{
  "source": "https://inferencerate.io",
  "updated_at": "2026-10-16T08:21:21.871650+00:00",
  "stats": {
    "total_models_tracked": 37,
    "total_comparisons_indexed": 666,
    "total_providers_monitored": 10,
    "total_internal_links": 8873,
    "cheapest_input_model": "Gemini 3.5 Flash-Lite",
    "cheapest_input_model_id": "gemini-3-5-flash-lite",
    "cheapest_input_cost": 0.075,
    "fastest_model": "Gemini 3.5 Flash-Lite",
    "fastest_model_id": "gemini-3-5-flash-lite",
    "fastest_ttft_ms": 78,
    "highest_tps_model": "Gemini 3.5 Flash-Lite",
    "highest_tps_score": 265.0,
    "highest_ivr_model": "Qwen 3.8 Flash Next",
    "highest_ivr_model_id": "qwen-3-8-flash-next",
    "highest_ivr_score": 85.1,
    "highest_swe_model": "Claude Opus 5.5",
    "highest_swe_model_id": "claude-opus-5-5",
    "highest_swe_score": 89.9,
    "highest_gpqa_model": "GPT-6 Astra",
    "highest_gpqa_score": 96.0,
    "highest_math_model": "GPT-6.1 Sol",
    "highest_math_score": 98.8,
    "best_open_weight_model": "Qwen 3.8 Flash Next",
    "best_open_weight_id": "qwen-3-8-flash-next",
    "best_open_weight_ivr": 85.1,
    "pareto_optimal_count": 5,
    "trending_comparison_slugs": [
      "deepseek-r1-vs-o3-mini",
      "claude-3-5-sonnet-vs-gpt-4o",
      "claude-4-5-sonnet-vs-gemini-4-argon",
      "claude-opus-5-5-vs-gemini-4-argon",
      "gemini-3-8-flash-vs-gemini-4-argon",
      "gemini-4-argon-vs-gpt-5",
      "gemini-4-argon-vs-gpt-6-astra",
      "deepseek-v4-pro-vs-gemini-4-argon",
      "claude-4-5-sonnet-vs-claude-opus-5-5",
      "claude-4-5-sonnet-vs-gemini-3-8-flash",
      "claude-4-5-sonnet-vs-gpt-5",
      "claude-4-5-sonnet-vs-gpt-6-astra"
    ],
    "active_agent_model": "gemini-4-argon",
    "last_build_timestamp": "2026-10-16T08:21:21.871650+00:00"
  },
  "pareto_frontier_ids": [
    "gemini-3-5-flash-lite",
    "qwen-3-8-flash-next",
    "gpt-6-1-sol",
    "claude-opus-5-5",
    "gemma-4-31b-it"
  ],
  "top_ivr_models": [
    {
      "id": "qwen-3-8-flash-next",
      "name": "Qwen 3.8 Flash Next",
      "organization": "Alibaba Cloud",
      "ivr_score": 85.1,
      "input_cost_per_m": 0.16,
      "output_cost_per_m": 0.47,
      "swe_bench": 81.0,
      "gpqa_diamond": 91.7,
      "avg_ttft_ms": 280,
      "tokens_per_sec": 185.0
    },
    {
      "id": "deepseek-v4-flash-vision-exp",
      "name": "DeepSeek V4 Flash Vision (Exp)",
      "organization": "DeepSeek",
      "ivr_score": 81.6,
      "input_cost_per_m": 0.22,
      "output_cost_per_m": 0.65,
      "swe_bench": 77.4,
      "gpqa_diamond": 91.3,
      "avg_ttft_ms": 310,
      "tokens_per_sec": 165.0
    },
    {
      "id": "gemini-3-5-flash-lite",
      "name": "Gemini 3.5 Flash-Lite",
      "organization": "Google",
      "ivr_score": 81.5,
      "input_cost_per_m": 0.075,
      "output_cost_per_m": 0.3,
      "swe_bench": 55.4,
      "gpqa_diamond": 68.4,
      "avg_ttft_ms": 78,
      "tokens_per_sec": 265.0
    },
    {
      "id": "gemma-4-31b-it",
      "name": "Gemma 4 31B IT",
      "organization": "Google",
      "ivr_score": 80.3,
      "input_cost_per_m": 0.12,
      "output_cost_per_m": 0.36,
      "swe_bench": 59.2,
      "gpqa_diamond": 71.2,
      "avg_ttft_ms": 115,
      "tokens_per_sec": 155.0
    },
    {
      "id": "gemini-2-0-flash",
