{"object":"list","data":[{"id":"claude-haiku-5-5","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":64000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125},"description":"Fastest, cheapest Claude 5.5 model: 1M context, 128k output, adaptive thinking with effort. Tiered pricing: $0.10/$0.50 per Mtok up to 100k-token prompts, $0.50/$2.50 above.","display_name":"Claude Haiku 5.5","reasoning_levels":["low","medium","high","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.06,"output":0.3,"cache_read":0.006,"cache_write":0.075},"is_primary":true},{"id":"claude-haiku-5.5","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":64000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125},"description":"Dotted-id alias of claude-haiku-5-5.","display_name":"Claude Haiku 5.5 (dotted alias)","reasoning_levels":["low","medium","high","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.06,"output":0.3,"cache_read":0.006,"cache_write":0.075},"is_primary":true},{"id":"claude-opus-4-6","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":5.0,"output":25.0,"cache_read":0.5,"cache_write":6.25},"description":"Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks. It is built for agents that operate across entire workflows rather than single prompts, making it especially effective...","display_name":"Claude Opus 4.6","quality_index":53.0,"reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":3.0,"output":15.0,"cache_read":0.3,"cache_write":3.75},"is_primary":true},{"id":"claude-opus-4-7","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":5.0,"output":25.0,"cache_read":0.5,"cache_write":6.25},"description":"Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents. Building on the coding and agentic strengths of Opus 4.6, it delivers stronger performance on...","display_name":"Claude Opus 4.7","quality_index":54.5,"reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":3.0,"output":15.0,"cache_read":0.3,"cache_write":3.75},"is_primary":true},{"id":"claude-opus-4-8","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":5.0,"output":25.0,"cache_read":0.5,"cache_write":6.25},"description":"Anthropic flagship model, May 28 2026. Best coding, agentic tasks, honesty.","display_name":"Claude Opus 4.8","quality_index":55.7,"reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":3.0,"output":15.0,"cache_read":0.3,"cache_write":3.75},"is_primary":true},{"id":"claude-opus-5","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":5.0,"output":25.0,"cache_read":0.5,"cache_write":6.25},"tags":["claude","opus","coding","agentic"],"description":"Anthropic Claude Opus 5: 1M context, 128K output, adaptive thinking","display_name":"Claude Opus 5","reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":3.0,"output":15.0,"cache_read":0.3,"cache_write":3.75},"is_primary":true},{"id":"claude-opus-5-5","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":4.0,"output":20.0,"cache_read":0.2,"cache_write":5.0},"description":"Anthropic Claude Opus 5.5 (2026-09-21): 1M context, 128K output, adaptive thinking. $4/$20.","display_name":"Claude Opus 5.5","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":2.4,"output":12.0,"cache_read":0.12,"cache_write":3.0},"is_primary":true},{"id":"claude-sonnet-4-6","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":3.0,"output":15.0,"cache_read":0.3,"cache_write":3.75},"description":"Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work. It excels at iterative development, complex codebase navigation, end-to-end project management with...","display_name":"Claude Sonnet 4.6","quality_index":47.2,"reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.8,"output":9.0,"cache_read":0.18,"cache_write":2.25},"is_primary":true},{"id":"claude-sonnet-5","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":3.0,"output":15.0,"cache_read":0.3,"cache_write":3.75},"display_name":"Claude Sonnet 5","quality_index":53.5,"reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.8,"output":9.0,"cache_read":0.18,"cache_write":2.25},"is_primary":true},{"id":"claude-sonnet-5-5","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.0,"output":10.0,"cache_read":0.2,"cache_write":2.5},"description":"Claude Sonnet 5.5 (2026-09-28) — best speed/intelligence combo, Sonnet-class; adaptive thinking (default effort high); knowledge cutoff Jun 2026","display_name":"Claude Sonnet 5.5","quality_index":53.5,"reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.2,"output":6.0,"cache_read":0.12,"cache_write":1.5},"is_primary":true},{"id":"deepseek-v4-pro","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":384000,"reasoning":true,"supports_vision":false,"supports_tools":true,"pricing":{"input":0.435,"output":0.87,"cache_read":0.0036,"cache_write":0.0},"description":"DeepSeek