input_token_rate": "0.0000003", "api_formats": ["openai", "responses"], "featured": false, "agentic": true, "tools": true, "code_agent": true}, {"id": "Qwen/Qwen3.8-Flash-Next", "name": "Qwen/Qwen3.8-Flash-Next", "display_name": "Qwen3.8 Flash Next", "description": "Qwen3.8 Flash Next is a 125B-parameter MoE model (6B active) from Alibaba's Qwen series, featuring hybrid attention (Gated DeltaNet + Qwen Sparse Attention) for ultra-efficient inference, native image and video input, a 1M token context window, and built-in configurable thinking.", "model_type": "text_generation", "input_modalities": ["text", "image", "video"], "output_modalities": ["text"], "input_token_rate": "0.00000015", "output_token_rate": "0.00000047", "context_length": 1000000, "max_completion_tokens": 131072, "provider": "qwen", "cutoff_date": "2025-07-01", "cached_input_token_rate": "0.000000016", "api_formats": ["openai"], "featured": false, "tools": true, "code_agent": true}, {"id": "Qwen/Qwen3.8-27B", "name": "Qwen/Qwen3.8-27B", "display_name": "Qwen3.8 27B", "description": "Qwen3.8 27B is an open-weight dense 27B-parameter vision-language model from Alibaba's Qwen3.8 series, with native image and video input, a 262K context window (1M with YaRN), and default-on configurable thinking. It is optimized for coding, research, and long-horizon agentic workflows.", "model_type": "text_generation", "input_modalities": ["text", "image", "video"], "output_modalities": ["text"], "input_token_rate": "0.00000045", "output_token_rate": "0.0000032", "context_length": 262144, "max_completion_tokens": 131072, "provider": "qwen", "cutoff_date": "2025-07-01", "thinking": true, "api_formats": ["openai"], "featured": false, "tools": true, "code_agent": true}, {"id": "qwen3.8-max", "name": "qwen3.8-max", "display_name": "Qwen3.8 Max", "description": "Qwen3.8-Max is Alibaba's flagship 2.4-trillion-parameter MoE model with native multimodal understanding and a 1M token context window. It excels at long-horizon agentic tasks, advanced coding, and complex reasoning with configurable thinking effort.", "model_type": "text_generation", "input_modalities": ["text", "image", "video"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.000006", "context_length": 262144, "max_completion_tokens": 131072, "provider": "qwen", "cutoff_date": "2025-07-01", "cached_input_token_rate": "0.00000025", "thinking": true, "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "abacusai/Smaug-Flash", "name": "abacusai/Smaug-Flash", "display_name": "Smaug Flash", "description": "Smaug Flash is an Abacus.AI finetune of DeepSeek V4 Flash, tuned for agentic and coding tasks.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.0000001", "output_token_rate": "0.0000004", "context_length": 1000000, "max_completion_tokens": 384000, "provider": "abacus", "cutoff_date": "2026-04-01", "cached_input_token_rate": "0.000000005", "thinking": true, "api_formats": ["openai", "responses"], "featured": false, "tools": true}, {"id": "gemini-3.6-flash", "name": "gemini-3.6-flash", "display_name": "Gemini 3.6 Flash", "description": "Gemini 3.6 Flash from Google is an advanced multimodal reasoning model that delivers better coding, knowledge work, and multimodal performance with greater token efficiency than Gemini 3.5 Flash.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio"], "output_modalities": ["text"], "input_token_rate": "0.00000075", "output_token_rate": "0.00000375", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2026-03-01", "cached_input_token_rate": "0.000000075", "thinking": true, "api_formats": ["openai"], "featured": false, "agentic": true, "tools": true}, {"id": "gemini-3.5-flash", "name": "gemini-3.5-flash", "display_name": "Gemini 3.5 Flash", "description": "Gemini 3.5 Flash from Google is an advanced multimodal model designed for high-quality reasoning and efficiency. It offers strong performance across text, image, audio, and video tasks.