{"models":[{"id":"alibaba:eu/deepseek-v4-flash","type":"llm","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"DeepSeek V4 Flash","model_id":"deepseek-v4-flash","capabilities":["text","tools","structured_output","reasoning","thinking"],"speed":"fast","cost":0.031,"quality":"good","context_window":1048576,"max_output_tokens":393216,"description":"DeepSeek V4 Flash lightweight MoE (284B params, 13B active) with 1M context — fast and cost-efficient, served EU-resident from Frankfurt (eu-central-1)","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.138],"output":[0.275],"cached_input":[0.028]},"region":"EU","model":"deepseek-v4-flash","family":"deepseek-v4-flash","pooled":true},{"id":"alibaba:eu/deepseek-v4-pro","type":"llm","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"DeepSeek V4 Pro","model_id":"deepseek-v4-pro","capabilities":["text","tools","structured_output","reasoning"],"speed":"medium","cost":0.21,"quality":"great","context_window":1048576,"max_output_tokens":393216,"description":"DeepSeek V4 Pro flagship MoE (1.6T params, 49B active) with 1M context, served EU-resident from Frankfurt (eu-central-1)","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[1.65],"output":[3.301],"cached_input":[0.138]},"region":"EU","model":"deepseek-v4-pro","family":"deepseek-v4-pro","pooled":true},{"id":"alibaba:eu/glm-5.1","type":"llm","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"GLM 5.1","model_id":"glm-5.1","capabilities":["text","tools","reasoning","structured_output","thinking"],"speed":"fast","cost":0.189,"quality":"good","context_window":202752,"max_output_tokens":131072,"description":"Zhipu AI GLM-5.1 (744B params) for long-horizon tasks with 128K max output, served EU-resident from Frankfurt (eu-central-1)","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","thresholds":[32768],"input":[0.825,1.1],"output":[3.301,3.851],"cached_input":[0.165,0.22]},"region":"EU","model":"glm-5.1","family":"glm-5.1","pooled":true},{"id":"alibaba:eu/happyhorse-1.0-i2v","type":"video","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"HappyHorse 1.0 Image-to-Video","model_id":"happyhorse-1.0-i2v","capabilities":["video_generation","image_to_video","audio"],"speed":"slow","quality":"best","context_window":0,"description":"Animate a first-frame image with text-described motion and synchronized audio. 720P/1080P, 3-15s.","params":{"default_max_tokens":0,"video":{"aspect_ratios":["16:9","9:16"],"resolutions":["720p","1080p"],"max_duration":15,"parameters":[{"name":"resolution","type":"string","description":"Output resolution tier (DashScope casing).","enum":["720P","1080P"],"example":"720P"},{"name":"duration","type":"integer","description":"Clip length in seconds (3-15).","example":"5"},{"name":"negative_prompt","type":"string","description":"Content to steer the generation away from."},{"name":"seed","type":"integer","description":"Seed for reproducible generation."}]}},"pricing":{"billing_unit":"per_second_by_resolution","rates":[{"unit":"per_second","amount":0.24,"dimensions":{"resolution":"1080p"}},{"unit":"per_second","amount":0.14,"dimensions":{"resolution":"720p"}}],"price_per_second_by_resolution":{"1080p":0.24,"720p":0.14}},"region":"EU","model":"happyhorse-1.0-i2v","family":"happyhorse-1.0-i2v","pooled":true},{"id":"alibaba:eu/happyhorse-1.0-r2v","type":"video","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"HappyHorse 1.0 Reference-to-Video","model_id":"happyhorse-1.0-r2v","capabilities":["video_generation","reference_images","audio"],"speed":"slow","quality":"best","context_window":0,"description":"Generate video keeping a subject from reference images consistent, with synchronized audio. 720P/1080P, 3-15s.","params":{"default_max_tokens":0,"video":{"aspect_ratios":["16:9","9:16"],"resolutions":["720p","1080p"],"max_duration":15,"parameters":[{"name":"resolution","type":"string","description":"Output resolution tier (DashScope casing).","enum":["720P","1080P"],"example":"720P"},{"name":"duration","type":"integer","description":"Clip length in seconds (3-15).","example":"5"},{"name":"negative_prompt","type":"string","description":"Content to steer the generation away from."},{"name":"seed","type":"integer","description":"Seed for reproducible generation."}]}},"pricing":{"billing_unit":"per_second_by_resolution","rates":[{"unit":"per_second","amount":0.24,"dimensions":{"resolution":"1080p"}},{"unit":"per_second","amount":0.14,"dimensions":{"resolution":"720p"}}],"price_per_second_by_resolution":{"1080p":0.24,"720p":0.14}},"region":"EU","model":"happyhorse-1.0-r2v","family":"happyhorse-1.0-r2v","pooled":true},{"id":"alibaba:eu/happyhorse-1.0-t2v","type":"video","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"HappyHorse 1.0 Text-to-Video","model_id":"happyhorse-1.0-t2v","capabilities":["video_generation","audio"],"speed":"slow","quality":"best","context_window":0,"description":"Generate video with synchronized audio from a text prompt. 