{ "anchors": { "arena_search_rating": { "a": 1050, "b": 1300, "direction": "higher" } }, "as_of": "2026-11-05", "attributions": [ { "source_id": "arena-leaderboard-dataset", "text": "Arena data: Arena (lmarena-ai), dataset lmarena-ai/leaderboard-dataset, https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset, licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Changes by Inadviser: the Arena scores (the Agent Arena IPS score and the text, text-to-image, text-to-video and search ratings) are rescaled to a 0-100 utility against fixed anchors set by Inadviser; the confidence interval and the observation or vote count are carried over; only a selection of models is shown on each board; where a board shows Arena's cost per task, it is shown unchanged. The dataset maintainers' method applies (https://arena.ai/blog/agent-arena-methodology/ for Agent Arena). Arena does not endorse Inadviser." } ], "board": { "category": "ai-search", "cost": { "kind": "metric", "label": "Cost of 1,000 queries", "metric": "cost_search_per_1000_queries", "unit": "USD per 1,000 queries" }, "half_life_days": 30, "job": "Answer questions with live web search through a vendor's own API, and judge the model by how often people prefer its answers in blind comparisons on Arena, against what 1,000 searches cost.", "price_limit_days": 30, "products": [ { "id": "gpt-5-6-sol-xhigh", "name": "gpt-5.6-sol-xhigh", "vendor": "openai" }, { "id": "qwen3-8-max", "name": "qwen3.8-max", "vendor": "alibaba" }, { "id": "claude-opus-4-6-search", "name": "claude-opus-4-6-search", "vendor": "anthropic" }, { "id": "gpt-5-5-search", "name": "gpt-5.5-search", "vendor": "openai" }, { "id": "claude-fable-5-1-high", "name": "claude-fable-5.1-high", "vendor": "anthropic" }, { "id": "claude-opus-5-high", "name": "claude-opus-5-high", "vendor": "anthropic" }, { "id": "gpt-6-astra-max", "name": "gpt-6-astra-max", "vendor": "openai" }, { "id": "gpt-6-1-sol-max", "name": "gpt-6.1-sol-max", "vendor": "openai" }, { "id": "claude-opus-4-7", "name": "claude-opus-4-7", "vendor": "anthropic" }, { "id": "claude-fable-5-high", "name": "claude-fable-5-high", "vendor": "anthropic" }, { "id": "ernie-5-1", "name": "ernie-5.1", "vendor": "baidu" }, { "id": "claude-sonnet-4-6-search", "name": "claude-sonnet-4-6-search", "vendor": "anthropic" }, { "id": "grok-4-5", "name": "grok-4.5", "vendor": "xai" }, { "id": "claude-opus-5-5-medium", "name": "claude-opus-5.5-medium", "vendor": "anthropic" }, { "id": "gemini-3-1-pro-grounding", "name": "gemini-3.1-pro-grounding", "vendor": "google" }, { "id": "gpt-5-1-search", "name": "gpt-5.1-search", "vendor": "openai" }, { "id": "gemini-3-flash-grounding", "name": "gemini-3-flash-grounding", "vendor": "google" }, { "id": "claude-sonnet-5-search", "name": "claude-sonnet-5-search", "vendor": "anthropic" }, { "id": "gpt-6-luna-max", "name": "gpt-6-luna-max", "vendor": "openai" } ], "question": "Which AI search model gives the most quality for the money?", "region": "global", "requirements": {}, "slug": "ai-search", "title": "AI search models", "workload": { "chars_per_month": null, "unit_note": "Cost of 1,000 queries. Each query makes one search, sends 3,000 input tokens (the question and the retrieved content) and returns 400 output tokens. The cost is the search fee per 1,000 plus the tokens at the model's list price. Free monthly search allowances, caching and the extra searches an agent may run are not counted. The search fees of xAI and the other vendors without a read price are unknown, so those models have unknown cost." } }, "code_commit": "6585be527486036d41ec57c349ffe2a97aa8cd24", "decision": { "disqualified": [], "leader": "gpt-5-6-sol-xhigh", "margin": 3, "needed": [ { "metric": "arena_search_rating", "need": "another independent origin or repeats", "product": "claude-fable-5-1-high", "type": "narrow_interval", "width": 5.642197 }, { "metric": "arena_search_rating", "need": "another independent origin or repeats", "product": "claude-opus-4-6-search", "type": "narrow_interval", "width": 3.631285 }, { "metric": "arena_search_rating", "need": "another independent origin or