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio"], "output_modalities": ["text"], "input_token_rate": "0.0000015", "output_token_rate": "0.000009", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.00000015", "thinking": true, "api_formats": ["openai"], "featured": false, "agentic": true, "tools": true}, {"id": "gemini-nano-banana-2.1", "name": "gemini-nano-banana-2.1", "display_name": "Nano Banana 2.1", "description": "Nano Banana 2.1", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text", "image"], "input_token_rate": "0.0000015", "output_token_rate": "0.0000075", "context_length": 65536, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.00000015", "api_formats": ["openai"], "featured": false}, {"id": "gemini-3-pro-image", "name": "gemini-3-pro-image", "display_name": "Nano Banana (Gemini 3 Pro Image)", "description": "Nano Banana (Gemini 3 Pro Image)", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text", "image"], "input_token_rate": "0.000002", "output_token_rate": "0.000012", "context_length": 65536, "max_completion_tokens": 32768, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.0000002", "api_formats": ["openai"], "featured": false}, {"id": "gemini-2.5-flash-image", "name": "gemini-2.5-flash-image", "display_name": "Nano Banana (Gemini 2.5 Flash Image)", "description": "Nano Banana (Gemini 2.5 Flash Image)", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text", "image"], "input_token_rate": "0.0000003", "output_token_rate": "0.00003", "context_length": 32768, "max_completion_tokens": 32768, "provider": "google", "cutoff_date": "2025-01-01", "api_formats": ["openai"], "featured": false}, {"id": "gemini-2.5-flash", "name": "gemini-2.5-flash", "display_name": "Gemini 2.5 Flash", "description": "Gemini 2.5 Flash is a large multimodal language model from Google. It is designed for speed, efficiency, and cost-effectiveness in AI tasks that require high volume and frequency.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio"], "output_modalities": ["text"], "input_token_rate": "0.0000003", "output_token_rate": "0.0000025", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.00000003", "thinking": true, "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "gemini-2.5-pro", "name": "gemini-2.5-pro", "display_name": "Gemini 2.5 Pro", "description": "Gemini 2.5 Pro is Google's advanced large language model. It is designed for complex tasks that require deep reasoning, advanced coding, and multimodal understanding.", "model_type": "text_generation", "input_modalities": ["text", "image", "audio"], "output_modalities": ["text"], "input_token_rate": "0.00000125", "output_token_rate": "0.00001", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "cached_input_token_rate": "0.000000125", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "claude-opus-5-5-fast", "name": "claude-opus-5-5-fast", "display_name": "Claude Opus 5.5", "description": "Claude Opus 5.5", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000008", "output_token_rate": "0.00004", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2026-06-01", "cached_input_token_rate": "0.0000004", "api_formats": ["openai", "anthropic"], "featured": false, "agentic": true, "computer_use": true, "tools": true}, {"id": "claude-fable-5", "name": "claude-fable-5", "display_name": "Claude Fable 5", "description": "Claude Fable 5 from Anthropic is the next-generation Claude model, delivering major advances in coding, reasoning, and agentic tasks. It excels in tasks requiring long, complex reasoning chains, but is also expensive.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00001", "output_token_rate": "0.00005", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2026-01-01", "api_formats": ["openai", "anthropic"], "featured": false, "agentic": true, "computer_use": true, "tools": true, "code_agent": true}, {"id": "claude-opus-4-8", "name": "claude-opus-4-8", "display_name": "Claude Opus 4.8", "description": "Claude Opus 4.8 from Anthropic is a highly capable AI model, focused on advanced coding, reasoning, and agentic tasks, with improved performance in multi-file code refactoring, debugging, and detail-oriented reasoning. It excels in tasks requiring long, complex reasoning chains, but is also expensive.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000005", "output_token_rate": "0.000025", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2026-01-01", "thinking": true, "api_formats": ["openai", "anthropic"], "featured": false, "agentic": true, "computer_use": true, "tools": true}, {"id": "claude-opus-4-7", "name": "claude-opus-4-7", "display_name": "Claude Opus 4.7", "description": "Claude Opus 4.7 from Anthropic is a highly capable AI model, focused on advanced coding, reasoning, and agentic tasks, with improved performance in multi-file code refactoring, debugging, and detail-oriented reasoning. It excels in tasks requiring long, complex reasoning chains, but is also expensive.