720P/1080P, 3-15s.","params":{"default_max_tokens":0,"video":{"aspect_ratios":["16:9","9:16"],"resolutions":["720p","1080p"],"max_duration":15,"parameters":[{"name":"resolution","type":"string","description":"Output resolution tier (DashScope casing).","enum":["720P","1080P"],"example":"720P"},{"name":"duration","type":"integer","description":"Clip length in seconds (3-15).","example":"5"},{"name":"negative_prompt","type":"string","description":"Content to steer the generation away from."},{"name":"seed","type":"integer","description":"Seed for reproducible generation."}]}},"pricing":{"billing_unit":"per_second_by_resolution","rates":[{"unit":"per_second","amount":0.24,"dimensions":{"resolution":"1080p"}},{"unit":"per_second","amount":0.14,"dimensions":{"resolution":"720p"}}],"price_per_second_by_resolution":{"1080p":0.24,"720p":0.14}},"region":"EU","model":"happyhorse-1.0-t2v","family":"happyhorse-1.0-t2v","pooled":true},{"id":"alibaba:eu/kimi-k2.5","type":"llm","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"Kimi K2.5","model_id":"kimi-k2.5","capabilities":["text","tools","reasoning","vision","structured_output","thinking"],"speed":"fast","cost":0.173,"quality":"good","context_window":262144,"max_output_tokens":16384,"description":"Moonshot AI Kimi K2.5 native-multimodal model with vision input and hybrid reasoning, served EU-resident from Frankfurt (eu-central-1)","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.574],"output":[3.011],"cached_input":[0.115]},"region":"EU","model":"kimi-k2.5","family":"kimi-k2.5","pooled":true},{"id":"alibaba:eu/qwen3-vl-flash","type":"llm","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"Qwen3-VL-Flash","model_id":"qwen3-vl-flash","capabilities":["text","tools","reasoning","vision","structured_output"],"speed":"fast","cost":0.034,"quality":"good","context_window":262144,"max_output_tokens":32768,"description":"Alibaba's Qwen3-VL-Flash low-latency vision-language model served from Frankfurt (eu-central-1); fast visual understanding for its tier","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","thresholds":[131072],"input":[0.05,0.12],"output":[0.4,0.96]},"region":"EU","model":"qwen3-vl-flash","family":"qwen3-vl-flash","pooled":true},{"id":"alibaba:eu/qwen3-vl-plus","type":"llm","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"Qwen3-VL-Plus","model_id":"qwen3-vl-plus","capabilities":["text","tools","reasoning","vision","structured_output"],"speed":"fast","cost":0.108,"quality":"good","context_window":262144,"max_output_tokens":32768,"description":"Alibaba's Qwen3-VL-Plus vision-language model served from Frankfurt (eu-central-1); strong visual agent and document understanding with thinking + non-thinking modes","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","thresholds":[131072],"input":[0.2,0.6],"output":[1.6,4.8]},"region":"EU","model":"qwen3-vl-plus","family":"qwen3-vl-plus","pooled":true},{"id":"alibaba:eu/qwen3.6-flash","type":"llm","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"Qwen3.6-Flash","model_id":"qwen3.6-flash","capabilities":["text","tools","reasoning","vision","structured_output"],"speed":"fast","cost":0.077,"quality":"good","context_window":1000000,"max_output_tokens":65536,"description":"Qwen3.6-Flash low-latency vision-language model served from Frankfurt (eu-central-1); strong agentic coding and reasoning for its tier","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","thresholds":[262144],"input":[0.165,0.66],"output":[0.99,3.961],"cached_input":[0.017,0.066]},"region":"EU","model":"qwen3.6-flash","family":"qwen3.6-flash","pooled":true},{"id":"alibaba:eu/qwen3.7-max","type":"llm","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"Qwen3.7-Max","model_id":"qwen3.7-max","capabilities":["text","tools","reasoning","structured_output","thinking"],"speed":"medium","cost":0.246,"quality":"great","context_window":1000000,"max_output_tokens":65536,"description":"Alibaba's Qwen3.7-Max flagship served from Frankfurt (eu-central-1); next-generation agent-centric model with deep thinking — pure-text interface during public experimentation","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[1.65],"output":[4.951],"cached_input":[0.33]},"region":"EU","aliases":["alibaba/qwen3.7-max"],"model":"qwen3.7-max","family":"qwen3.7-max","pooled":true},{"id":"alibaba:eu/qwen3.7-plus","type":"llm","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"Qwen3.7-Plus","model_id":"qwen3.7-plus","capabilities":["text","tools","reasoning","vision","structured_output"],"speed":"fast","cost":0.088,"quality":"good","context_window":1000000,"max_output_tokens":65536,"description":"Qwen3.7-Plus