V4 Pro (GA 0813, 2026-08-12): 1M ctx, 384K max output, thinking+non-thinking, Responses API + Codex. The deepseek-v4-pro endpoint now serves DeepSeek-V4-Pro-0813.","display_name":"DeepSeek V4 Pro","quality_index":44.3,"reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort"],"architecture":{"input_modalities":["text"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.261,"output":0.522,"cache_read":0.0022,"cache_write":0.0},"is_primary":true},{"id":"deepseek-v4.1-flash","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":384000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.15,"output":0.6,"cache_read":0.003,"cache_write":0.0},"description":"DeepSeek V4.1-Flash (native id deepseek-flash): multimodal MoE 552B/8-16B active, 1M ctx, 384k out. Pricing=off-peak native headline; PEAK (01-04 & 06-10 UTC Mon-Fri) doubles to 0.30/1.20 native. OR/fireworks bill flat ~0.20-0.22/0.60-0.66.","display_name":"DeepSeek V4.1 Flash","quality_index":40.3,"reasoning_levels":["low","medium","high","xhigh"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.09,"output":0.36,"cache_read":0.0018,"cache_write":0.0},"is_primary":true},{"id":"fable-5","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":10.0,"output":50.0,"cache_read":1.0,"cache_write":12.5},"description":"Claude Fable 5 alias","display_name":"Claude Fable 5","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":6.0,"output":30.0,"cache_read":0.6,"cache_write":7.5},"is_primary":true},{"id":"fable-5.1","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":10.0,"output":50.0,"cache_read":0.25,"cache_write":12.5},"display_name":"Claude Fable 5.1","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":6.0,"output":30.0,"cache_read":0.15,"cache_write":7.5},"is_primary":true},{"id":"gemini-3.1-pro-preview","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":65536,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.0,"output":12.0,"cache_read":0.2,"cache_write":0.0},"description":"Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows. Building on the multimodal foundation...","display_name":"Gemini 3.1 Pro Preview","quality_index":46.5,"reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.2,"output":7.2,"cache_read":0.12,"cache_write":0.0},"is_primary":true},{"id":"gemini-3.7-flash","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":65536,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.375,"output":1.875,"cache_read":0.0375,"cache_write":0.0},"description":"Google Gemini 3.7 Flash — workhorse multimodal (text/image/video/audio/file), 1M context.","display_name":"Gemini 3.7 Flash","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.225,"output":1.125,"cache_read":0.0225,"cache_write":0.0},"is_primary":true},{"id":"gemini-3.8-flash","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":65536,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.375,"output":1.875,"cache_read":0.0375,"cache_write":0.0},"display_name":"Gemini 3.8 Flash","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.225,"output":1.125,"cache_read":0.0225,"cache_write":0.0},"is_primary":true},{"id":"gemini-4-argon","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":65536,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.0,"output":10.0,"cache_read":0.1,"cache_write":0.0},"description":"Google flagship reasoning model (Gemini 4 Argon). Intro pricing 2/10 per Mtok through 2026-11-30.","display_name":"Gemini 4 Argon","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.2,"output":6.0,"cache_read":0.06,"cache_write":0.0},"is_primary":true},{"id":"glm-5.3","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":131072,"reasoning":true,"supports_vision":false,"supports_tools":true,"pricing":{"input":1.4,"output":4.4,"cache_read":0.26,"cache_write":0.0},"description":"Z.ai GLM-5.3 (2026-08-14): same 743B base as GLM-5.2, all gains from post-training. Coding + agentic + cyber. Effort levels validated live: none/minimal/low/medium/high/xhigh/max.","display_name":"GLM-5.3","quality_index":57.0,"reasoning_levels":["low","high","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort"],"architecture":{"input_modalities":["text"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.84,"output":2.64,"cache_read":0.156,"cache_write":0.0},"is_primary":true},{"id":"glm-5.3-flash","object":"model","created":0,"is_free":false,"context_window":1310720,"max_output_tokens":131072,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.15,"output":0.5,"cache_read":0.03,"cache_write":0.0},"description":"GLM 5.3 Flash — the unmasked Ox Alpha stealth model. Cheap