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000005", "output_token_rate": "0.000025", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2026-01-01", "thinking": true, "api_formats": ["openai", "anthropic"], "featured": false, "agentic": true, "computer_use": true, "tools": true}, {"id": "claude-opus-4-6", "name": "claude-opus-4-6", "display_name": "Claude Opus 4.6", "description": "Claude Opus 4.6 from Anthropic is a highly capable AI model, focused on advanced coding, reasoning, and agentic tasks, with improved performance in multi-file code refactoring, debugging, and detail-oriented reasoning. It excels in tasks requiring long, complex reasoning chains, but is also expensive.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000005", "output_token_rate": "0.000025", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "anthropic", "cutoff_date": "2025-05-01", "thinking": true, "api_formats": ["openai", "anthropic"], "featured": false, "agentic": true, "computer_use": true, "tools": true}, {"id": "claude-opus-4-5-20251101", "name": "claude-opus-4-5-20251101", "display_name": "Claude Opus 4.5", "description": "Claude Opus 4.5 from Anthropic is an AI model focused on advanced coding, reasoning, and agentic tasks, with improved performance in multi-file code refactoring, debugging, and detail-oriented reasoning. It excels in tasks requiring long, complex reasoning chains, but is also expensive.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000005", "output_token_rate": "0.000025", "context_length": 200000, "max_completion_tokens": 64000, "provider": "anthropic", "cutoff_date": "2025-03-01", "thinking": true, "api_formats": ["openai", "anthropic"], "featured": false, "tools": true}, {"id": "claude-haiku-4-5-20251001", "name": "claude-haiku-4-5-20251001", "display_name": "Claude Haiku 4.5", "description": "Claude Haiku 4.5", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000001", "output_token_rate": "0.000005", "context_length": 200000, "max_completion_tokens": 64000, "provider": "anthropic", "cutoff_date": "2025-02-01", "thinking": true, "api_formats": ["openai", "anthropic"], "featured": false, "tools": true, "code_agent": true}, {"id": "claude-sonnet-4-5-20250929", "name": "claude-sonnet-4-5-20250929", "display_name": "Claude Sonnet 4.5", "description": "Claude Sonnet 4.5 advances over its predecessor, Sonnet 4, delivering stronger coding and reasoning performance with greater precision and control. Makes agentic programming better. It excels in tasks requiring long, complex reasoning chains, but is also expensive.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000003", "output_token_rate": "0.000015", "context_length": 200000, "max_completion_tokens": 64000, "provider": "anthropic", "cutoff_date": "2025-01-01", "thinking": true, "api_formats": ["openai", "anthropic"], "featured": false, "tools": true, "code_agent": true}, {"id": "openai/gpt-oss-120b", "name": "openai/gpt-oss-120b", "display_name": "GPT-OSS 120B", "description": "GPT-OSS from OpenAI is a powerful, open-weight language model designed for high-reasoning, agentic tasks. It supports advanced capabilities like configurable reasoning effort levels, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation.", "model_type": "text_generation", "input_modalities": ["text"], "output_modalities": ["text"], "input_token_rate": "0.000000037", "output_token_rate": "0.00000017", "context_length": 128000, "max_completion_tokens": 16384, "provider": "openai", "cutoff_date": "2024-06-01", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "gpt-5.6-luna", "name": "gpt-5.6-luna", "display_name": "GPT-5.6 Luna", "description": "GPT-5.6 Luna from OpenAI is a flagship model with advanced reasoning capabilities and a 1M context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.0000002", "output_token_rate": "0.0000012", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2026-02-16", "cached_input_token_rate": "0.00000002", "thinking": true, "api_formats": ["openai", "responses"], "featured": false, "agentic": true, "computer_use": true, "tools": true}, {"id": "gpt-5.6-terra", "name": "gpt-5.6-terra", "display_name": "GPT-5.6 Terra", "description": "GPT-5.6 Terra from OpenAI is a flagship model with advanced reasoning capabilities and a 1M