repeats", "product": "claude-opus-5-high", "type": "narrow_interval", "width": 5.496899 }, { "metric": "arena_search_rating", "need": "another independent origin or repeats", "product": "gpt-5-5-search", "type": "narrow_interval", "width": 3.920321 }, { "metric": "arena_search_rating", "need": "another independent origin or repeats", "product": "gpt-5-6-sol-xhigh", "type": "narrow_interval", "width": 5.032124 }, { "metric": "arena_search_rating", "need": "another independent origin or repeats", "product": "qwen3-8-max", "type": "narrow_interval", "width": 10.547231 } ], "reasons": [ { "code": "bounds_overlap", "gap": 0.77697, "other": "claude-fable-5-1-high", "product": "gpt-5-6-sol-xhigh", "required": 3 }, { "code": "bounds_overlap", "gap": -0.426706, "other": "claude-opus-4-6-search", "product": "gpt-5-6-sol-xhigh", "required": 3 }, { "code": "bounds_overlap", "gap": 2.560767, "other": "claude-opus-5-high", "product": "gpt-5-6-sol-xhigh", "required": 3 }, { "code": "bounds_overlap", "gap": 1.354281, "other": "gpt-5-5-search", "product": "gpt-5-6-sol-xhigh", "required": 3 }, { "code": "bounds_overlap", "gap": -7.767848, "other": "qwen3-8-max", "product": "gpt-5-6-sol-xhigh", "required": 3 } ], "shortlist": [ "gpt-5-6-sol-xhigh", "qwen3-8-max", "claude-opus-4-6-search", "gpt-5-5-search", "claude-fable-5-1-high", "claude-opus-5-high" ], "unmeasured": [], "verdict": "shortlist", "winner": null }, "evidence_ids": [ "arena-search.claude-fable-5-1-high.arena_search_rating.2026-10-13", "arena-search.claude-fable-5-high.arena_search_rating.2026-10-13", "arena-search.claude-opus-4-6-search.arena_search_rating.2026-10-13", "arena-search.claude-opus-4-7.arena_search_rating.2026-10-13", "arena-search.claude-opus-5-5-medium.arena_search_rating.2026-10-13", "arena-search.claude-opus-5-high.arena_search_rating.2026-10-13", "arena-search.claude-sonnet-4-6-search.arena_search_rating.2026-10-13", "arena-search.claude-sonnet-5-search.arena_search_rating.2026-10-13", "arena-search.ernie-5-1.arena_search_rating.2026-10-13", "arena-search.gemini-3-1-pro-grounding.arena_search_rating.2026-10-13", "arena-search.gemini-3-flash-grounding.arena_search_rating.2026-10-13", "arena-search.gpt-5-1-search.arena_search_rating.2026-10-13", "arena-search.gpt-5-5-search.arena_search_rating.2026-10-13", "arena-search.gpt-5-6-sol-xhigh.arena_search_rating.2026-10-13", "arena-search.gpt-6-1-sol-max.arena_search_rating.2026-10-13", "arena-search.gpt-6-astra-max.arena_search_rating.2026-10-13", "arena-search.gpt-6-luna-max.arena_search_rating.2026-10-13", "arena-search.grok-4-5.arena_search_rating.2026-10-13", "arena-search.qwen3-8-max.arena_search_rating.2026-10-13", "price-search.claude-fable-5-1-high.cost_search_per_1000_queries.2026-11-05", "price-search.claude-fable-5-high.cost_search_per_1000_queries.2026-11-05", "price-search.claude-opus-4-6-search.cost_search_per_1000_queries.2026-11-05", "price-search.claude-opus-4-7.cost_search_per_1000_queries.2026-11-05", "price-search.claude-opus-5-5-medium.cost_search_per_1000_queries.2026-11-05", "price-search.claude-opus-5-high.cost_search_per_1000_queries.2026-11-05", "price-search.claude-sonnet-4-6-search.cost_search_per_1000_queries.2026-11-05", "price-search.claude-sonnet-5-search.cost_search_per_1000_queries.2026-11-05", "price-search.gemini-3-1-pro-grounding.cost_search_per_1000_queries.2026-11-05", "price-search.gemini-3-flash-grounding.cost_search_per_1000_queries.2026-11-05", "price-search.gpt-5-1-search.cost_search_per_1000_queries.2026-11-05", "price-search.gpt-5-5-search.cost_search_per_1000_queries.2026-11-05", "price-search.gpt-5-6-sol-xhigh.cost_search_per_1000_queries.2026-11-05", "price-search.gpt-6-1-sol-max.cost_search_per_1000_queries.2026-11-05", "price-search.gpt-6-astra-max.cost_search_per_1000_queries.2026-11-05", "price-search.gpt-6-luna-max.cost_search_per_1000_queries.2026-11-05" ], "pareto": { "dominated": [ { "dominated_by": [ "claude-opus-4-6-search", "gpt-5-5-search", "gpt-5-6-sol-xhigh" ], "id": "claude-fable-5-1-high" }, { "dominated_by": [ "claude-fable-5-1-high", "claude-opus-4-6-search", "claude-opus-4-7", "claude-opus-5-high", "gpt-5-5-search", "gpt-5-6-sol-xhigh", "gpt-6-1-sol-max", "gpt-6-astra-max" ], "id": "claude-fable-5-high" }, { "dominated_by": [ "gpt-5-6-sol-xhigh" ], "id": "claude-opus-4-6-search" }, { "dominated_by": [ "claude-opus-4-6-search", "claude-opus-5-high", "gpt-5-6-sol-xhigh", "gpt-6-1-sol-max" ], "id": "claude-opus-4-7" }, { "dominated_by": [ "claude-sonnet-4-6-search", "gpt-5-6-sol-xhigh", "gpt-6-1-sol-max" ], "id": "claude-opus-5-5-medium" }, { "dominated_by": [ "claude-opus-4-6-search", "gpt-5-6-sol-xhigh" ], "id": "claude-opus-5-high" }, { "dominated_by": [ "gpt-6-1-sol-max" ], "id": "claude-sonnet-4-6-search" }, { "dominated_by": [ "gemini-3-flash-grounding", "gpt-5-1-search", "gpt-6-1-sol-max" ], "id": "claude-sonnet-5-search" }, { "dominated_by": [ "gpt-6-1-sol-max" ], "id": "gemini-3-1-pro-grounding" }, { "dominated_by": [ "claude-opus-4-6-search", "gpt-5-6-sol-xhigh" ], "id": "gpt-5-5-search" }, { "dominated_by": [ "claude-fable-5-1-high", "claude-opus-4-6-search", "claude-opus-5-high", "gpt-5-5-search", "gpt-5-6-sol-xhigh" ], "id": "gpt-6-astra-max" } ], "excluded": [ { "id": "ernie-5-1", "reason": "unknown_cost" }, { "id": "grok-4-5", "reason": "unknown_cost" }, { "id": "qwen3-8-max", "reason": "unknown_cost" } ], "frontier": [ "gpt-6-luna-max", "gemini-3-flash-grounding", "gpt-5-1-search", "gpt-6-1-sol-max", "gpt-5-6-sol-xhigh" ] }, "prices": [ { "confirmed": true, "cost": 30, "plan_id": "price-search.gpt-5-6-sol-xhigh.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "gpt-5-6-sol-xhigh", "retrieved_at": "2026-11-05", "source_url": "https://developers.openai.com/api/docs/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 35, "plan_id": "price-search.claude-opus-4-6-search.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "claude-opus-4-6-search", "retrieved_at": "2026-11-05", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 37, "plan_id": "price-search.gpt-5-5-search.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "gpt-5-5-search", "retrieved_at": "2026-11-05", "source_url": "https://developers.openai.com/api/docs/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 60, "plan_id": "price-search.claude-fable-5-1-high.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "claude-fable-5-1-high", "retrieved_at": "2026-11-05", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 35, "plan_id": "price-search.claude-opus-5-high.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "claude-opus-5-high", "retrieved_at": "2026-11-05", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 60, "plan_id": "price-search.gpt-6-astra-max.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "gpt-6-astra-max", "retrieved_at": "2026-11-05", "source_url": "https://developers.openai.com/api/docs/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 20, "plan_id": "price-search.gpt-6-1-sol-max.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "gpt-6-1-sol-max", "retrieved_at": "2026-11-05", "source_url": "https://developers.openai.com/api/docs/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 35, "plan_id": "price-search.claude-opus-4-7.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "claude-opus-4-7", "retrieved_at": "2026-11-05", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 60, "plan_id": "price-search.claude-fable-5-high.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "claude-fable-5-high", "retrieved_at": "2026-11-05", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 25, "plan_id": "price-search.claude-sonnet-4-6-search.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "claude-sonnet-4-6-search", "retrieved_at": "2026-11-05", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 30, "plan_id": "price-search.claude-opus-5-5-medium.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "claude-opus-5-5-medium", "retrieved_at": "2026-11-05", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 24.8, "plan_id": "price-search.gemini-3-1-pro-grounding.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "gemini-3-1-pro-grounding", "retrieved_at": "2026-11-05", "source_url": "https://aistudio.google.com/docs/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 17.75, "plan_id": "price-search.gpt-5-1-search.