fast reasoning, 1.3M context.","display_name":"GLM 5.3 Flash","reasoning_levels":["low","high","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.09,"output":0.3,"cache_read":0.018,"cache_write":0.0},"is_primary":true},{"id":"gpt-5.4","object":"model","created":0,"is_free":false,"context_window":400000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.5,"output":15.0,"cache_read":0.25,"cache_write":0.0},"recommended_api":"openai-responses","description":"GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system. It features a 1M+ token context window (922K input, 128K output) with support for...","display_name":"GPT-5.4","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.5,"output":9.0,"cache_read":0.15,"cache_write":0.0},"is_primary":true},{"id":"gpt-5.5","object":"model","created":0,"is_free":false,"context_window":1050000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":5.0,"output":30.0,"cache_read":0.5,"cache_write":0.0},"recommended_api":"openai-responses","description":"GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token...","display_name":"GPT-5.5","quality_index":54.8,"reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":3.0,"output":18.0,"cache_read":0.3,"cache_write":0.0},"is_primary":true},{"id":"gpt-5.6-luna","object":"model","created":0,"is_free":false,"context_window":1050000,"max_output_tokens":128000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":1.0,"output":6.0,"cache_read":0.1,"cache_write":0.0},"recommended_api":"openai-responses","display_name":"GPT-5.6 Luna","quality_index":42.0,"reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.6,"output":3.6,"cache_read":0.06,"cache_write":0.0},"is_primary":true},{"id":"gpt-5.6-sol","object":"model","created":0,"is_free":false,"context_window":1050000,"max_output_tokens":128000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":5.0,"output":30.0,"cache_read":0.5,"cache_write":0.0},"recommended_api":"openai-responses","display_name":"GPT-5.6 Sol","quality_index":56.0,"reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":3.0,"output":18.0,"cache_read":0.3,"cache_write":0.0},"is_primary":true},{"id":"gpt-5.6-terra","object":"model","created":0,"is_free":false,"context_window":1050000,"max_output_tokens":128000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.5,"output":15.0,"cache_read":0.25,"cache_write":0.0},"recommended_api":"openai-responses","display_name":"GPT-5.6 Terra","quality_index":54.8,"reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.5,"output":9.0,"cache_read":0.15,"cache_write":0.0},"is_primary":true},{"id":"gpt-6-astra","object":"model","created":0,"is_free":false,"context_window":1050000,"max_output_tokens":128000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":10.0,"output":50.0,"cache_read":1.0,"cache_write":12.5},"recommended_api":"openai-responses","description":"OpenAI GPT-6 Astra (2026-09-03) — frontier reasoning model for complex agentic workflows, coding, and computer use. 1.05M context.","display_name":"GPT-6 Astra","reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":6.0,"output":30.0,"cache_read":0.6,"cache_write":7.5},"is_primary":true},{"id":"gpt-6-luna","object":"model","created":0,"is_free":false,"context_window":1050000,"max_output_tokens":128000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125},"recommended_api":"openai-responses","description":"OpenAI GPT-6 Luna (2026-09-22) — most efficient GPT-6 for focused high-volume tasks; strong at high reasoning effort. 1.05M context.","display_name":"GPT-6 Luna","reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.06,"output":0.3,"cache_read":0.006,"cache_write":0.075},"is_primary":true},{"id":"gpt-6-sol","object":"model","created":0,"is_free":false,"context_window":1050000,"max_output_tokens":128000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.0,"output":10.0,"cache_read":0.2,"cache_write":2.5},"recommended_api":"openai-responses","description":"OpenAI GPT-6 Sol (2026-09-22) — cost-efficient high-end GPT-6 for demanding agentic coding, computer use, long-horizon SWE. 1.05M context.","display_name":"GPT-6 Sol","reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.2,"output":6.0,"cache_read":0.12,"cache_write":1.5},"is_primary":true},{"id":"gpt-6.1-sol","object":"model","created":0,"is_free":false,"context_window":1050000,"max_output_tokens":128000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.0,"output":10.0,"cache_read":0.1,"cache_write":2.5},"recommended_api":"openai-responses","description":"OpenAI GPT-6.1 Sol (2026-09-29) — upgrade to GPT-6 Sol; near-Astra quality at ~1/5 cost for coding, computer use, professional work. 1.05M context.","display_name":"GPT-6.1 