{"name":"Polarison AI API pricing and capability dataset","description":"Official API list prices (USD per 1M tokens), context limits and Epoch AI capability scores for leading AI models from OpenAI, Anthropic, Google, xAI and DeepSeek.","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"citation":"Polarison, “AI API pricing and capability dataset”, https://polarison.com/data/models.json. Benchmark data: Epoch AI, “Capabilities & Benchmarking”, epoch.ai/benchmarks (CC BY 4.0).","jsonUrl":"https://polarison.com/data/models.json","csvUrl":"https://polarison.com/data/models.csv","updated":"2026-09-15","units":{"prices":"USD per 1M tokens, standard tier","blended":"(3 × input + output) ÷ 4","benchmarks":"percent","eci":"Epoch Capabilities Index (higher is more capable)"},"sources":{"pricing":[{"provider":"OpenAI","url":"https://developers.openai.com/api/docs/pricing"},{"provider":"Anthropic","url":"https://platform.claude.com/docs/en/about-claude/pricing"},{"provider":"Google","url":"https://ai.google.dev/gemini-api/docs/pricing"},{"provider":"xAI","url":"https://docs.x.ai/docs/pricing"},{"provider":"DeepSeek","url":"https://api-docs.deepseek.com/quick_start/pricing"}],"benchmarks":{"name":"Epoch AI","url":"https://epoch.ai/benchmarks","license":"CC BY 4.0","retrieved":"2026-09-15"}},"benchmarks":[{"id":"gpqa","name":"GPQA Diamond","category":"Science reasoning","description":"Graduate-level biology, chemistry and physics questions designed to be hard to look up."},{"id":"frontiermath","name":"FrontierMath (Tiers 1–3)","category":"Advanced math","description":"Original, unpublished math problems ranging from advanced undergraduate to research level."},{"id":"simpleqa","name":"SimpleQA Verified","category":"Factual accuracy","description":"Short fact-seeking questions that reward correct answers over confident guesses."},{"id":"arcagi2","name":"ARC-AGI-2","category":"Abstract reasoning","description":"Visual pattern puzzles that are easy for people but hard for AI systems."},{"id":"deepswe","name":"DeepSWE","category":"Coding","description":"Agentic software engineering tasks that require changing real codebases."},{"id":"apexagents","name":"APEX-Agents","category":"Agentic work","description":"Long, multi-step professional tasks completed by an AI agent."}],"models":[{"id":"gpt-6-astra","name":"GPT-6 Astra","provider":"OpenAI","api_model_id":"gpt-6-astra","preview":false,"input_usd_per_mtok":10,"cached_input_usd_per_mtok":1,"output_usd_per_mtok":50,"blended_usd_per_mtok":20,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":20,"long_context_output_usd_per_mtok":75,"batch_discount":0.5,"context_window_tokens":1050000,"max_output_tokens":128000,"price_verified":"2026-09-14","pricing_source":"https://developers.openai.com/api/docs/pricing","eci":166.31,"eci_ci_low":163,"eci_ci_high":171.88,"gpqa_pct":94.4,"frontiermath_pct":93.7,"simpleqa_pct":75.6,"arcagi2_pct":95,"deepswe_pct":74.1,"apexagents_pct":46.7,"best_value":true,"notes":["Cache writes are billed at $12.50 per 1M tokens.","Fast mode costs 2x the standard rate; Flex costs half.","Regional (data residency) endpoints add a 10% uplift."],"url":"https://polarison.com/models/gpt-6-astra"},{"id":"gpt-5-6-sol","name":"GPT-5.6 Sol","provider":"OpenAI","api_model_id":"gpt-5.6-sol","preview":false,"input_usd_per_mtok":4,"cached_input_usd_per_mtok":0.4,"output_usd_per_mtok":20,"blended_usd_per_mtok":8,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":8,"long_context_output_usd_per_mtok":30,"batch_discount":0.5,"context_window_tokens":1050000,"max_output_tokens":128000,"price_verified":"2026-09-14","pricing_source":"https://developers.openai.com/api/docs/pricing","eci":161.81,"eci_ci_low":159.57,"eci_ci_high":165.2,"gpqa_pct":91.3,"frontiermath_pct":89.1,"simpleqa_pct":69.7,"arcagi2_pct":92.5,"deepswe_pct":72.7,"apexagents_pct":40,"best_value":true,"notes":["Promotional