served from Frankfurt (eu-central-1); cost-effective agent-centric model with deep thinking and native vision-language input","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","thresholds":[262144],"input":[0.276,0.826],"output":[1.101,3.301],"cached_input":[0.056,0.166]},"region":"EU","model":"qwen3.7-plus","family":"qwen3.7-plus","pooled":true},{"id":"alibaba:eu/wan2.6-i2v","type":"video","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"Wan 2.6 Image-to-Video","model_id":"wan2.6-i2v","capabilities":["video_generation","image_to_video","audio"],"speed":"slow","quality":"great","context_window":0,"description":"Animate a first-frame image with text-described motion and synchronized audio. 720P/1080P, 2-15s.","params":{"default_max_tokens":0,"video":{"aspect_ratios":["16:9","9:16"],"resolutions":["720p","1080p"],"max_duration":15,"parameters":[{"name":"resolution","type":"string","description":"Output resolution tier (DashScope casing).","enum":["720P","1080P"],"example":"720P"},{"name":"duration","type":"integer","description":"Clip length in seconds (2-15).","example":"5"},{"name":"negative_prompt","type":"string","description":"Content to steer the generation away from."},{"name":"audio_url","type":"string","description":"Optional public URL to a WAV/MP3 driving audio track."},{"name":"prompt_extend","type":"boolean","description":"LLM prompt rewriting (default true)."},{"name":"seed","type":"integer","description":"Seed for reproducible generation."}]}},"pricing":{"billing_unit":"per_second_by_resolution","rates":[{"unit":"per_second","amount":0.15,"dimensions":{"resolution":"1080p"}},{"unit":"per_second","amount":0.1,"dimensions":{"resolution":"720p"}}],"price_per_second_by_resolution":{"1080p":0.15,"720p":0.1}},"region":"EU","model":"wan2.6-i2v","family":"wan2.6-i2v","pooled":true},{"id":"alibaba:eu/wan2.6-r2v","type":"video","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"Wan 2.6 Reference-to-Video","model_id":"wan2.6-r2v","capabilities":["video_generation","reference_video","audio"],"speed":"slow","quality":"great","context_window":0,"description":"Re-animate a subject from a reference video with text-described motion and synchronized audio. Pass the reference as the `video` input. 720P/1080P, 2-10s.","params":{"default_max_tokens":0,"video":{"aspect_ratios":["16:9","9:16"],"resolutions":["720p","1080p"],"max_duration":10,"parameters":[{"name":"resolution","type":"string","description":"Output resolution tier (DashScope casing).","enum":["720P","1080P"],"example":"720P"},{"name":"duration","type":"integer","description":"Clip length in seconds (2-10 with a reference video).","example":"5"},{"name":"negative_prompt","type":"string","description":"Content to steer the generation away from."},{"name":"seed","type":"integer","description":"Seed for reproducible generation."}]}},"pricing":{"billing_unit":"per_second_by_resolution","rates":[{"unit":"per_second","amount":0.15,"dimensions":{"resolution":"1080p"}},{"unit":"per_second","amount":0.1,"dimensions":{"resolution":"720p"}}],"price_per_second_by_resolution":{"1080p":0.15,"720p":0.1}},"region":"EU","model":"wan2.6-r2v","family":"wan2.6-r2v","pooled":true},{"id":"alibaba:eu/wan2.6-t2v","type":"video","provider":"alibaba:eu","provider_display_name":"Alibaba Cloud","name":"Wan 2.6 Text-to-Video","model_id":"wan2.6-t2v","capabilities":["video_generation","audio"],"speed":"slow","quality":"great","context_window":0,"description":"Text-to-video with synchronized audio and multi-shot support. 720P/1080P, 5/10/15s.","params":{"default_max_tokens":0,"video":{"aspect_ratios":["16:9","9:16","1:1"],"resolutions":["720p","1080p"],"max_duration":15,"parameters":[{"name":"size","type":"string","description":"Output WxH (Wan uses size, not a resolution tier).","enum":["1280*720","720*1280","960*960","1920*1080","1080*1920","1440*1440"],"example":"1280*720"},{"name":"duration","type":"integer","description":"Clip length in seconds (2-15; commonly 5/10/15).","example":"5"},{"name":"negative_prompt","type":"string","description":"Content to steer the generation away from."},{"name":"audio_url","type":"string","description":"Optional public URL to a WAV/MP3 driving audio track."},{"name":"prompt_extend","type":"boolean","description":"LLM prompt rewriting (default true)."},{"name":"shot_type","type":"string","description":"Single- or multi-shot (requires prompt_extend).","enum":["single","multi"]},{"name":"seed","type":"integer","description":"Seed for reproducible generation."