context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.000012", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2026-02-16", "cached_input_token_rate": "0.0000002", "thinking": true, "api_formats": ["openai", "responses"], "featured": false, "agentic": true, "computer_use": true, "tools": true}, {"id": "gpt-5.6-sol", "name": "gpt-5.6-sol", "display_name": "GPT-5.6 Sol", "description": "GPT-5.6 Sol from OpenAI is a flagship model with advanced reasoning capabilities and a 1M context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000004", "output_token_rate": "0.00002", "context_length": 1000000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2026-02-16", "cached_input_token_rate": "0.0000004", "thinking": true, "api_formats": ["openai", "responses"], "featured": false, "agentic": true, "computer_use": true, "tools": true}, {"id": "gpt-5.3-codex", "name": "gpt-5.3-codex", "display_name": "GPT-5.3 Codex", "description": "GPT-5.3 Codex", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000175", "output_token_rate": "0.000014", "context_length": 400000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2025-08-31", "cached_input_token_rate": "0.000000175", "api_formats": ["openai", "responses"], "featured": false, "tools": true, "code_agent": true}, {"id": "gpt-5.2", "name": "gpt-5.2", "display_name": "GPT-5.2", "description": "GPT-5.2", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000175", "output_token_rate": "0.000014", "context_length": 400000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2025-08-31", "cached_input_token_rate": "0.000000175", "thinking": true, "api_formats": ["openai", "responses"], "featured": false, "tools": true}, {"id": "gpt-5.1", "name": "gpt-5.1", "display_name": "GPT-5.1", "description": "GPT-5.1", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000125", "output_token_rate": "0.00001", "context_length": 400000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2024-09-30", "cached_input_token_rate": "0.000000125", "thinking": true, "api_formats": ["openai", "responses"], "featured": false, "tools": true}, {"id": "gpt-5-nano", "name": "gpt-5-nano", "display_name": "GPT-5 Nano", "description": "GPT-5 Nano is the fastest and cheapest model in the GPT-5 family. It is a good model for summarization and classification tasks.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000005", "output_token_rate": "0.0000004", "context_length": 400000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2024-09-30", "cached_input_token_rate": "0.000000005", "api_formats": ["openai", "responses"], "featured": false, "tools": true}, {"id": "gpt-5-mini", "name": "gpt-5-mini", "display_name": "GPT-5 Mini", "description": "GPT-5 Mini is a streamlined variant of GPT-5 built for lighter reasoning workloads. It delivers GPT-5 instruction adherence and safety tuning while offering lower latency and cost.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000025", "output_token_rate": "0.000002", "context_length": 400000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2024-09-30", "cached_input_token_rate": "0.000000025", "api_formats": ["openai", "responses"], "featured": false, "tools": true}, {"id": "gpt-5", "name": "gpt-5", "display_name": "GPT-5", "description": "GPT-5 from OpenAI is an advanced model, delivering significant gains in reasoning, code generation quality, and overall user experience. It is a very good model for coding and agentic tasks across industries.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000125", "output_token_rate": "0.00001", "context_length": 400000, "max_completion_tokens": 128000, "provider": "openai", "cutoff_date": "2024-09-30", "cached_input_token_rate": "0.000000125", "api_formats": ["openai", "responses"], "featured": false, "tools": true}, {"id": "gpt-4.1-mini", "name": "gpt-4.1-mini", "display_name": "GPT-4.1 Mini", "description": "GPT-4.1 Mini is a smaller, faster, and cheaper version of GPT-4.1, designed for lightweight tasks and lower latency requirements.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.0000004", "output_token_rate": "0.0000016", "context_length": 1047576, "max_completion_tokens": 32768, "provider": "openai", "cutoff_date": "2024-05-31", "cached_input_token_rate": "0.0000001", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "gpt-4.1", "name": "gpt-4.1", "display_name": "GPT-4.1", "description": "GPT-4.1 is designed to be faster, cheaper, and more reliable for specific tasks compared to the o-series models, especially for coding and tool-based operations.