pricing, available at least through November 21, 2026.","Cache writes are billed at $5 per 1M tokens.","Regional (data residency) endpoints add a 10% uplift."],"url":"https://polarison.com/models/gpt-5-6-sol"},{"id":"gpt-5-6-terra","name":"GPT-5.6 Terra","provider":"OpenAI","api_model_id":"gpt-5.6-terra","preview":false,"input_usd_per_mtok":2,"cached_input_usd_per_mtok":0.2,"output_usd_per_mtok":12,"blended_usd_per_mtok":4.5,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":4,"long_context_output_usd_per_mtok":18,"batch_discount":0.5,"context_window_tokens":1050000,"max_output_tokens":128000,"price_verified":"2026-09-14","pricing_source":"https://developers.openai.com/api/docs/pricing","eci":159.14,"eci_ci_low":157.01,"eci_ci_high":161.84,"gpqa_pct":91.1,"frontiermath_pct":86,"simpleqa_pct":43.2,"arcagi2_pct":83.9,"deepswe_pct":69.6,"apexagents_pct":null,"best_value":true,"notes":["Cache writes are billed at $2.50 per 1M tokens.","Fast mode costs 2x the standard rate; Flex costs half."],"url":"https://polarison.com/models/gpt-5-6-terra"},{"id":"gpt-5-6-luna","name":"GPT-5.6 Luna","provider":"OpenAI","api_model_id":"gpt-5.6-luna","preview":false,"input_usd_per_mtok":0.2,"cached_input_usd_per_mtok":0.02,"output_usd_per_mtok":1.2,"blended_usd_per_mtok":0.45,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":0.4,"long_context_output_usd_per_mtok":1.8,"batch_discount":0.5,"context_window_tokens":1050000,"max_output_tokens":128000,"price_verified":"2026-09-14","pricing_source":"https://developers.openai.com/api/docs/pricing","eci":156.31,"eci_ci_low":154.06,"eci_ci_high":158.86,"gpqa_pct":88.8,"frontiermath_pct":82.1,"simpleqa_pct":41,"arcagi2_pct":59.5,"deepswe_pct":67.2,"apexagents_pct":null,"best_value":true,"notes":["Cache writes are billed at $0.25 per 1M tokens.","Fast mode costs 2x the standard rate; Flex costs half."],"url":"https://polarison.com/models/gpt-5-6-luna"},{"id":"claude-fable-5-1","name":"Claude Fable 5.1","provider":"Anthropic","api_model_id":"claude-fable-5-1","preview":false,"input_usd_per_mtok":10,"cached_input_usd_per_mtok":0.25,"output_usd_per_mtok":50,"blended_usd_per_mtok":20,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":null,"long_context_output_usd_per_mtok":null,"batch_discount":0.5,"context_window_tokens":1000000,"max_output_tokens":128000,"price_verified":"2026-09-14","pricing_source":"https://platform.claude.com/docs/en/about-claude/pricing","eci":164.47,"eci_ci_low":161.36,"eci_ci_high":168.28,"gpqa_pct":null,"frontiermath_pct":90.2,"simpleqa_pct":70.8,"arcagi2_pct":90,"deepswe_pct":null,"apexagents_pct":47.4,"best_value":false,"notes":["Cache hits cost 2.5% of the base input price, versus 10% on other Claude models.","Cache writes cost $12.50 (5-minute cache) or $20 (1-hour cache) per 1M tokens.","Uses Anthropic’s newer tokenizer, which produces roughly 30% more tokens for the same text.","US-only inference (inference_geo: \"us\") costs 1.1x."],"url":"https://polarison.com/models/claude-fable-5-1"},{"id":"claude-opus-5","name":"Claude Opus 5","provider":"Anthropic","api_model_id":"claude-opus-5","preview":false,"input_usd_per_mtok":5,"cached_input_usd_per_mtok":0.5,"output_usd_per_mtok":25,"blended_usd_per_mtok":10,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":null,"long_context_output_usd_per_mtok":null,"batch_discount":0.5,"context_window_tokens":1000000,"max_output_tokens":128000,"price_verified":"2026-09-14","pricing_source":"https://platform.claude.com/docs/en/about-claude/pricing","eci":162.3,"eci_ci_low":159.9,"eci_ci_high":165.74,"gpqa_pct":91.8,"frontiermath_pct":85.6,"simpleqa_pct":59.9,"arcagi2_pct":90.4,"deepswe_pct":73.6,"apexagents_pct":43.5,"best_value":true,"notes":["Cache