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "gpt-5-1-search", "retrieved_at": "2026-11-05", "source_url": "https://developers.openai.com/api/docs/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 16.7, "plan_id": "price-search.gemini-3-flash-grounding.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "gemini-3-flash-grounding", "retrieved_at": "2026-11-05", "source_url": "https://aistudio.google.com/docs/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 20, "plan_id": "price-search.claude-sonnet-5-search.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "claude-sonnet-5-search", "retrieved_at": "2026-11-05", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing", "unit": "USD per 1,000 queries" }, { "confirmed": true, "cost": 10.5, "plan_id": "price-search.gpt-6-luna-max.cost_search_per_1000_queries.2026-11-05", "price_expired": false, "product": "gpt-6-luna-max", "retrieved_at": "2026-11-05", "source_url": "https://developers.openai.com/api/docs/pricing", "unit": "USD per 1,000 queries" } ], "products": [ { "cost": { "confirmed": true, "cost": 30, "currency": null, "flags": [], "plan": { "id": "price-search.gpt-5-6-sol-xhigh.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-openai-api", "source_url": "https://developers.openai.com/api/docs/pricing" }, "price_age_days": 0, "price_expired": false, "product": "gpt-5-6-sol-xhigh", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "gpt-5-6-sol-xhigh", "name": "gpt-5.6-sol-xhigh", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 86.968366, "key": "arena_search_rating", "kind": "ci95", "low": 81.936242, "missing_reason": null, "origins": 1, "present": true, "u_hat": 84.452304, "used": [ "arena-search.gpt-5-6-sol-xhigh.arena_search_rating.2026-10-13" ], "value": 1261.130759, "weight": 1, "weight_total": 0.293887, "width_contribution": 5.032124 } ], "complete": true, "high": 86.968366, "label": "ci95", "low": 81.936242, "missing": [], "q": 84.452304, "spread_unknown": [] }, "vendor": "openai" }, { "cost": { "confirmed": false, "cost": null, "currency": null, "flags": [], "plan": null, "price_age_days": null, "price_expired": false, "product": "qwen3-8-max", "rejected": [], "status": "no_price", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "qwen3-8-max", "name": "qwen3.8-max", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 89.70409, "key": "arena_search_rating", "kind": "ci95", "low": 79.156859, "missing_reason": null, "origins": 1, "present": true, "u_hat": 84.430474, "used": [ "arena-search.qwen3-8-max.arena_search_rating.2026-10-13" ], "value": 1261.076186, "weight": 1, "weight_total": 0.293887, "width_contribution": 10.547231 } ], "complete": true, "high": 89.70409, "label": "ci95", "low": 79.156859, "missing": [], "q": 84.430474, "spread_unknown": [] }, "vendor": "alibaba" }, { "cost": { "confirmed": true, "cost": 35, "currency": null, "flags": [], "plan": { "id": "price-search.claude-opus-4-6-search.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-anthropic-api", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing" }, "price_age_days": 0, "price_expired": false, "product": "claude-opus-4-6-search", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "claude-opus-4-6-search", "name": "claude-opus-4-6-search", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 82.362948, "key": "arena_search_rating", "kind": "ci95", "low": 78.731663, "missing_reason": null, "origins": 1, "present": true, "u_hat": 80.547305, "used": [ "arena-search.claude-opus-4-6-search.arena_search_rating.2026-10-13" ], "value": 1251.368263, "weight": 1, "weight_total": 0.293887, "width_contribution": 3.631285 } ], "complete": true, "high": 82.362948, "label": "ci95", "low": 78.731663, "missing": [], "q": 80.547305, "spread_unknown": [] }, "vendor": "anthropic" }, { "cost": { "confirmed": true, "cost": 37, "currency": null, "flags": [], "plan": { "id": "price-search.gpt-5-5-search.