}]}},"pricing":{"billing_unit":"per_second_by_resolution","rates":[{"unit":"per_second","amount":0.15,"dimensions":{"resolution":"1080p"}},{"unit":"per_second","amount":0.1,"dimensions":{"resolution":"720p"}}],"price_per_second_by_resolution":{"1080p":0.15,"720p":0.1}},"region":"EU","model":"wan2.6-t2v","family":"wan2.6-t2v","pooled":true},{"id":"anthropic/claude-haiku-4-5","type":"llm","provider":"anthropic","provider_display_name":"Anthropic","name":"Claude Haiku 4.5","model_id":"claude-haiku-4-5-20251001","capabilities":["text","vision","structured_output","tools","pdf"],"speed":"fast","cost":0.234,"quality":"good","context_window":200000,"max_output_tokens":64000,"description":"Fast and efficient Claude model","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096},"pricing":{"billing_unit":"per_mtok","input":[1],"output":[5],"cached_input":[0.1],"cache_creation":[1.25],"cache_creation_1h":[2],"web_search_price_per_request":0.01},"featured":true,"region":"US","country":"United States","aliases":["anthropic/claude-haiku-4.5"],"server_tools":{"web_search":"native"},"model":"claude-haiku-4-5-20251001","family":"claude-haiku-4-5-20251001","pooled":true},{"id":"anthropic/claude-opus-4-1","type":"llm","provider":"anthropic","provider_display_name":"Anthropic","name":"Claude Opus 4.1","model_id":"claude-opus-4-1-20250805","capabilities":["text","vision","structured_output","tools","pdf"],"speed":"slow","cost":0.646,"quality":"best","context_window":200000,"max_output_tokens":32000,"description":"Claude Opus 4.1 model","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096,"reasoning":{"supported":["low","medium","high"],"default":"medium"}},"pricing":{"billing_unit":"per_mtok","input":[15],"output":[75],"cached_input":[1.5],"cache_creation":[18.75],"cache_creation_1h":[30],"web_search_price_per_request":0.01},"region":"US","country":"United States","aliases":["anthropic/claude-opus-4.1"],"server_tools":{"web_search":"native"},"model":"claude-opus-4-1-20250805","family":"claude-opus-4-1-20250805","pooled":true},{"id":"anthropic/claude-opus-4-5","type":"llm","provider":"anthropic","provider_display_name":"Anthropic","name":"Claude Opus 4.5","model_id":"claude-opus-4-5-20251101","capabilities":["text","vision","structured_output","tools","pdf"],"speed":"slow","cost":0.467,"quality":"best","context_window":200000,"max_output_tokens":64000,"description":"Most capable Claude model with extended thinking","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096,"reasoning":{"supported":["low","medium","high"],"default":"medium"}},"pricing":{"billing_unit":"per_mtok","input":[5],"output":[25],"cached_input":[0.5],"cache_creation":[6.25],"cache_creation_1h":[10],"web_search_price_per_request":0.01},"region":"US","country":"United States","aliases":["anthropic/claude-opus-4.5"],"server_tools":{"web_search":"native"},"model":"claude-opus-4-5-20251101","family":"claude-opus-4-5-20251101","pooled":true},{"id":"anthropic/claude-opus-4-6","type":"llm","provider":"anthropic","provider_display_name":"Anthropic","name":"Claude Opus 4.6","model_id":"claude-opus-4-6","capabilities":["text","vision","structured_output","tools","pdf"],"speed":"slow","cost":0.467,"quality":"best","context_window":1000000,"max_output_tokens":128000,"description":"Most intelligent Claude model for agents and coding","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096,"reasoning":{"supported":["low","medium","high","max"],"default":"high","mode":"adaptive","max_budget":64000}},"pricing":{"billing_unit":"per_mtok","input":[5],"output":[25],"cached_input":[0.5],"cache_creation":[6.25],"cache_creation_1h":[10],"web_search_price_per_request":0.01},"region":"US","country":"United States","aliases":["anthropic/claude-opus-4.6"],"server_tools":{"web_search":"native"},"model":"claude-opus-4-6","family":"claude-opus-4-6","pooled":true},{"id":"anthropic/claude-opus-4-7","type":"llm","provider":"anthropic","provider_display_name":"Anthropic","name":"Claude Opus 4.7","model_id":"claude-opus-4-7","capabilities":["text","vision","structured_output","tools","pdf","reasoning"],"speed":"slow","cost":0.467,"quality":"best","context_window":1000000,"max_output_tokens":128000,"description":"Most capable Claude model for complex reasoning and agentic coding","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096,"reasoning":{"supported":["low","medium","high","xhigh","max"],"default":"high","mode":"adaptive","max_budget":64000}},"pricing":{"billing_unit":"per_mtok","input":[5],"output":[25],"cached_input":[0.5],"cache_creation":[6.25],"cache_creation_1h":[10],"web_search_price_per_request":0.01},"region":"US","country":"United