Sol","reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.2,"output":6.0,"cache_read":0.06,"cache_write":1.5},"is_primary":true},{"id":"gpt-6.1-terra","object":"model","created":0,"is_free":false,"context_window":1050000,"max_output_tokens":128000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.5,"output":2.5,"cache_read":0.0,"cache_write":0.0},"recommended_api":"openai-responses","description":"OpenAI GPT-6.1 Terra — fast coding & reasoning model.","display_name":"GPT-6.1 Terra","reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.3,"output":1.5,"cache_read":0.0,"cache_write":0.0},"is_primary":true},{"id":"grok-4.6","object":"model","created":0,"is_free":false,"context_window":500000,"max_output_tokens":16384,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.0,"output":6.0,"cache_read":0.5,"cache_write":0.0},"description":"xAI Grok 4.6 (2026-08-12): long-running agents, coding, visual work. 500K ctx, xhigh reasoning tier.","display_name":"Grok 4.6","quality_index":61.0,"reasoning_levels":["minimal","low","medium","high","xhigh"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.2,"output":3.6,"cache_read":0.3,"cache_write":0.0},"is_primary":true},{"id":"grok-4.7","object":"model","created":0,"is_free":false,"context_window":500000,"max_output_tokens":16384,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.0,"output":6.0,"cache_read":0.5,"cache_write":0.0},"description":"xAI Grok 4.7 (2026-09-21): larger base, longer-horizon coding/agentic + knowledge work, self-verification. 500K ctx. OpenRouter primary (cheaper $1.6/$4.8), xAI native fallback.","display_name":"Grok 4.7","quality_index":62.0,"reasoning_levels":["minimal","low","medium","high","xhigh"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.2,"output":3.6,"cache_read":0.3,"cache_write":0.0},"is_primary":true},{"id":"hy3","object":"model","created":0,"is_free":false,"context_window":262144,"max_output_tokens":32000,"reasoning":true,"supports_vision":false,"supports_tools":true,"pricing":{"input":0.132,"output":0.528,"cache_read":0.033,"cache_write":0.0},"description":"Tencent Hy3 — 295B MoE (21B active) for reasoning and agentic workflows. Text-only, 256K context. Preview auto-routes here.","display_name":"Hy3","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort"],"architecture":{"input_modalities":["text"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.0792,"output":0.3168,"cache_read":0.0198,"cache_write":0.0},"is_primary":true},{"id":"kimi-for-coding","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":131072,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.95,"output":4.0,"cache_read":0.19,"cache_write":0.0},"display_name":"Kimi For Coding (K2.8 Preview)","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.57,"output":2.4,"cache_read":0.114,"cache_write":0.0},"is_primary":true},{"id":"kimi-k2.7-code","object":"model","created":0,"is_free":false,"context_window":262144,"max_output_tokens":131072,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.95,"output":4.0,"cache_read":0.19,"cache_write":0.0},"tags":["coding","kimi","moonshot"],"description":"Moonshot's Kimi K2.7, tuned for long-horizon coding and agentic workflows.","display_name":"Kimi K2.7 Code","quality_index":42.0,"reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.57,"output":2.4,"cache_read":0.114,"cache_write":0.0},"is_primary":true},{"id":"kimi-k3","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":131072,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":3.0,"output":15.0,"cache_read":0.6,"cache_write":0.0},"tags":["overloaded"],"display_name":"Kimi K3","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.8,"output":9.0,"cache_read":0.36,"cache_write":0.0},"is_primary":true},{"id":"kimi-k3.1-pro","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":131072,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.0,"output":10.0,"cache_read":0.4,"cache_write":0.0},"tags":["kimi"],"description":"Moonshot flagship K3.1 Pro (2026-10-23): 3T-class MoE, 1M ctx, thinking-only (requests with thinking disabled are rejected). Disabled until a provider grants access; price = launch $2/$10 per 1M (source: orcarouter/AA, verify with Moonshot).","display_name":"Kimi K3.1 Pro","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.2,"output":6.0,"cache_read":0.24,"cache_write":0.0},"is_primary":true},{"id":"mimo-v2.6-flash","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":131072,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.14,"output":0.28,"cache_read":0.0028,"cache_write":0.0},"description":"MiMo-V2.6-Flash: 309B MoE (15B active), 1M context, omni-modal — cheap high-frequency reasoning.","display_name":"Xiaomi: MiMo-V2.6-Flash","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.084,"output":0.168,"cache_read":0.0017,"cache_write":0.0},"is_primary":true},{"id":"mimo-v2.6-pro","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":131072,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.435,"output":0.87,"cache_read":0.0036,"cache_write":0.0},"description":"MiMo-V2.6-Pro: 1.02T MoE (42B active), 1M context, omni-modal, RL-scaled reasoning; best open-weights class.","display_name":"Xiaomi: MiMo-V2.6-Pro","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.261,"output":0.522,"cache_read":0.0022,"cache_write":0.0},"is_primary":true},{"id":"mimo-v2.6-pro-ultraspeed","object":"model","created":0,"is_free":false,"context_window":1048576,"max_output_tokens":131072,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":4.35,"output":8.7,"cache_read":0.036,"cache_write":0.0},"description":"MiMo-V2.6-Pro-UltraSpeed: same Pro checkpoint, ~10x output speed. Premier price tier.","display_name":"Xiaomi: MiMo-V2.6-Pro-UltraSpeed","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":2.61,"output":5.22,"cache_read":0.0216,"cache_write":0.0},"is_primary":true},{"id":"minimax-m3","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.3,"output":1.2,"cache_read":0.06,"cache_write":0.0},"description":"Frontier multimodal coding model, 1M context, MSA attention, May 31 2026","display_name":"MiniMax M3","quality_index":44.4,"reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.18,"output":0.72,"cache_read":0.036,"cache_write":0.0},"is_primary":true},{"id":"minimax-m3.1-flash","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":16384,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":0.15,"output":0.6,"cache_read":0.015,"cache_write":0.0},"description":"MiniMax M3.1 Flash. 1M ctx. reasoning_effort low/medium/high/xhigh/max (default max). Thinking cannot be disabled.","display_name":"MiniMax M3.1 Flash","reasoning_levels":["low","medium","high","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.09,"output":0.36,"cache_read":0.009,"cache_write":0.0},"is_primary":true},{"id":"nemotron-3-ultra","object":"model","created":0,"is_free":false,"context_window":262144,"max_output_tokens":16384,"reasoning":true,"supports_vision":false,"supports_tools":true,"pricing":{"input":0.5,"output":2.2,"cache_read":0.0,"cache_write":0.0},"description":"NVIDIA Nemotron 3 Ultra — large reasoning model for agentic tasks.","display_name":"Nemotron 3 Ultra","quality_index":37.8,"reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort"],"architecture":{"input_modalities":["text"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":0.3,"output":1.32,"cache_read":0.0,"cache_write":0.0},"is_primary":true},{"id":"qwen3.8-max","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":131072,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":2.0,"output":6.0,"cache_read":0.25,"cache_write":0.0},"description":"Qwen3.8 Max — flagship of Alibaba Qwen3.8 series. Multimodal reasoning (text/image/video), 1M context.","display_name":"Qwen3.8 Max","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"discount_pct":40.0,"effective_pricing":{"input":1.2,"output":3.6,"cache_read":0.15,"cache_write":0.0},"is_primary":true},{"id":"claude-fable-5-1","object":"model","created":0,"is_free":false,"context_window":1000000,"max_output_tokens":32000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":10.0,"output":50.0,"cache_read":0.25,"cache_write":12.5},"description":"Historical Fable alias","display_name":"Claude Fable 5.1","reasoning_levels":["low","medium","high"],"reasoning_levels_declared":false,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"alias_of":"fable-5.1","discount_pct":40.0,"effective_pricing":{"input":6.0,"output":30.0,"cache_read":0.15,"cache_write":7.5},"is_primary":false},{"id":"gpt-6","object":"model","created":0,"is_free":false,"context_window":1050000,"max_output_tokens":128000,"reasoning":true,"supports_vision":true,"supports_tools":true,"pricing":{"input":10.0,"output":50.0,"cache_read":1.0,"cache_write":12.5},"recommended_api":"openai-responses","description":"GPT-6 Astra bare alias","display_name":"GPT-6 Astra","reasoning_levels":["low","medium","high","xhigh","max"],"reasoning_levels_declared":true,"picker_enabled":true,"policy":{"state":"enabled"},"supported_parameters":["messages","temperature","top_p","stream","tools","tool_choice","reasoning","reasoning_effort","image_url"],"architecture":{"input_modalities":["text","image"],"output_modalities":["text"]},"alias_of":"gpt-6-astra","discount_pct":40.0,"effective_pricing":{"input":6.0,"output":30.0,"cache_read":0.6,"cache_write":7.5},"is_primary":false}]}