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-openai-api", "source_url": "https://developers.openai.com/api/docs/pricing" }, "price_age_days": 0, "price_expired": false, "product": "gpt-5-5-search", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "gpt-5-5-search", "name": "gpt-5.5-search", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 80.581961, "key": "arena_search_rating", "kind": "ci95", "low": 76.66164, "missing_reason": null, "origins": 1, "present": true, "u_hat": 78.6218, "used": [ "arena-search.gpt-5-5-search.arena_search_rating.2026-10-13" ], "value": 1246.554501, "weight": 1, "weight_total": 0.293887, "width_contribution": 3.920321 } ], "complete": true, "high": 80.581961, "label": "ci95", "low": 76.66164, "missing": [], "q": 78.6218, "spread_unknown": [] }, "vendor": "openai" }, { "cost": { "confirmed": true, "cost": 60, "currency": null, "flags": [], "plan": { "id": "price-search.claude-fable-5-1-high.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-anthropic-api", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing" }, "price_age_days": 0, "price_expired": false, "product": "claude-fable-5-1-high", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "claude-fable-5-1-high", "name": "claude-fable-5.1-high", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 81.159272, "key": "arena_search_rating", "kind": "ci95", "low": 75.517075, "missing_reason": null, "origins": 1, "present": true, "u_hat": 78.338174, "used": [ "arena-search.claude-fable-5-1-high.arena_search_rating.2026-10-13" ], "value": 1245.845434, "weight": 1, "weight_total": 0.293887, "width_contribution": 5.642197 } ], "complete": true, "high": 81.159272, "label": "ci95", "low": 75.517075, "missing": [], "q": 78.338174, "spread_unknown": [] }, "vendor": "anthropic" }, { "cost": { "confirmed": true, "cost": 35, "currency": null, "flags": [], "plan": { "id": "price-search.claude-opus-5-high.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-anthropic-api", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing" }, "price_age_days": 0, "price_expired": false, "product": "claude-opus-5-high", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "claude-opus-5-high", "name": "claude-opus-5-high", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 79.375475, "key": "arena_search_rating", "kind": "ci95", "low": 73.878576, "missing_reason": null, "origins": 1, "present": true, "u_hat": 76.627026, "used": [ "arena-search.claude-opus-5-high.arena_search_rating.2026-10-13" ], "value": 1241.567564, "weight": 1, "weight_total": 0.293887, "width_contribution": 5.496899 } ], "complete": true, "high": 79.375475, "label": "ci95", "low": 73.878576, "missing": [], "q": 76.627026, "spread_unknown": [] }, "vendor": "anthropic" }, { "cost": { "confirmed": true, "cost": 60, "currency": null, "flags": [], "plan": { "id": "price-search.gpt-6-astra-max.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-openai-api", "source_url": "https://developers.openai.com/api/docs/pricing" }, "price_age_days": 0, "price_expired": false, "product": "gpt-6-astra-max", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "gpt-6-astra-max", "name": "gpt-6-astra-max", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 76.583194, "key": "arena_search_rating", "kind": "ci95", "low": 70.309455, "missing_reason": null, "origins": 1, "present": true, "u_hat": 73.446324, "used": [ "arena-search.gpt-6-astra-max.arena_search_rating.2026-10-13" ], "value": 1233.615811, "weight": 1, "weight_total": 0.293887, "width_contribution": 6.273739 } ], "complete": true, "high": 76.583194, "label": "ci95", "low": 70.309455, "missing": [], "q": 73.446324, "spread_unknown": [] }, "vendor": "openai" }, { "cost": { "confirmed": true, "cost": 20, "currency": null, "flags": [], "plan": { "id": "price-search.gpt-6-1-sol-max.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-openai-api", "source_url": "https://developers.openai.com/api/docs/pricing" }, "price_age_days": 0, "price_expired": false, "product": "gpt-6-1-sol-max", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "gpt-6-1-sol-max", "name": "gpt-6.1-sol-max", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 76.785298, "key": "arena_search_rating", "kind": "ci95", "low": 69.696582, "missing_reason": null, "origins": 1, "present": true, "u_hat": 73.24094, "used": [ "arena-search.gpt-6-1-sol-max.arena_search_rating.2026-10-13" ], "value": 1233.10235, "weight": 1, "weight_total": 0.293887, "width_contribution": 7.088716 } ], "complete": true, "high": 76.785298, "label": "ci95", "low": 69.696582, "missing": [], "q": 73.24094, "spread_unknown": [] }, "vendor": "openai" }, { "cost": { "confirmed": true, "cost": 35, "currency": null, "flags": [], "plan": { "id": "price-search.claude-opus-4-7.