States","aliases":["anthropic/claude-opus-4.7"],"server_tools":{"web_search":"native"},"model":"claude-opus-4-7","family":"claude-opus-4-7","pooled":true},{"id":"anthropic/claude-opus-4-8","type":"llm","provider":"anthropic","provider_display_name":"Anthropic","name":"Claude Opus 4.8","model_id":"claude-opus-4-8","capabilities":["text","vision","structured_output","tools","pdf","reasoning"],"speed":"slow","cost":0.467,"quality":"best","context_window":1000000,"max_output_tokens":128000,"description":"Anthropic's most capable model for complex reasoning and agentic coding","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096,"reasoning":{"supported":["low","medium","high","xhigh","max"],"default":"high","mode":"adaptive"}},"pricing":{"billing_unit":"per_mtok","input":[5],"output":[25],"cached_input":[0.5],"cache_creation":[6.25],"cache_creation_1h":[10],"web_search_price_per_request":0.01},"featured":true,"region":"US","country":"United States","server_tools":{"web_search":"native"},"model":"claude-opus-4-8","family":"claude-opus-4-8","pooled":true},{"id":"anthropic/claude-sonnet-4-5","type":"llm","provider":"anthropic","provider_display_name":"Anthropic","name":"Claude Sonnet 4.5","model_id":"claude-sonnet-4-5-20250929","capabilities":["text","vision","structured_output","tools","pdf","reasoning"],"speed":"medium","cost":0.388,"quality":"great","context_window":200000,"max_output_tokens":64000,"description":"Balanced Claude model for most tasks","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096,"reasoning":{"supported":["low","medium","high"],"default":"medium"}},"pricing":{"billing_unit":"per_mtok","input":[3],"output":[15],"cached_input":[0.3],"cache_creation":[3.75],"cache_creation_1h":[6],"web_search_price_per_request":0.01},"region":"US","country":"United States","aliases":["anthropic/claude-sonnet-4.5"],"server_tools":{"web_search":"native"},"model":"claude-sonnet-4-5","family":"claude-sonnet-4-5","pooled":true},{"id":"anthropic/claude-sonnet-4-6","type":"llm","provider":"anthropic","provider_display_name":"Anthropic","name":"Claude Sonnet 4.6","model_id":"claude-sonnet-4-6","capabilities":["text","vision","structured_output","tools","pdf","reasoning"],"speed":"medium","cost":0.388,"quality":"great","context_window":1000000,"max_output_tokens":64000,"description":"Latest Sonnet model, balanced for most tasks","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096,"reasoning":{"supported":["low","medium","high","max"],"default":"high","mode":"adaptive","max_budget":64000}},"pricing":{"billing_unit":"per_mtok","input":[3],"output":[15],"cached_input":[0.3],"cache_creation":[3.75],"cache_creation_1h":[6],"web_search_price_per_request":0.01},"featured":true,"region":"US","country":"United States","aliases":["anthropic/claude-sonnet-4.6"],"server_tools":{"web_search":"native"},"model":"claude-sonnet-4-6","family":"claude-sonnet-4-6","pooled":true},{"id":"anthropic/claude-sonnet-5","type":"llm","provider":"anthropic","provider_display_name":"Anthropic","name":"Claude Sonnet 5","model_id":"claude-sonnet-5","capabilities":["text","vision","structured_output","tools","pdf","reasoning"],"speed":"medium","cost":0.328,"quality":"great","context_window":1000000,"max_output_tokens":64000,"description":"Latest Sonnet model, balanced for most tasks","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096,"reasoning":{"supported":["low","medium","high","max"],"default":"high","mode":"adaptive","max_budget":64000}},"pricing":{"billing_unit":"per_mtok","input":[2],"output":[10],"cached_input":[0.2],"cache_creation":[2.5],"cache_creation_1h":[4],"web_search_price_per_request":0.01},"featured":true,"region":"US","country":"United States","server_tools":{"web_search":"native"},"model":"claude-sonnet-5","family":"claude-sonnet-5","pooled":true},{"id":"arcee/trinity-large-preview","type":"llm","provider":"arcee","provider_display_name":"Arcee","name":"Trinity Large Preview","model_id":"trinity-large-preview","capabilities":["text","tools","structured_output"],"speed":"fast","cost":0.082,"quality":"good","context_window":131072,"max_output_tokens":8192,"description":"Arcee Trinity Large 400B preview model","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.25],"output":[1]},"region":"US","country":"United States","model":"trinity-large-preview","family":"trinity-large-preview","pooled":true},{"id":"arcee/trinity-large-thinking","type":"llm","provider":"arcee","provider_display_name":"Arcee","name":"Trinity Large Thinking","model_id":"trinity-large-thinking","capabilities":["text","tools","structured_output","reasoning"],"speed":"fast","cost":0.076,"quality":"good","context_window":262144,"max_output_tokens":8192,"description":"Arcee Trinity Large 400B thinking model with reasoning