writes cost $6.25 (5-minute cache) or $10 (1-hour cache) per 1M tokens.","Uses Anthropic’s newer tokenizer, which produces roughly 30% more tokens for the same text.","US-only inference (inference_geo: \"us\") costs 1.1x."],"url":"https://polarison.com/models/claude-opus-5"},{"id":"claude-sonnet-5","name":"Claude Sonnet 5","provider":"Anthropic","api_model_id":"claude-sonnet-5","preview":false,"input_usd_per_mtok":2,"cached_input_usd_per_mtok":0.2,"output_usd_per_mtok":10,"blended_usd_per_mtok":4,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":null,"long_context_output_usd_per_mtok":null,"batch_discount":0.5,"context_window_tokens":1000000,"max_output_tokens":128000,"price_verified":"2026-09-14","pricing_source":"https://platform.claude.com/docs/en/about-claude/pricing","eci":156.22,"eci_ci_low":153.63,"eci_ci_high":158.63,"gpqa_pct":87.4,"frontiermath_pct":65.6,"simpleqa_pct":33.7,"arcagi2_pct":null,"deepswe_pct":53.8,"apexagents_pct":32.5,"best_value":false,"notes":["The $2 / $10 launch price is now the standard price; the planned rise to $3 / $15 will not happen.","Cache writes cost $2.50 (5-minute cache) or $4 (1-hour cache) per 1M tokens.","Uses Anthropic’s newer tokenizer, which produces roughly 30% more tokens for the same text."],"url":"https://polarison.com/models/claude-sonnet-5"},{"id":"claude-haiku-4-5","name":"Claude Haiku 4.5","provider":"Anthropic","api_model_id":"claude-haiku-4-5-20251001","preview":false,"input_usd_per_mtok":1,"cached_input_usd_per_mtok":0.1,"output_usd_per_mtok":5,"blended_usd_per_mtok":2,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":null,"long_context_output_usd_per_mtok":null,"batch_discount":0.5,"context_window_tokens":200000,"max_output_tokens":64000,"price_verified":"2026-09-14","pricing_source":"https://platform.claude.com/docs/en/about-claude/pricing","eci":142.37,"eci_ci_low":139.8,"eci_ci_high":144.07,"gpqa_pct":61.6,"frontiermath_pct":null,"simpleqa_pct":13.2,"arcagi2_pct":4,"deepswe_pct":null,"apexagents_pct":8.9,"best_value":false,"notes":["Cache writes cost $1.25 (5-minute cache) or $2 (1-hour cache) per 1M tokens.","Uses Anthropic’s previous tokenizer."],"url":"https://polarison.com/models/claude-haiku-4-5"},{"id":"gemini-3-8-flash","name":"Gemini 3.8 Flash","provider":"Google","api_model_id":"gemini-3.8-flash","preview":false,"input_usd_per_mtok":0.75,"cached_input_usd_per_mtok":0.075,"output_usd_per_mtok":3.75,"blended_usd_per_mtok":1.5,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":null,"long_context_output_usd_per_mtok":null,"batch_discount":0.5,"context_window_tokens":1048576,"max_output_tokens":65536,"price_verified":"2026-09-14","pricing_source":"https://ai.google.dev/gemini-api/docs/pricing","eci":156.54,"eci_ci_low":154.58,"eci_ci_high":161.97,"gpqa_pct":93.9,"frontiermath_pct":68.4,"simpleqa_pct":69.7,"arcagi2_pct":null,"deepswe_pct":73.8,"apexagents_pct":null,"best_value":true,"notes":["Introductory price through December 31, 2026. From January 1, 2027: $1.50 input, $0.15 cached, $7.50 output per 1M tokens.","Output price includes thinking tokens.","Priority inference costs 1.8x the standard rate."],"url":"https://polarison.com/models/gemini-3-8-flash"},{"id":"gemini-3-5-flash","name":"Gemini 3.5 Flash","provider":"Google","api_model_id":"gemini-3.5-flash","preview":false,"input_usd_per_mtok":1.5,"cached_input_usd_per_mtok":0.15,"output_usd_per_mtok":9,"blended_usd_per_mtok":3.375,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":null,"long_context_output_usd_per_mtok":null,"batch_discount":0.5,"context_window_tokens":1048576,"max_output_tokens":65536,"price_verified":"2026-09-14","pricing_source":"https://ai.google.dev/gemini-api/docs/pricing","eci":154.69,"eci_ci_low":152.78,"eci_ci_high":157.03,"gpqa_pct":90.4,"frontiermath_pct":62.8,"simpleqa_pct":66.2,"arcagi2_pct":72.1,"deepswe_pct":37.4,"apexagents_pct":null,"best_value":false,"notes":["Output