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.000002", "output_token_rate": "0.000008", "context_length": 1047576, "max_completion_tokens": 32768, "provider": "openai", "cutoff_date": "2024-05-31", "cached_input_token_rate": "0.0000005", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "gpt-realtime-whisper", "name": "gpt-realtime-whisper", "display_name": "GPT Realtime Whisper", "description": "GPT Realtime Whisper", "model_type": "text_generation", "input_modalities": ["text"], "output_modalities": ["text"], "input_token_rate": "0.0002833333", "output_token_rate": "0", "context_length": 16000, "max_completion_tokens": 2000, "provider": "openai", "cutoff_date": "2025-06-01", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "gpt-transcribe", "name": "gpt-transcribe", "display_name": "GPT Transcribe", "description": "GPT Transcribe", "model_type": "text_generation", "input_modalities": ["text"], "output_modalities": ["text"], "input_token_rate": "0.000075", "output_token_rate": "0", "context_length": 16000, "max_completion_tokens": 2000, "provider": "openai", "cutoff_date": "2025-06-01", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "gpt-4o-transcribe", "name": "gpt-4o-transcribe", "display_name": "GPT 4o Transcribe", "description": "GPT 4o Transcribe", "model_type": "text_generation", "input_modalities": ["text"], "output_modalities": ["text"], "input_token_rate": "0.0000025", "output_token_rate": "0.00001", "context_length": 16000, "max_completion_tokens": 2000, "provider": "openai", "cutoff_date": "2024-09-30", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "gpt-4o-mini-transcribe", "name": "gpt-4o-mini-transcribe", "display_name": "GPT 4o Mini Transcribe", "description": "GPT 4o Mini Transcribe", "model_type": "text_generation", "input_modalities": ["text"], "output_modalities": ["text"], "input_token_rate": "0.00000125", "output_token_rate": "0.000005", "context_length": 16000, "max_completion_tokens": 2000, "provider": "openai", "cutoff_date": "2024-09-30", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "gpt-4o-mini", "name": "gpt-4o-mini", "display_name": "GPT-4o Mini", "description": "GPT-4o mini is a highly efficient and cost-effective small language model. t excels in instruction following, multimodal reasoning, and supports a 128k context window.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.00000015", "output_token_rate": "0.0000006", "context_length": 128000, "max_completion_tokens": 16384, "provider": "openai", "cutoff_date": "2023-10-01", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "gpt-4o", "name": "gpt-4o", "display_name": "GPT-4o", "description": "GPT-4o from OpenAI is an advanced multimodal AI model that can seamlessly process and generate content across text, images, and audio. This allows for more natural human-computer interaction, as users can engage with it through talking, typing, showing images, and even videos.", "model_type": "text_generation", "input_modalities": ["text", "image"], "output_modalities": ["text"], "input_token_rate": "0.0000025", "output_token_rate": "0.00001", "context_length": 128000, "max_completion_tokens": 16384, "provider": "openai", "cutoff_date": "2023-10-01", "cached_input_token_rate": "0.00000125", "api_formats": ["openai"], "featured": false, "tools": true}, {"id": "gpt_image15", "name": "gpt_image15", "display_name": "GPT Image 1.5", "description": "GPT Image 1.5", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "openai", "text_input_rate": "0.0005", "image_input_rate": "0.0008", "text_output_rate": "0.001", "image_output_rate": "0.0032"}, {"id": "gpt_image15_edit", "name": "gpt_image15_edit", "display_name": "GPT Image 1.5 [Edit]", "description": "GPT Image 1.5 [Edit]", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "openai", "text_input_rate": "0.0005", "image_input_rate": "0.0008", "text_output_rate": "0.001", "image_output_rate": "0.0032"}, {"id": "gpt_image2", "name": "gpt_image2", "display_name": "GPT Image 2", "description": "GPT Image 2", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "openai", "text_input_rate": "0.0005", "image_input_rate": "0.0008", "image_output_rate": "0.003"}, {"id": "gpt_image2_edit", "name": "gpt_image2_edit", "display_name": "GPT Image 2 [Edit]", "description": "GPT Image 2 [Edit]", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "openai", "text_input_rate": "0.0005", "image_input_rate": "0.0008", "image_output_rate": "0.003"}, {"id": "gpt_image25", "name": "gpt_image25", "display_name": "GPT Image 2.5", "description": "GPT Image 2.5", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "openai", "text_input_rate": "0.0005", "image_input_rate": "0.0008", "image_output_rate": "0.003"}, {"id": "gpt_image25_edit", "name": "gpt_image25_edit", "display_name": "GPT Image 2.5 [Edit]", "description": "GPT Image 2.5 [Edit]", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "openai", "text_input_rate": "0.0005", "image_input_rate": "0.0008", "image_output_rate": "0.003"}, {"id": "gpt_image26", "name": "gpt_image26", "display_name": "GPT Image 2.6", "description": "GPT Image 2.6", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "openai", "text_input_rate": "0.0005", "image_input_rate": "0.0008", "image_output_rate": "0.003"}, {"id": "flux_kontext", "name": "flux_kontext", "display_name": "FLUX.1 Kontext", "description": "FLUX.1 Kontext", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "black_forest_labs", "pro_rate": 4, "max_rate": 8}, {"id": "flux_kontext_edit", "name": "flux_kontext_edit", "display_name": "FLUX.1 Kontext [Edit]", "description": "FLUX.1 Kontext [Edit]", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "black_forest_labs", "pro_rate": 4, "max_rate": 8}, {"id": "flux_pro", "name": "flux_pro", "display_name": "FLUX 1.1 [pro]", "description": "FLUX 1.1 [pro]", "model_type": "image_generation", "input_modalities": ["text"], "output_modalities": ["image"], "provider": "black_forest_labs", "rate": 4}, {"id": "flux_pro_ultra", "name": "flux_pro_ultra", "display_name": "FLUX 1.1 [pro] Ultra", "description": "FLUX 1.1 [pro] Ultra", "model_type": "image_generation", "input_modalities": ["text"], "output_modalities": ["image"], "provider": "black_forest_labs", "rate": 6}, {"id": "ideogram", "name": "ideogram", "display_name": "Ideogram 3.0", "description": "Ideogram 3.0", "model_type": "image_generation", "input_modalities": ["text"], "output_modalities": ["image"], "provider": "ideogram", "TURBO_rate": 3, "BALANCED_rate": 6, "QUALITY_rate": 9}, {"id": "ideogram45", "name": "ideogram45", "display_name": "Ideogram 4.5", "description": "Ideogram 4.5", "model_type": "image_generation", "input_modalities": ["text"], "output_modalities": ["image"], "provider": "ideogram", "low_rate": 3, "medium_rate": 6, "high_rate": 22}, {"id": "ideogram_character", "name": "ideogram_character", "display_name": "Ideogram Character", "description": "Ideogram Character", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "ideogram", "TURBO_rate": 10, "BALANCED_rate": 15, "QUALITY_rate": 20}, {"id": "recraft", "name": "recraft", "display_name": "Recraft", "description": "Recraft", "model_type": "image_generation", "input_modalities": ["text"], "output_modalities": ["image"], "provider": "recraft", "rate": 4}, {"id": "recraft_svg", "name": "recraft_svg", "display_name": "Recraft SVG", "description": "Recraft SVG", "model_type": "image_generation", "input_modalities": ["text"], "output_modalities": ["image"], "provider": "recraft", "rate": 8}, {"id": "recraft_vectorize", "name": "recraft_vectorize", "display_name": "Recraft Vectorize", "description": "Recraft Vectorize", "model_type": "image_generation", "input_modalities": ["image"], "output_modalities": ["image"], "provider": "recraft", "rate": 1}, {"id": "remove_background", "name": "remove_background", "display_name": "Remove Background", "description": "Remove Background", "model_type": "image_generation", "input_modalities": ["image"], "output_modalities": ["image"], "provider": "pixelcut", "rate": "1.6", "bria_rate": "1.8"}, {"id": "magnific", "name": "magnific", "display_name": "Magnific Upscaler", "description": "Magnific Upscaler", "model_type": "image_generation", "input_modalities": ["image"], "output_modalities": ["image"], "provider": "magnific", "2k_rate": "11.9", "4k_rate": "23.8", "8k_rate": "81.7", "10k_rate": "81.7", "topaz_fallback_24mp_rate": 8, "topaz_fallback_48mp_rate": 16, "topaz_fallback_96mp_rate": 32, "topaz_fallback_512mp_rate": 136}, {"id": "flux_pro_canny", "name": "flux_pro_canny", "display_name": "FLUX 1.1 [pro] Canny [Edit]", "description": "FLUX 1.1 [pro] Canny [Edit]", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "black_forest_labs", "rate": 5}, {"id": "flux_pro_depth", "name": "flux_pro_depth", "display_name": "FLUX 