[
  {
    "id": "deepseek-v4-1-flash",
    "name": "DeepSeek V4.1 Flash",
    "provider": "DeepSeek",
    "release_date": "2026-09-10",
    "context_window": 1048576,
    "max_output": 384000,
    "input_cost": 0.15,
    "output_cost": 0.6,
    "cache_read": 0.003,
    "cache_write": null,
    "reasoning": false,
    "modalities": [
      "text"
    ],
    "description": "DeepSeek's 1M context efficiency flagship, refreshed as the 1017 checkpoint (October 17, 2026) with unchanged off-peak rates of $0.15/$0.60 and $0.003 cache-hit input; standard rate is $0.30/$1.20 with $0.006 cache read. 384K max output.",
    "popular": true,
    "popularity_rank": 1
  },
  {
    "id": "claude-opus-5-5",
    "name": "Claude Opus 5.5",
    "provider": "Anthropic",
    "release_date": "2026-09-22",
    "context_window": 1000000,
    "max_output": 128000,
    "input_cost": 4,
    "output_cost": 20,
    "cache_read": 0.2,
    "cache_write": 5,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Anthropic's frontier reasoning flagship (September 22, 2026) topping the Artificial Analysis Intelligence Index at 58.",
    "popular": true,
    "popularity_rank": 2
  },
  {
    "id": "glm-5-3-flash",
    "name": "GLM 5.3 Flash",
    "provider": "Zhipu AI",
    "release_date": "2026-08",
    "context_window": 1310720,
    "max_output": 16384,
    "input_cost": 0.15,
    "output_cost": 0.5,
    "cache_read": 0.03,
    "cache_write": 0.075,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Z.ai's premier high-speed multimodal reasoning model (August 26, 2026) with a 1.31M context window. The September launch promotion ended, so the live list rate is $0.15/$0.50 per 1M tokens with $0.03 cache reads.",
    "popular": true,
    "popularity_rank": 3
  },
  {
    "id": "kimi-k3-1-pro",
    "name": "Kimi K3.1 Pro",
    "provider": "Moonshot",
    "release_date": "2026-10-23",
    "context_window": 1048576,
    "max_output": 943718,
    "input_cost": 2,
    "output_cost": 10,
    "cache_read": 0.2,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text",
      "vision",
      "video"
    ],
    "description": "Moonshot AI's flagship open-weight multimodal reasoning model (October 23, 2026) and post-trained successor to Kimi K3, with a 1M-token context window and native image/video understanding. Priced 33% below K3 at $2/$10 per 1M tokens with $0.20 cache reads.",
    "popular": true,
    "popularity_rank": 11
  },
  {
    "id": "minimax-m3-1-flash",
    "name": "MiniMax M3.1 Flash",
    "provider": "MiniMax",
    "release_date": "2026-10-19",
    "context_window": 1048576,
    "max_output": 524288,
    "input_cost": 0.15,
    "output_cost": 0.6,
    "cache_read": 0.015,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text",
      "vision",
      "video"
    ],
    "description": "MiniMax's low-cost agentic reasoning model (October 19, 2026) with a 1M-token context window and text/image/video input, at $0.15/$0.60 per 1M tokens with $0.015 cache reads.",
    "popular": true,
    "popularity_rank": 10
  },
  {
    "id": "mercury-3",
    "name": "Mercury 3",
    "provider": "Inception",
    "release_date": "2026-10-19",
    "context_window": 1000000,
    "max_output": 131072,
    "input_cost": 0.1,
    "output_cost": 0.35,
    "cache_read": 0.01,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text"
    ],
    "description": "Inception's flagship diffusion LLM (October 19, 2026) running at 1,068 tokens/sec on NVIDIA GPUs with a 1M-token context window, at $0.10/$0.35 per 1M tokens with $0.01 cache reads.",
    "popular": false,
    "popularity_rank": 30
  },
  {
    "id": "step-5-preview",