price includes thinking tokens.","Context cache storage costs $1.00 per 1M tokens per hour.","Priority inference costs 1.8x the standard rate."],"url":"https://polarison.com/models/gemini-3-5-flash"},{"id":"gemini-3-5-flash-lite","name":"Gemini 3.5 Flash-Lite","provider":"Google","api_model_id":"gemini-3.5-flash-lite","preview":false,"input_usd_per_mtok":0.3,"cached_input_usd_per_mtok":0.03,"output_usd_per_mtok":2.5,"blended_usd_per_mtok":0.85,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":null,"long_context_output_usd_per_mtok":null,"batch_discount":0.5,"context_window_tokens":1048576,"max_output_tokens":65536,"price_verified":"2026-09-14","pricing_source":"https://ai.google.dev/gemini-api/docs/pricing","eci":145.13,"eci_ci_low":142.76,"eci_ci_high":146.76,"gpqa_pct":77.8,"frontiermath_pct":26,"simpleqa_pct":null,"arcagi2_pct":10.3,"deepswe_pct":null,"apexagents_pct":null,"best_value":false,"notes":["The same input price applies to text, image, video and audio.","Output price includes thinking tokens."],"url":"https://polarison.com/models/gemini-3-5-flash-lite"},{"id":"gemini-3-1-pro","name":"Gemini 3.1 Pro","provider":"Google","api_model_id":"gemini-3.1-pro-preview","preview":true,"input_usd_per_mtok":2,"cached_input_usd_per_mtok":0.2,"output_usd_per_mtok":12,"blended_usd_per_mtok":4.5,"long_context_threshold_tokens":200000,"long_context_input_usd_per_mtok":4,"long_context_output_usd_per_mtok":18,"batch_discount":null,"context_window_tokens":1048576,"max_output_tokens":65536,"price_verified":"2026-09-14","pricing_source":"https://ai.google.dev/gemini-api/docs/pricing","eci":155,"eci_ci_low":152.72,"eci_ci_high":157.6,"gpqa_pct":92.6,"frontiermath_pct":59.6,"simpleqa_pct":73.5,"arcagi2_pct":77.1,"deepswe_pct":11.7,"apexagents_pct":33.5,"best_value":false,"notes":["Prompts over 200K tokens are billed at $4 input / $18 output per 1M tokens.","No free tier on the Gemini API.","Output price includes thinking tokens."],"url":"https://polarison.com/models/gemini-3-1-pro"},{"id":"gemini-2-5-pro","name":"Gemini 2.5 Pro","provider":"Google","api_model_id":"gemini-2.5-pro","preview":false,"input_usd_per_mtok":1.25,"cached_input_usd_per_mtok":0.125,"output_usd_per_mtok":10,"blended_usd_per_mtok":3.4375,"long_context_threshold_tokens":200000,"long_context_input_usd_per_mtok":2.5,"long_context_output_usd_per_mtok":15,"batch_discount":null,"context_window_tokens":1048576,"max_output_tokens":65536,"price_verified":"2026-09-14","pricing_source":"https://ai.google.dev/gemini-api/docs/pricing","eci":145.26,"eci_ci_low":143.82,"eci_ci_high":146.79,"gpqa_pct":80.4,"frontiermath_pct":24.6,"simpleqa_pct":null,"arcagi2_pct":4.9,"deepswe_pct":null,"apexagents_pct":6.6,"best_value":false,"notes":["Prompts over 200K tokens are billed at $2.50 input / $15 output per 1M tokens.","Output price includes thinking tokens."],"url":"https://polarison.com/models/gemini-2-5-pro"},{"id":"gemini-2-5-flash","name":"Gemini 2.5 Flash","provider":"Google","api_model_id":"gemini-2.5-flash","preview":false,"input_usd_per_mtok":0.3,"cached_input_usd_per_mtok":0.03,"output_usd_per_mtok":2.5,"blended_usd_per_mtok":0.85,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":null,"long_context_output_usd_per_mtok":null,"batch_discount":null,"context_window_tokens":1048576,"max_output_tokens":65536,"price_verified":"2026-09-14","pricing_source":"https://ai.google.dev/gemini-api/docs/pricing","eci":140.53,"eci_ci_low":138.48,"eci_ci_high":141.94,"gpqa_pct":null,"frontiermath_pct":null,"simpleqa_pct":null,"arcagi2_pct":null,"deepswe_pct":null,"apexagents_pct":1.8,"best_value":false,"notes":["Audio