capabilities","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.25],"output":[0.9],"cached_input":[0.06]},"region":"US","country":"United States","model":"trinity-large-thinking","family":"trinity-large-thinking","pooled":true},{"id":"arcee/trinity-mini","type":"llm","provider":"arcee","provider_display_name":"Arcee","name":"Trinity Mini","model_id":"trinity-mini","capabilities":["text","tools","structured_output"],"speed":"fast","cost":0.016,"quality":"good","context_window":131072,"max_output_tokens":8192,"description":"Arcee Trinity Mini 26B model","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.045],"output":[0.15]},"region":"US","country":"United States","model":"trinity-mini","family":"trinity-mini","pooled":true},{"id":"aws/claude-3-haiku-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Claude 3 Haiku","model_id":"eu.anthropic.claude-3-haiku-20240307-v1:0","capabilities":["text","tools","vision"],"speed":"fast","cost":0.094,"quality":"good","context_window":200000,"max_output_tokens":4096,"description":"Claude 3 Haiku via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096},"pricing":{"billing_unit":"per_mtok","input":[0.25],"output":[1.25]},"region":"EU","country":"France","model":"claude-3-haiku","family":"claude-3-haiku","pooled":true},{"id":"aws/claude-haiku-4-5-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Claude Haiku 4.5","model_id":"eu.anthropic.claude-haiku-4-5-20251001-v1:0","capabilities":["text","tools","vision","structured_output"],"speed":"medium","cost":0.246,"quality":"great","context_window":200000,"max_output_tokens":64000,"description":"Claude Haiku 4.5 via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096},"pricing":{"billing_unit":"per_mtok","input":[1.1],"output":[5.5],"cached_input":[0.11]},"region":"EU","country":"France","aliases":["aws/claude-haiku-4.5-eu"],"model":"claude-haiku-4-5","family":"claude-haiku-4-5","pooled":true},{"id":"aws/claude-opus-4-5-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Claude Opus 4.5","model_id":"eu.anthropic.claude-opus-4-5-20251101-v1:0","capabilities":["text","tools","vision","pdf","structured_output","reasoning"],"speed":"medium","cost":0.467,"quality":"best","context_window":200000,"max_output_tokens":64000,"description":"Claude Opus 4.5 via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096},"pricing":{"billing_unit":"per_mtok","input":[5],"output":[25],"cached_input":[0.5]},"region":"EU","country":"France","aliases":["aws/claude-opus-4.5-eu"],"model":"claude-opus-4-5","family":"claude-opus-4-5","pooled":true},{"id":"aws/claude-opus-4-6-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Claude Opus 4.6","model_id":"eu.anthropic.claude-opus-4-6-v1","capabilities":["text","vision","structured_output","tools","pdf"],"speed":"medium","cost":0.467,"quality":"best","context_window":1000000,"max_output_tokens":128000,"description":"Claude Opus 4.6 via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096},"pricing":{"billing_unit":"per_mtok","input":[5],"output":[25],"cached_input":[0.5]},"region":"EU","country":"Sweden","aliases":["aws/claude-opus-4.6-eu"],"model":"claude-opus-4-6","family":"claude-opus-4-6","pooled":true},{"id":"aws/claude-opus-4-7","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Claude Opus 4.7","model_id":"eu.anthropic.claude-opus-4-7","capabilities":["text","vision","structured_output","tools","pdf","reasoning"],"speed":"medium","cost":0.467,"quality":"best","context_window":1000000,"max_output_tokens":128000,"description":"Claude Opus 4.7 via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096},"pricing":{"billing_unit":"per_mtok","input":[5],"output":[25],"cached_input":[0.5]},"region":"EU","country":"Sweden","aliases":["aws/claude-opus-4.7"],"model":"claude-opus-4-7","family":"claude-opus-4-7","pooled":true},{"id":"aws/claude-opus-4-8","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Claude Opus 4.8","model_id":"eu.anthropic.claude-opus-4-8","capabilities":["text","vision","structured_output","tools","pdf","reasoning"],"speed":"medium","cost":0.467,"quality":"best","context_window":1000000,"max_output_tokens":128000,"description":"Claude Opus 4.8 via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096},"pricing":{"billing_unit":"per_mtok","input":[5],"output":[25],"cached_input":[0.5]},"region":"EU","country":"Sweden","model":"claude-opus-4-8","family":"claude-opus-4-8","pooled":true},{"id":"aws/claude-sonnet-4-5-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Claude Sonnet 4.5","model_id":"eu.anthropic.claude-sonnet-4-5-20250929-v1:0","capabilities":["text","vision","structured_output","tools","pdf","reasoning"],"speed":"medium","cost":0.403,"quality":"great","context_window":200000,"max_output_tokens":64000,"description":"Claude