    "name": "Step 5 Preview",
    "provider": "StepFun",
    "release_date": "2026-10-08",
    "context_window": 1000000,
    "max_output": 65536,
    "input_cost": 1,
    "output_cost": 2.7,
    "cache_read": 0.1,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "StepFun's flagship agentic MoE model (27B active / 600B total, October 8, 2026) at $1/$2.70 per 1M tokens with 1M context, $0.10 cache reads, and strength in software engineering and finance.",
    "popular": true,
    "popularity_rank": 17
  },
  {
    "id": "mimo-v2-6-flash",
    "name": "MiMo-V2.6 Flash",
    "provider": "Xiaomi",
    "release_date": "2026-09-22",
    "context_window": 1050000,
    "max_output": 131072,
    "input_cost": 0.14,
    "output_cost": 0.28,
    "cache_read": 0.0028,
    "cache_write": null,
    "reasoning": false,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Xiaomi's MIT-licensed open-weights efficiency tier (September 22, 2026) at $0.14/$0.28 per 1M tokens.",
    "popular": true,
    "popularity_rank": 5
  },
  {
    "id": "gpt-6-astra",
    "name": "GPT-6 Astra",
    "provider": "OpenAI",
    "release_date": "2026-09-04",
    "context_window": 1050000,
    "max_output": 128000,
    "input_cost": 10,
    "output_cost": 50,
    "cache_read": 1,
    "cache_write": 12.5,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "OpenAI's top-tier frontier model (September 4, 2026) at $10/$50 per 1M tokens with 1.05M context and 128K output.",
    "popular": true,
    "popularity_rank": 13
  },
  {
    "id": "claude-fable-5-1",
    "name": "Claude Fable 5.1",
    "provider": "Anthropic",
    "release_date": "2026-08",
    "context_window": 1000000,
    "max_output": 64000,
    "input_cost": 10,
    "output_cost": 50,
    "cache_read": 0.25,
    "cache_write": 12.5,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Anthropic's frontier reasoning and coding flagship leading LMSYS Arena Elo in August 2026, now at $10/$50 per 1M tokens.",
    "popular": true,
    "popularity_rank": 16
  },
  {
    "id": "claude-sonnet-5-5",
    "name": "Claude Sonnet 5.5",
    "provider": "Anthropic",
    "release_date": "2026-09-28",
    "context_window": 1000000,
    "max_output": 128000,
    "input_cost": 2,
    "output_cost": 10,
    "cache_read": 0.1,
    "cache_write": 2.5,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Anthropic's workhorse flagship (September 28, 2026) at $2/$10 with 1M context, running 30% faster and up to 30% cheaper per task.",
    "popular": true,
    "popularity_rank": 9
  },
  {
    "id": "gpt-6-luna",
    "name": "GPT-6 Luna",
    "provider": "OpenAI",
    "release_date": "2026-09-22",
    "context_window": 1050000,
    "max_output": 128000,
    "input_cost": 0.1,
    "output_cost": 0.5,
    "cache_read": 0.01,
    "cache_write": 0.125,
    "reasoning": false,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "OpenAI's ultra-low-cost 1.05M context budget tier (September 22, 2026) at $0.10/$0.50 per 1M tokens with $0.01 cached input.",
    "popular": true,
    "popularity_rank": 4
  },
  {
    "id": "qwen-3-8-27b",
    "name": "Qwen 3.8 27B",
    "provider": "Qwen",
    "release_date": "2026-09",
    "context_window": 1000000,
    "max_output": 8192,
    "input_cost": 0.2,
    "output_cost": 2.5,
    "cache_read": 0.09,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Qwen's dense 27B vision-language open-weight model with controllable thinking and native speculative decoding, repriced in October 2026 to $0.20/$2.50 per 1M tokens with $0.09 cache reads.",
    "popular": true,
    "popularity_rank": 18
  },
  {
    "id": "gpt-6-1-sol",
    "name": "GPT-6.1 Sol",