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-anthropic-api", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing" }, "price_age_days": 0, "price_expired": false, "product": "claude-opus-4-7", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "claude-opus-4-7", "name": "claude-opus-4-7", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 74.004723, "key": "arena_search_rating", "kind": "ci95", "low": 70.112552, "missing_reason": null, "origins": 1, "present": true, "u_hat": 72.058638, "used": [ "arena-search.claude-opus-4-7.arena_search_rating.2026-10-13" ], "value": 1230.146594, "weight": 1, "weight_total": 0.293887, "width_contribution": 3.892171 } ], "complete": true, "high": 74.004723, "label": "ci95", "low": 70.112552, "missing": [], "q": 72.058638, "spread_unknown": [] }, "vendor": "anthropic" }, { "cost": { "confirmed": true, "cost": 60, "currency": null, "flags": [], "plan": { "id": "price-search.claude-fable-5-high.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-anthropic-api", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing" }, "price_age_days": 0, "price_expired": false, "product": "claude-fable-5-high", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "claude-fable-5-high", "name": "claude-fable-5-high", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 74.259797, "key": "arena_search_rating", "kind": "ci95", "low": 69.052004, "missing_reason": null, "origins": 1, "present": true, "u_hat": 71.655901, "used": [ "arena-search.claude-fable-5-high.arena_search_rating.2026-10-13" ], "value": 1229.139752, "weight": 1, "weight_total": 0.293887, "width_contribution": 5.207793 } ], "complete": true, "high": 74.259797, "label": "ci95", "low": 69.052004, "missing": [], "q": 71.655901, "spread_unknown": [] }, "vendor": "anthropic" }, { "cost": { "confirmed": false, "cost": null, "currency": null, "flags": [], "plan": null, "price_age_days": null, "price_expired": false, "product": "ernie-5-1", "rejected": [], "status": "no_price", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "ernie-5-1", "name": "ernie-5.1", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 74.474676, "key": "arena_search_rating", "kind": "ci95", "low": 66.897367, "missing_reason": null, "origins": 1, "present": true, "u_hat": 70.686021, "used": [ "arena-search.ernie-5-1.arena_search_rating.2026-10-13" ], "value": 1226.715053, "weight": 1, "weight_total": 0.293887, "width_contribution": 7.577309 } ], "complete": true, "high": 74.474676, "label": "ci95", "low": 66.897367, "missing": [], "q": 70.686021, "spread_unknown": [] }, "vendor": "baidu" }, { "cost": { "confirmed": true, "cost": 25, "currency": null, "flags": [], "plan": { "id": "price-search.claude-sonnet-4-6-search.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-anthropic-api", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing" }, "price_age_days": 0, "price_expired": false, "product": "claude-sonnet-4-6-search", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "claude-sonnet-4-6-search", "name": "claude-sonnet-4-6-search", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 69.991619, "key": "arena_search_rating", "kind": "ci95", "low": 66.412208, "missing_reason": null, "origins": 1, "present": true, "u_hat": 68.201914, "used": [ "arena-search.claude-sonnet-4-6-search.arena_search_rating.2026-10-13" ], "value": 1220.504784, "weight": 1, "weight_total": 0.293887, "width_contribution": 3.579411 } ], "complete": true, "high": 69.991619, "label": "ci95", "low": 66.412208, "missing": [], "q": 68.201914, "spread_unknown": [] }, "vendor": "anthropic" }, { "cost": { "confirmed": false, "cost": null, "currency": null, "flags": [], "plan": null, "price_age_days": null, "price_expired": false, "product": "grok-4-5", "rejected": [], "status": "no_price", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "grok-4-5", "name": "grok-4.5", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 