Sonnet 4.5 via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096},"pricing":{"billing_unit":"per_mtok","input":[3.3],"output":[16.5],"cached_input":[0.33]},"region":"EU","country":"France","aliases":["aws/claude-sonnet-4.5-eu"],"model":"claude-sonnet-4-5","family":"claude-sonnet-4-5","pooled":true},{"id":"aws/claude-sonnet-4-6-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Claude Sonnet 4.6","model_id":"eu.anthropic.claude-sonnet-4-6","capabilities":["text","vision","structured_output","tools","pdf","reasoning"],"speed":"medium","cost":0.403,"quality":"great","context_window":1000000,"max_output_tokens":64000,"description":"Claude Sonnet 4.6 via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096},"pricing":{"billing_unit":"per_mtok","input":[3.3],"output":[16.5],"cached_input":[0.33]},"region":"EU","country":"Sweden","aliases":["aws/claude-sonnet-4.6-eu"],"model":"claude-sonnet-4-6","family":"claude-sonnet-4-6","pooled":true},{"id":"aws/cohere/rerank-v3.5","type":"rerank","provider":"bedrock","provider_display_name":"AWS","name":"Cohere Rerank v3.5","model_id":"cohere.rerank-v3-5:0","capabilities":["rerank"],"speed":"fast","quality":"best","context_window":0,"description":"Cohere Rerank v3.5 via AWS Bedrock (EU). Per-request pricing.","pricing":{"billing_unit":"per_generation","rates":[{"unit":"per_generation","amount":0.002}],"price_per_generation":0.002},"region":"EU","country":"Germany","model":"rerank-v3.5","family":"rerank-v3.5","pooled":true},{"id":"aws/gpt-oss-120b-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"GPT OSS 120B","model_id":"openai.gpt-oss-120b-1:0","capabilities":["text","structured_output","tools","reasoning","thinking","vision"],"speed":"fast","cost":0.054,"quality":"good","context_window":128000,"max_output_tokens":8192,"description":"Open source GPT 120B via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.15],"output":[0.6]},"region":"EU","country":"Sweden","model":"gpt-oss-120b","family":"gpt-oss-120b","pooled":true},{"id":"aws/gpt-oss-20b-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"GPT OSS 20B","model_id":"openai.gpt-oss-20b-1:0","capabilities":["text","tools","structured_output","reasoning","thinking","vision"],"speed":"fast","cost":0.028,"quality":"good","context_window":128000,"max_output_tokens":8192,"description":"Open source GPT 20B via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.07],"output":[0.3]},"region":"EU","country":"Sweden","model":"gpt-oss-20b","family":"gpt-oss-20b","pooled":true},{"id":"aws/ministral-14b","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Ministral 14B","model_id":"mistral.ministral-3-14b-instruct","capabilities":["text","tools"],"speed":"fast","cost":0.031,"quality":"good","context_window":128000,"max_output_tokens":8192,"description":"Mistral Ministral 14B via AWS Bedrock","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.2],"output":[0.2]},"region":"US","country":"United States","model":"ministral-14b","family":"ministral-14b","pooled":true},{"id":"aws/ministral-3b","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Ministral 3B","model_id":"mistral.ministral-3-3b-instruct","capabilities":["text","tools"],"speed":"fast","cost":0.016,"quality":"good","context_window":128000,"max_output_tokens":8192,"description":"Mistral Ministral 3B via AWS Bedrock","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.1],"output":[0.1]},"region":"US","country":"United States","model":"ministral-3b","family":"ministral-3b","pooled":true},{"id":"aws/ministral-8b","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Ministral 8B","model_id":"mistral.ministral-3-8b-instruct","capabilities":["text","tools"],"speed":"fast","cost":0.023,"quality":"good","context_window":128000,"max_output_tokens":8192,"description":"Mistral Ministral 8B via AWS Bedrock","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.15],"output":[0.15]},"region":"US","country":"United States","model":"ministral-8b","family":"ministral-8b","pooled":true},{"id":"aws/mistral-large-3","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Mistral Large 3","model_id":"mistral.mistral-large-3-675b-instruct","capabilities":["text","tools"],"speed":"fast","cost":0.117,"quality":"good","context_window":128000,"max_output_tokens":8192,"description":"Mistral Large 3 via AWS Bedrock","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.5],"output":[1.5]},"region":"US","country":"United States","model":"mistral-large-3","family":"mistral-large-3","pooled":true},{"id":"aws/nemotron-nano-12b-v2-vl","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Nemotron Nano 12B