1.1 [pro] Depth [Edit]", "description": "FLUX 1.1 [pro] Depth [Edit]", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "black_forest_labs", "rate": 5}, {"id": "midjourney", "name": "midjourney", "display_name": "Midjourney", "description": "Midjourney", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "midjourney", "v7_turbo_rate": 12, "v6_turbo_rate": 14, "fast_rate": 7, "relax_rate": 4}, {"id": "seedream", "name": "seedream", "display_name": "Seedream 4.5", "description": "Seedream 4.5", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "bytedance", "rate": 4}, {"id": "seedream5_lite", "name": "seedream5_lite", "display_name": "Seedream 5 Lite", "description": "Seedream 5 Lite", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "bytedance", "rate": "3.5"}, {"id": "seedream5_pro", "name": "seedream5_pro", "display_name": "Seedream 5 Pro", "description": "Seedream 5 Pro", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "bytedance", "rate": "6.75", "high_res_rate": "13.5", "additional_image_rate": "0.45"}, {"id": "dreamina", "name": "dreamina", "display_name": "Dreamina", "description": "Dreamina", "model_type": "image_generation", "input_modalities": ["text"], "output_modalities": ["image"], "provider": "bytedance", "rate": 3}, {"id": "nano_banana", "name": "nano_banana", "display_name": "Nano Banana", "description": "Nano Banana", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["text", "image"], "provider": "abacus", "rate": "3.9", "input_rate": "0.00003", "output_rate": "0.0030232558"}, {"id": "nano_banana_pro", "name": "nano_banana_pro", "display_name": "Nano Banana Pro", "description": "Nano Banana Pro", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["text", "image"], "provider": "abacus", "rate": "13.44", "input_rate": "0.0002", "text_output_rate": "0.0012", "image_output_rate": "0.012", "grounding_rate": "1.4"}, {"id": "nano_banana2", "name": "nano_banana2", "display_name": "Nano Banana 2", "description": "Nano Banana 2", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["text", "image"], "provider": "abacus", "rate": "6.72", "input_rate": "0.00005", "text_output_rate": "0.0003", "image_output_rate": "0.006", "grounding_rate": "1.4"}, {"id": "nano_banana21", "name": "nano_banana21", "display_name": "Nano Banana 2.1", "description": "Nano Banana 2.1", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["text", "image"], "provider": "abacus", "rate": "3.36", "input_rate": "0.00015", "text_output_rate": "0.00075", "image_output_rate": "0.003", "grounding_rate": "1.4"}, {"id": "nano_banana_lite", "name": "nano_banana_lite", "display_name": "Nano Banana Lite", "description": "Nano Banana Lite", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["text", "image"], "provider": "abacus", "rate": "3.36", "input_rate": "0.000025", "text_output_rate": "0.00015", "image_output_rate": "0.003"}, {"id": "qwen_image_edit", "name": "qwen_image_edit", "display_name": "Qwen Image Edit", "description": "Qwen Image Edit", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "qwen", "rate": 3, "unit": "megapixel"}, {"id": "hunyuan_image", "name": "hunyuan_image", "display_name": "Hunyuan Image 3.0", "description": "Hunyuan Image 3.0", "model_type": "image_generation", "input_modalities": ["text"], "output_modalities": ["image"], "provider": "tencent", "rate": 10}, {"id": "flux2", "name": "flux2", "display_name": "FLUX.2", "description": "FLUX.2", "model_type": "image_generation", "input_modalities": ["text"], "output_modalities": ["image"], "provider": "black_forest_labs", "rate": "0.96", "edit_rate": "1.2"}, {"id": "flux2_pro", "name": "flux2_pro", "display_name": "FLUX.2 [Pro]", "description": "FLUX.2 [Pro]", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "black_forest_labs", "rate": 3, "extra_mp_rate": "1.5", "input_rate": "1.5"}, {"id": "grok_imagine_image", "name": "grok_imagine_image", "display_name": "Grok Imagine Image", "description": "Grok Imagine Image", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "xai", "rate": 2, "image_input_rate": "0.2"}, {"id": "grok_imagine_image2", "name": "grok_imagine_image2", "display_name": "Grok Imagine Image 2", "description": "Grok Imagine Image 2", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "xai", "low_1k_rate": 4, "low_2k_rate": 6, "medium_1k_rate": 6, "medium_2k_rate": 8, "image_input_rate": 1}, {"id": "grok_imagine_image_quality", "name": "grok_imagine_image_quality", "display_name": "Grok