    "provider": "OpenAI",
    "release_date": "2026-09-29",
    "context_window": 1050000,
    "max_output": 128000,
    "input_cost": 2,
    "output_cost": 10,
    "cache_read": 0.1,
    "cache_write": 2.5,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "OpenAI's DevDay flagship (September 29, 2026) at $2/$10 per 1M tokens with 1.05M context and 128K output.",
    "popular": true,
    "popularity_rank": 7
  },
  {
    "id": "qwen-3-8-flash",
    "name": "Qwen 3.8 Flash",
    "provider": "Qwen",
    "release_date": "2026-09",
    "context_window": 1000000,
    "max_output": 8192,
    "input_cost": 0.15,
    "output_cost": 0.47,
    "cache_read": 0.016,
    "cache_write": null,
    "reasoning": false,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "High-efficiency multimodal lightweight model from Qwen (September 2026) with 1M context window at sub-$0.50 output pricing.",
    "popular": true,
    "popularity_rank": 20
  },
  {
    "id": "nemotron-3-ultra",
    "name": "Nemotron 3 Ultra",
    "provider": "NVIDIA",
    "release_date": "2026-06",
    "context_window": 256000,
    "max_output": 8192,
    "input_cost": 0.6,
    "output_cost": 2.4,
    "cache_read": 0.12,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text"
    ],
    "description": "NVIDIA's flagship open-weights model scoring 48 on Artificial Analysis Intelligence Index.",
    "popular": true,
    "popularity_rank": 8
  },
  {
    "id": "gemini-4-argon",
    "name": "Gemini 4 Argon",
    "provider": "Google",
    "release_date": "2026-10-13",
    "context_window": 1048576,
    "max_output": 65536,
    "input_cost": 2,
    "output_cost": 10,
    "cache_read": 0.1,
    "cache_write": 0.1875,
    "reasoning": true,
    "modalities": [
      "text",
      "vision",
      "audio",
      "video"
    ],
    "description": "Google's Gemini 4 generation flagship (October 13, 2026) at $2/$10 per 1M tokens with 1M context, native multimodal input and $0.10 cache reads.",
    "popular": true,
    "popularity_rank": 12
  },
  {
    "id": "mimo-v2-6-pro",
    "name": "MiMo-V2.6 Pro",
    "provider": "Xiaomi",
    "release_date": "2026-09-21",
    "context_window": 1050000,
    "max_output": 131072,
    "input_cost": 0.435,
    "output_cost": 0.87,
    "cache_read": 0.0036,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Xiaomi's MIT-licensed open-weights flagship (September 21, 2026) and the top open-weights model on the Artificial Analysis index.",
    "popular": true,
    "popularity_rank": 15
  },
  {
    "id": "glm-5-3",
    "name": "GLM 5.3",
    "provider": "Zhipu AI",
    "release_date": "2026-07",
    "context_window": 1310720,
    "max_output": 16384,
    "input_cost": 0.0699,
    "output_cost": 1.4,
    "cache_read": 0.0689,
    "cache_write": 0.8775,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Zhipu AI's flagship general-purpose foundation model with 1.31M context and top Chinese-English benchmarks, repriced in October 2026 to $0.079/$1.40 per 1M tokens.",
    "popular": true,
    "popularity_rank": 6
  },
  {
    "id": "deepseek-v4-pro",
    "name": "DeepSeek V4 Pro",
    "provider": "DeepSeek",
    "release_date": "2026-04",
    "context_window": 1048576,
    "max_output": 32000,
    "input_cost": 0.955,
    "output_cost": 1.911,
    "cache_read": 0.0796,
    "cache_write": 0.948,
    "reasoning": true,
    "modalities": [
      "text"
    ],
    "description": "DeepSeek's frontier MoE open architecture model with native long-context reasoning.",
    "popular": false,
    "popularity_rank": 27
  },
  {
    "id": "llama-4-scout",
    "name": "Llama 4 Scout",
    "provider": "Meta",