      "name": "Gemini 2.0 Flash",
      "organization": "Google DeepMind",
      "ivr_score": 77.7,
      "input_cost_per_m": 0.1,
      "output_cost_per_m": 0.4,
      "swe_bench": 51.8,
      "gpqa_diamond": 63.6,
      "avg_ttft_ms": 280,
      "tokens_per_sec": 185.0
    },
    {
      "id": "llama-4-maverick",
      "name": "Llama 4 Maverick",
      "organization": "Meta",
      "ivr_score": 77.0,
      "input_cost_per_m": 0.2,
      "output_cost_per_m": 0.6,
      "swe_bench": 62.4,
      "gpqa_diamond": 73.8,
      "avg_ttft_ms": 135,
      "tokens_per_sec": 175.0
    },
    {
      "id": "glm-5-3-flash",
      "name": "GLM-5.3-Flash",
      "organization": "Zhipu AI",
      "ivr_score": 76.7,
      "input_cost_per_m": 0.12,
      "output_cost_per_m": 0.4,
      "swe_bench": 52.8,
      "gpqa_diamond": 63.5,
      "avg_ttft_ms": 90,
      "tokens_per_sec": 175.0
    },
    {
      "id": "deepseek-v3",
      "name": "DeepSeek V3",
      "organization": "DeepSeek",
      "ivr_score": 75.0,
      "input_cost_per_m": 0.14,
      "output_cost_per_m": 0.28,
      "swe_bench": 42.0,
      "gpqa_diamond": 59.1,
      "avg_ttft_ms": 410,
      "tokens_per_sec": 78.0
    },
    {
      "id": "qwen-3-8-27b",
      "name": "Qwen 3.8 27B",
      "organization": "Alibaba Cloud",
      "ivr_score": 74.9,
      "input_cost_per_m": 0.2,
      "output_cost_per_m": 0.6,
      "swe_bench": 58.4,
      "gpqa_diamond": 68.0,
      "avg_ttft_ms": 120,
      "tokens_per_sec": 140.0
    },
    {
      "id": "llama-3-3-70b",
      "name": "Llama 3.3 70B Instruct",
      "organization": "Meta",
      "ivr_score": 74.6,
      "input_cost_per_m": 0.12,
      "output_cost_per_m": 0.3,
      "swe_bench": 42.5,
      "gpqa_diamond": 55.6,
      "avg_ttft_ms": 145,
      "tokens_per_sec": 165.0
    },
    {
      "id": "gemini-3-8-flash",
      "name": "Gemini 3.8 Flash",
      "organization": "Google",
      "ivr_score": 74.0,
      "input_cost_per_m": 0.25,
      "output_cost_per_m": 1.5,
      "swe_bench": 68.5,
      "gpqa_diamond": 79.2,
      "avg_ttft_ms": 105,
      "tokens_per_sec": 210.0
    },
    {
      "id": "glm-5-3",
      "name": "GLM-5.3",
      "organization": "Zhipu AI",
      "ivr_score": 69.9,
      "input_cost_per_m": 0.35,
      "output_cost_per_m": 1.0,
      "swe_bench": 56.0,
      "gpqa_diamond": 67.0,
      "avg_ttft_ms": 160,
      "tokens_per_sec": 120.0
    },
    {
      "id": "grok-3-mini",
      "name": "Grok 3 Mini",
      "organization": "xAI",
      "ivr_score": 69.2,
      "input_cost_per_m": 0.3,
      "output_cost_per_m": 1.2,
      "swe_bench": 54.0,
      "gpqa_diamond": 68.0,
      "avg_ttft_ms": 150,
      "tokens_per_sec": 160.0
    },
    {
      "id": "gpt-4o-mini",
      "name": "GPT-4o Mini",
      "organization": "OpenAI",
      "ivr_score": 67.2,
      "input_cost_per_m": 0.15,
      "output_cost_per_m": 0.6,
      "swe_bench": 41.0,
      "gpqa_diamond": 40.2,
      "avg_ttft_ms": 220,
      "tokens_per_sec": 140.0
    },
    {
      "id": "deepseek-v4-pro",
      "name": "DeepSeek V4 Pro",
      "organization": "DeepSeek",
      "ivr_score": 67.0,
      "input_cost_per_m": 1.32,
      "output_cost_per_m": 3.96,
      "swe_bench": 80.6,
      "gpqa_diamond": 90.1,
      "avg_ttft_ms": 520,
      "tokens_per_sec": 78.0
    }
  ]
}