input costs $1.00 per 1M tokens; text, image and video cost $0.30.","Output price includes thinking tokens."],"url":"https://polarison.com/models/gemini-2-5-flash"},{"id":"grok-4-6","name":"Grok 4.6","provider":"xAI","api_model_id":"grok-4.6","preview":false,"input_usd_per_mtok":2,"cached_input_usd_per_mtok":0.5,"output_usd_per_mtok":6,"blended_usd_per_mtok":3,"long_context_threshold_tokens":200000,"long_context_input_usd_per_mtok":4,"long_context_output_usd_per_mtok":12,"batch_discount":null,"context_window_tokens":500000,"max_output_tokens":null,"price_verified":"2026-09-14","pricing_source":"https://docs.x.ai/docs/pricing","eci":156.35,"eci_ci_low":154.77,"eci_ci_high":158.88,"gpqa_pct":92,"frontiermath_pct":66,"simpleqa_pct":49.3,"arcagi2_pct":67.1,"deepswe_pct":67.5,"apexagents_pct":41.2,"best_value":false,"notes":["Requests with 200K or more prompt tokens are billed at $4 input / $1 cached / $12 output for all tokens.","Web Search and X Search tools cost $5 per 1,000 calls."],"url":"https://polarison.com/models/grok-4-6"},{"id":"grok-4-3","name":"Grok 4.3","provider":"xAI","api_model_id":"grok-4.3","preview":false,"input_usd_per_mtok":1.25,"cached_input_usd_per_mtok":0.2,"output_usd_per_mtok":2.5,"blended_usd_per_mtok":1.5625,"long_context_threshold_tokens":200000,"long_context_input_usd_per_mtok":2.5,"long_context_output_usd_per_mtok":5,"batch_discount":null,"context_window_tokens":1000000,"max_output_tokens":null,"price_verified":"2026-09-14","pricing_source":"https://docs.x.ai/docs/pricing","eci":null,"eci_ci_low":null,"eci_ci_high":null,"gpqa_pct":null,"frontiermath_pct":null,"simpleqa_pct":null,"arcagi2_pct":null,"deepswe_pct":null,"apexagents_pct":null,"best_value":false,"notes":["Requests with 200K or more prompt tokens are billed at $2.50 input / $0.40 cached / $5 output for all tokens."],"url":"https://polarison.com/models/grok-4-3"},{"id":"deepseek-flash","name":"DeepSeek V4.1 Flash","provider":"DeepSeek","api_model_id":"deepseek-flash","preview":false,"input_usd_per_mtok":0.3,"cached_input_usd_per_mtok":0.006,"output_usd_per_mtok":1.2,"blended_usd_per_mtok":0.525,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":null,"long_context_output_usd_per_mtok":null,"batch_discount":null,"context_window_tokens":1000000,"max_output_tokens":384000,"price_verified":"2026-09-14","pricing_source":"https://api-docs.deepseek.com/quick_start/pricing","eci":null,"eci_ci_low":null,"eci_ci_high":null,"gpqa_pct":null,"frontiermath_pct":null,"simpleqa_pct":null,"arcagi2_pct":null,"deepswe_pct":null,"apexagents_pct":null,"best_value":false,"notes":["Prices shown are peak rates. Off-peak rates are half: peak hours are 01:00–04:00 and 06:00–10:00 UTC, Monday to Friday.","Compatible with both OpenAI-format and Anthropic-format APIs."],"url":"https://polarison.com/models/deepseek-flash"},{"id":"deepseek-v4-pro","name":"DeepSeek V4 Pro","provider":"DeepSeek","api_model_id":"deepseek-v4-pro","preview":false,"input_usd_per_mtok":1.32,"cached_input_usd_per_mtok":0.044,"output_usd_per_mtok":3.96,"blended_usd_per_mtok":1.98,"long_context_threshold_tokens":null,"long_context_input_usd_per_mtok":null,"long_context_output_usd_per_mtok":null,"batch_discount":null,"context_window_tokens":1000000,"max_output_tokens":384000,"price_verified":"2026-09-14","pricing_source":"https://api-docs.deepseek.com/quick_start/pricing","eci":155.47,"eci_ci_low":153.78,"eci_ci_high":157.83,"gpqa_pct":88.9,"frontiermath_pct":64.6,"simpleqa_pct":52.9,"arcagi2_pct":61.3,"deepswe_pct":null,"apexagents_pct":null,"best_value":false,"notes":["Prices shown are peak rates. Off-peak rates are half: peak hours are 01:00–04:00 and 06:00–10:00 UTC, Monday to Friday.","Vision input is not supported."],"url":"https://polarison.com/models/deepseek-v4-pro"}]}