68.717836, "key": "arena_search_rating", "kind": "ci95", "low": 64.03183, "missing_reason": null, "origins": 1, "present": true, "u_hat": 66.374833, "used": [ "arena-search.grok-4-5.arena_search_rating.2026-10-13" ], "value": 1215.937083, "weight": 1, "weight_total": 0.293887, "width_contribution": 4.686006 } ], "complete": true, "high": 68.717836, "label": "ci95", "low": 64.03183, "missing": [], "q": 66.374833, "spread_unknown": [] }, "vendor": "xai" }, { "cost": { "confirmed": true, "cost": 30, "currency": null, "flags": [], "plan": { "id": "price-search.claude-opus-5-5-medium.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-anthropic-api", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing" }, "price_age_days": 0, "price_expired": false, "product": "claude-opus-5-5-medium", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "claude-opus-5-5-medium", "name": "claude-opus-5.5-medium", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 67.936781, "key": "arena_search_rating", "kind": "ci95", "low": 62.265239, "missing_reason": null, "origins": 1, "present": true, "u_hat": 65.10101, "used": [ "arena-search.claude-opus-5-5-medium.arena_search_rating.2026-10-13" ], "value": 1212.752524, "weight": 1, "weight_total": 0.293887, "width_contribution": 5.671542 } ], "complete": true, "high": 67.936781, "label": "ci95", "low": 62.265239, "missing": [], "q": 65.10101, "spread_unknown": [] }, "vendor": "anthropic" }, { "cost": { "confirmed": true, "cost": 24.8, "currency": null, "flags": [], "plan": { "id": "price-search.gemini-3-1-pro-grounding.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-google-gemini-api", "source_url": "https://aistudio.google.com/docs/pricing" }, "price_age_days": 0, "price_expired": false, "product": "gemini-3-1-pro-grounding", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "gemini-3-1-pro-grounding", "name": "gemini-3.1-pro-grounding", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 65.393348, "key": "arena_search_rating", "kind": "ci95", "low": 61.651188, "missing_reason": null, "origins": 1, "present": true, "u_hat": 63.522268, "used": [ "arena-search.gemini-3-1-pro-grounding.arena_search_rating.2026-10-13" ], "value": 1208.80567, "weight": 1, "weight_total": 0.293887, "width_contribution": 3.74216 } ], "complete": true, "high": 65.393348, "label": "ci95", "low": 61.651188, "missing": [], "q": 63.522268, "spread_unknown": [] }, "vendor": "google" }, { "cost": { "confirmed": true, "cost": 17.75, "currency": null, "flags": [], "plan": { "id": "price-search.gpt-5-1-search.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-openai-api", "source_url": "https://developers.openai.com/api/docs/pricing" }, "price_age_days": 0, "price_expired": false, "product": "gpt-5-1-search", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "gpt-5-1-search", "name": "gpt-5.1-search", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 61.956356, "key": "arena_search_rating", "kind": "ci95", "low": 57.805024, "missing_reason": null, "origins": 1, "present": true, "u_hat": 59.88069, "used": [ "arena-search.gpt-5-1-search.arena_search_rating.2026-10-13" ], "value": 1199.701725, "weight": 1, "weight_total": 0.293887, "width_contribution": 4.151332 } ], "complete": true, "high": 61.956356, "label": "ci95", "low": 57.805024, "missing": [], "q": 59.88069, "spread_unknown": [] }, "vendor": "openai" }, { "cost": { "confirmed": true, "cost": 16.7, "currency": null, "flags": [], "plan": { "id": "price-search.gemini-3-flash-grounding.