VL","model_id":"nvidia.nemotron-nano-12b-v2","capabilities":["text","vision"],"speed":"fast","cost":0.023,"quality":"good","context_window":128000,"max_output_tokens":8192,"description":"NVIDIA Nemotron Nano 12B Vision-Language via AWS Bedrock","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.06],"output":[0.24]},"region":"US","country":"United States","model":"nemotron-nano-12b-v2-vl","family":"nemotron-nano-12b-v2-vl","pooled":true},{"id":"aws/nemotron-nano-9b-v2","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Nemotron Nano 9B v2","model_id":"nvidia.nemotron-nano-9b-v2","capabilities":["text","tools","thinking"],"speed":"fast","cost":0.016,"quality":"good","context_window":128000,"max_output_tokens":8192,"description":"NVIDIA Nemotron Nano 9B via AWS Bedrock","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.04],"output":[0.16]},"region":"US","country":"United States","model":"nemotron-nano-9b-v2","family":"nemotron-nano-9b-v2","pooled":true},{"id":"aws/nova-2-lite-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Amazon Nova 2 Lite","model_id":"eu.amazon.nova-2-lite-v1:0","capabilities":["text","tools","vision","structured_output"],"speed":"fast","cost":0.148,"quality":"good","context_window":1000000,"max_output_tokens":64000,"description":"Amazon's Nova 2 multimodal model (1M context, prompt caching) via AWS Bedrock (EU cross-region inference)","params":{"temperature":{"min":0,"max":1,"default":0.7},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.3],"output":[2.5]},"region":"EU","country":"Sweden","model":"nova-2-lite","family":"nova-2-lite","pooled":true},{"id":"aws/nova-lite-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Amazon Nova Lite","model_id":"eu.amazon.nova-lite-v1:0","capabilities":["text","tools","vision","structured_output"],"speed":"fast","cost":0.023,"quality":"good","context_window":300000,"max_output_tokens":10000,"description":"Amazon's fast, cost-effective multimodal Nova model via AWS Bedrock (EU cross-region inference)","params":{"temperature":{"min":0,"max":1,"default":0.7},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.06],"output":[0.24]},"region":"EU","country":"Sweden","model":"nova-lite","family":"nova-lite","pooled":true},{"id":"aws/nova-micro-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Amazon Nova Micro","model_id":"eu.amazon.nova-micro-v1:0","capabilities":["text","tools","structured_output"],"speed":"fast","cost":0.014,"quality":"good","context_window":128000,"max_output_tokens":5000,"description":"Amazon's fast, low-cost Nova text model via AWS Bedrock (EU cross-region inference)","params":{"temperature":{"min":0,"max":1,"default":0.7},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.035],"output":[0.14]},"region":"EU","country":"Sweden","model":"nova-micro","family":"nova-micro","pooled":true},{"id":"aws/nova-pro-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Amazon Nova Pro","model_id":"eu.amazon.nova-pro-v1:0","capabilities":["text","tools","vision","structured_output"],"speed":"fast","cost":0.185,"quality":"good","context_window":300000,"max_output_tokens":10000,"description":"Amazon's high-capability multimodal Nova model via AWS Bedrock (EU cross-region inference)","params":{"temperature":{"min":0,"max":1,"default":0.7},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[0.8],"output":[3.2]},"region":"EU","country":"Sweden","model":"nova-pro","family":"nova-pro","pooled":true},{"id":"aws/pixtral-large-2502-eu","type":"llm","provider":"bedrock","provider_display_name":"AWS","name":"Pixtral Large","model_id":"eu.mistral.pixtral-large-2502-v1:0","capabilities":["text","tools","vision"],"speed":"medium","cost":0.271,"quality":"great","context_window":128000,"max_output_tokens":8192,"description":"Mistral Pixtral Large via AWS Bedrock (EU region)","params":{"temperature":{"min":0,"max":2,"default":1},"default_max_tokens":0},"pricing":{"billing_unit":"per_mtok","input":[2],"output":[6]},"region":"EU","country":"France","model":"pixtral-large-2502","family":"pixtral-large-2502","pooled":true},{"id":"azure/claude-haiku-4-5","type":"llm","provider":"azure-anthropic","provider_display_name":"Azure","name":"Claude Haiku 4.5","model_id":"claude-haiku-4-5","capabilities":["text","tools","vision","structured_output"],"speed":"medium","cost":0.234,"quality":"great","context_window":200000,"max_output_tokens":64000,"description":"Anthropic Claude Haiku 4.5 via Azure","params":{"temperature":{"min":0,"max":1,"default":1},"default_max_tokens":4096},"pricing":{"billing_unit":"per_mtok","input":[1],"output":[5],"cached_input":[0.1]},"region":"EU","country":"Sweden","aliases":["azure/claude-haiku-4.5"],"model":"claude-haiku-4-5","family":"claude-haiku-4-5","pooled":true}],"total":670,"offset":0,"limit":50}