    "release_date": "2026-05",
    "context_window": 1300000,
    "max_output": 8192,
    "input_cost": 0.1,
    "output_cost": 0.3,
    "cache_read": null,
    "cache_write": null,
    "reasoning": false,
    "modalities": [
      "text"
    ],
    "description": "Meta's flagship open-weights release with 1.3M context window and leading developer adoption.",
    "popular": true,
    "popularity_rank": 23
  },
  {
    "id": "gemini-3-8-flash",
    "name": "Gemini 3.8 Flash",
    "provider": "Google",
    "release_date": "2026-09-02",
    "context_window": 1048576,
    "max_output": 65536,
    "input_cost": 0.825,
    "output_cost": 4.125,
    "cache_read": 0.0825,
    "cache_write": 0.0417,
    "reasoning": true,
    "modalities": [
      "text",
      "vision",
      "audio",
      "video"
    ],
    "description": "Google's multimodal Flash tier (September 2, 2026) at $0.825/$4.125 per 1M tokens after the introductory $0.75/$3.75 pricing ended.",
    "popular": true,
    "popularity_rank": 22
  },
  {
    "id": "grok-4-7",
    "name": "Grok 4.7",
    "provider": "xAI",
    "release_date": "2026-09-21",
    "context_window": 500000,
    "max_output": 16384,
    "input_cost": 2,
    "output_cost": 6,
    "cache_read": 0.5,
    "cache_write": 2,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "xAI's current flagship (September 21, 2026) holding Grok 4.6's $2/$6 rate card with a 500K context window.",
    "popular": true,
    "popularity_rank": 24
  },
  {
    "id": "glm-5-3-prime",
    "name": "GLM 5.3 Prime",
    "provider": "Zhipu AI",
    "release_date": "2026-09-23",
    "context_window": 1000000,
    "max_output": 131072,
    "input_cost": 2.8,
    "output_cost": 8.8,
    "cache_read": 0.56,
    "cache_write": 2.8,
    "reasoning": true,
    "modalities": [
      "text"
    ],
    "description": "Z.ai's high-speed reasoning variant of GLM-5.3 (September 23, 2026) delivering 1.5-2x throughput with 1M context and 128K output tokens.",
    "popular": false,
    "popularity_rank": 28
  },
  {
    "id": "qwen-3-8-max",
    "name": "Qwen 3.8 Max",
    "provider": "Qwen",
    "release_date": "2026-09-02",
    "context_window": 1000000,
    "max_output": 16384,
    "input_cost": 2,
    "output_cost": 6,
    "cache_read": 0.25,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Alibaba Qwen's frontier flagship model (September 2026) featuring preserved thinking traces, 1M context, and top agentic performance.",
    "popular": true,
    "popularity_rank": 14
  },
  {
    "id": "claude-sonnet-4-6",
    "name": "Claude Sonnet 4.6",
    "provider": "Anthropic",
    "release_date": "2026-02",
    "context_window": 1000000,
    "max_output": 32000,
    "input_cost": 3,
    "output_cost": 15,
    "cache_read": 0.3,
    "cache_write": 3.75,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Anthropic's high-efficiency 1M context Sonnet generation.",
    "popular": false,
    "popularity_rank": 33
  },
  {
    "id": "gemini-3-5-flash-lite",
    "name": "Gemini 3.5 Flash-Lite",
    "provider": "Google",
    "release_date": "2026-04",
    "context_window": 1000000,
    "max_output": 8192,
    "input_cost": 0.3,
    "output_cost": 2.5,
    "cache_read": 0.03,
    "cache_write": 0.0833,
    "reasoning": false,
    "modalities": [
      "text",
      "vision",
      "audio",
      "video"
    ],
    "description": "Google's ultra-lightweight 1M context model with 362 tokens/sec median throughput.",
    "popular": false,
    "popularity_rank": 32
  },
  {
    "id": "ling-3-0-flash",
    "name": "Ling 3.0 Flash",
    "provider": "InclusionAI",
    "release_date": "2026-07",
    "context_window": 262144,