Imagine Quality", "description": "Grok Imagine Quality", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "xai", "_1k_rate": 5, "_2k_rate": 7, "image_input_rate": 1}, {"id": "wan27", "name": "wan27", "display_name": "Wan 2.7", "description": "Wan 2.7", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "alibaba", "rate": 3}, {"id": "meta_muse_image", "name": "meta_muse_image", "display_name": "Meta Muse Image", "description": "Meta Muse Image", "model_type": "image_generation", "input_modalities": ["text", "image"], "output_modalities": ["image"], "provider": "meta", "rate": 1}, {"id": "gpt-audio-1.5", "name": "gpt-audio-1.5", "display_name": "GPT Audio 1.5", "description": "GPT Audio 1.5 supports audio input and output via the chat completions API, enabling speech understanding and generation in a single model call.", "model_type": "audio_generation", "input_modalities": ["text", "audio"], "output_modalities": ["text", "audio"], "input_token_rate": "0.0000025", "output_token_rate": "0.00001", "context_length": 128000, "max_completion_tokens": 16384, "provider": "openai", "cutoff_date": "2023-10-01", "api_formats": ["openai"], "featured": false}, {"id": "gpt-audio-mini", "name": "gpt-audio-mini", "display_name": "GPT Audio Mini", "description": "GPT Audio Mini is a cost-efficient model supporting audio input and output via the chat completions API, ideal for high-volume speech applications.", "model_type": "audio_generation", "input_modalities": ["text", "audio"], "output_modalities": ["text", "audio"], "input_token_rate": "0.0000006", "output_token_rate": "0.0000024", "context_length": 128000, "max_completion_tokens": 16384, "provider": "openai", "cutoff_date": "2023-10-01", "api_formats": ["openai"], "featured": false}, {"id": "gemini-2.5-flash-preview-tts", "name": "gemini-2.5-flash-preview-tts", "display_name": "Gemini 2.5 Flash TTS", "description": "Gemini 2.5 Flash TTS is a dedicated text-to-speech model from Google, supporting audio output generation via the Gemini API.", "model_type": "audio_generation", "input_modalities": ["text"], "output_modalities": ["text", "audio"], "input_token_rate": "0.0000005", "output_token_rate": "0", "context_length": 1048576, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "api_formats": ["openai"], "featured": false}, {"id": "gemini-2.5-pro-preview-tts", "name": "gemini-2.5-pro-preview-tts", "display_name": "Gemini 2.5 Pro TTS", "description": "Gemini 2.5 Pro TTS is a high-quality dedicated text-to-speech model from Google, supporting audio output generation via the Gemini API.", "model_type": "audio_generation", "input_modalities": ["text"], "output_modalities": ["text", "audio"], "input_token_rate": "0.000001", "output_token_rate": "0", "context_length": 2097152, "max_completion_tokens": 65536, "provider": "google", "cutoff_date": "2025-01-01", "api_formats": ["openai"], "featured": false}, {"id": "jev-latest", "name": "jev-latest", "display_name": "TypeSafe Jev", "description": "TypeSafe's Jev: a fast, calibrated decision model for classification, routing and scoring rather than text generation. Score a state against typed questions (choice, yes/no, score) and get back probabilities in under a second. Call it through /v1/decisions (OpenAI Decisions schema) or /v1/systemone (Jev schema); billed on input tokens only.", "model_type": "classification", "input_modalities": ["text"], "output_modalities": ["text"], "input_token_rate": "0.000000042", "output_token_rate": "0", "context_length": 32000, "provider": "typesafe", "api_formats": ["decisions", "systemone"]}, {"id": "gemini-decision-1", "name": "gemini-decision-1", "display_name": "Gemini Decisions", "description": "Google's Gemini Decisions: a calibrated decision model for classification, moderation, routing and scoring rather than text generation. Score text, images, audio, video or PDFs against typed questions (yes/no, choice, score; up to 1000 per /v1/decisions call) and get back probabilities. Call it through /v1/decisions (OpenAI Decisions schema, or Gemini's own) or /v1/systemone (Jev schema); billed on input tokens only.", "model_type": "classification", "input_modalities": ["text", "image", "audio", "video"], "output_modalities": ["text"], "input_token_rate": "0.0000001", "output_token_rate": "0", "context_length": 262144, "provider": "google", "api_formats": ["decisions", "systemone"]}]}