    "max_output": 8192,
    "input_cost": 0.021,
    "output_cost": 0.063,
    "cache_read": 0.0042,
    "cache_write": null,
    "reasoning": false,
    "modalities": [
      "text"
    ],
    "description": "Ultra-low cost high-throughput Asian LLM at $0.021/1M tokens with $0.0042 cache reads.",
    "popular": false,
    "popularity_rank": 31
  },
  {
    "id": "llama-4-maverick",
    "name": "Llama 4 Maverick",
    "provider": "Meta",
    "release_date": "2026-05",
    "context_window": 1000000,
    "max_output": 8192,
    "input_cost": 0.1875,
    "output_cost": 0.6525,
    "cache_read": 0.05,
    "cache_write": null,
    "reasoning": false,
    "modalities": [
      "text"
    ],
    "description": "Meta's balanced open-weight model with 1M context window and broad tool-use support.",
    "popular": false,
    "popularity_rank": 29
  },
  {
    "id": "claude-haiku-5-5",
    "name": "Claude Haiku 5.5",
    "provider": "Anthropic",
    "release_date": "2026-10-07",
    "context_window": 1000000,
    "max_output": 64000,
    "input_cost": 0.1,
    "output_cost": 0.5,
    "cache_read": 0.01,
    "cache_write": 0.125,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Anthropic's small, fast model for high-volume cost-sensitive work (October 7, 2026) at $0.10/$0.50 per 1M tokens with 1M context and adjustable reasoning effort.",
    "popular": true,
    "popularity_rank": 19
  },
  {
    "id": "mistral-large-4",
    "name": "Mistral Large 4",
    "provider": "Mistral",
    "release_date": "2026-10-06",
    "context_window": 1048576,
    "max_output": 65536,
    "input_cost": 1.36,
    "output_cost": 4.18,
    "cache_read": 0.14,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "Mistral AI's frontier multimodal flagship (October 6, 2026) with a 1M context window. The public-preview discount ended, so the live list rate is $1.36/$4.18 per 1M tokens with $0.14 cache reads.",
    "popular": false,
    "popularity_rank": 26
  },
  {
    "id": "nemotron-3-5-super-vl",
    "name": "Nemotron 3.5 Super VL",
    "provider": "NVIDIA",
    "release_date": "2026-10-15",
    "context_window": 1048576,
    "max_output": 32768,
    "input_cost": 0.1,
    "output_cost": 0.4,
    "cache_read": 0.08,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text",
      "vision"
    ],
    "description": "NVIDIA's open-weights vision-language model (October 15, 2026) at $0.10/$0.40 per 1M tokens with 1M context.",
    "popular": false,
    "popularity_rank": 25
  },
  {
    "id": "muse-wonder",
    "name": "Muse Wonder",
    "provider": "Meta",
    "release_date": "2026-10-12",
    "context_window": 1048576,
    "max_output": 131072,
    "input_cost": 4,
    "output_cost": 20,
    "cache_read": 0.2,
    "cache_write": null,
    "reasoning": true,
    "modalities": [
      "text",
      "vision",
      "audio"
    ],
    "description": "Meta's new frontier-tier multimodal model (October 12, 2026) at $4/$20 per 1M tokens with 1M context, sitting above Muse Spark in the Muse family.",
    "popular": true,
    "popularity_rank": 21
  },
  {
    "id": "ling-3-1-flash",
    "name": "Ling 3.1 Flash",
    "provider": "InclusionAI",
    "release_date": "2026-10",
    "context_window": 262144,
    "max_output": 8192,
    "input_cost": 0.075,
    "output_cost": 0.225,
    "cache_read": 0.015,
    "cache_write": null,
    "reasoning": false,
    "modalities": [
      "text"
    ],
    "description": "InclusionAI's high-throughput efficiency LLM (October 2026), now publicly priced at $0.075/$0.225 per 1M tokens with $0.015 cache reads.",
    "popular": false,
    "popularity_rank": 34
  }
]