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-google-gemini-api", "source_url": "https://aistudio.google.com/docs/pricing" }, "price_age_days": 0, "price_expired": false, "product": "gemini-3-flash-grounding", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "gemini-3-flash-grounding", "name": "gemini-3-flash-grounding", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 59.998335, "key": "arena_search_rating", "kind": "ci95", "low": 56.557482, "missing_reason": null, "origins": 1, "present": true, "u_hat": 58.277909, "used": [ "arena-search.gemini-3-flash-grounding.arena_search_rating.2026-10-13" ], "value": 1195.694771, "weight": 1, "weight_total": 0.293887, "width_contribution": 3.440853 } ], "complete": true, "high": 59.998335, "label": "ci95", "low": 56.557482, "missing": [], "q": 58.277909, "spread_unknown": [] }, "vendor": "google" }, { "cost": { "confirmed": true, "cost": 20, "currency": null, "flags": [], "plan": { "id": "price-search.claude-sonnet-5-search.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-anthropic-api", "source_url": "https://platform.claude.com/docs/en/about-claude/pricing" }, "price_age_days": 0, "price_expired": false, "product": "claude-sonnet-5-search", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "claude-sonnet-5-search", "name": "claude-sonnet-5-search", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 58.840615, "key": "arena_search_rating", "kind": "ci95", "low": 53.911469, "missing_reason": null, "origins": 1, "present": true, "u_hat": 56.376042, "used": [ "arena-search.claude-sonnet-5-search.arena_search_rating.2026-10-13" ], "value": 1190.940105, "weight": 1, "weight_total": 0.293887, "width_contribution": 4.929146 } ], "complete": true, "high": 58.840615, "label": "ci95", "low": 53.911469, "missing": [], "q": 56.376042, "spread_unknown": [] }, "vendor": "anthropic" }, { "cost": { "confirmed": true, "cost": 10.5, "currency": null, "flags": [], "plan": { "id": "price-search.gpt-6-luna-max.cost_search_per_1000_queries.2026-11-05", "retrieved_at": "2026-11-05", "source_id": "pricing-openai-api", "source_url": "https://developers.openai.com/api/docs/pricing" }, "price_age_days": 0, "price_expired": false, "product": "gpt-6-luna-max", "rejected": [], "status": "confirmed", "unconfirmed_plans": [], "unit": "USD per 1,000 queries", "workload_chars": null }, "id": "gpt-6-luna-max", "name": "gpt-6-luna-max", "quality": { "breakdown": [ { "auditable_own": false, "conflicts": [ "unknown" ], "critical": true, "excluded": [], "high": 46.713265, "key": "arena_search_rating", "kind": "ci95", "low": 40.93223, "missing_reason": null, "origins": 1, "present": true, "u_hat": 43.822748, "used": [ "arena-search.gpt-6-luna-max.arena_search_rating.2026-10-13" ], "value": 1159.556869, "weight": 1, "weight_total": 0.293887, "width_contribution": 5.781035 } ], "complete": true, "high": 46.713265, "label": "ci95", "low": 40.93223, "missing": [], "q": 43.822748, "spread_unknown": [] }, "vendor": "openai" } ], "ranking": [ { "cost": 30, "cost_confirmed": true, "high": 86.968366, "label": "ci95", "low": 81.936242, "product": "gpt-5-6-sol-xhigh", "q": 84.452304, "rank": 1, "status": "active" }, { "cost": null, "cost_confirmed": false, "high": 89.70409, "label": "ci95", "low": 79.156859, "product": "qwen3-8-max", "q": 84.430474, "rank": 2, "status": "active" }, { "cost": 35, "cost_confirmed": true, "high": 82.362948, "label": "ci95", "low": 78.731663, "product": "claude-opus-4-6-search", "q": 80.547305, "rank": 3, "status": "active" }, { "cost": 37, "cost_confirmed": true, "high": 80.581961, "label": "ci95", "low": 76.66164, "product": "gpt-5-5-search", "q": 78.6218, "rank": 4, "status": "active" }, { "cost": 60, "cost_confirmed": true, "high": 81.159272, "label": "ci95", "low": 75.517075, "product": "claude-fable-5-1-high", "q": 78.338174, "rank": 5, "status": "active" }, { "cost": 35, "cost_confirmed": true, "high": 79.375475, "label": "ci95", "low": 73.878576, "product": "claude-opus-5-high", "q": 76.627026, "rank": 6, "status": "active" }, { "cost": 60, "cost_confirmed": true, "high": 76.583194, "label": "ci95", "low": 70.309455, "product": "gpt-6-astra-max", "q": 73.446324, "rank": 7, "status": "active" }, { "cost": 20, "cost_confirmed": true, "high": 76.785298, "label": "ci95", "low": 69.696582, "product": "gpt-6-1-sol-max", "q": 73.24094, "rank": 8, "status": "active" }, { "cost": 35, "cost_confirmed": true, "high": 74.004723, "label": "ci95", "low": 70.112552, "product": "claude-opus-4-7", "q": 72.058638, "rank": 9, "status": "active" }, { "cost": 60, "cost_confirmed": true, "high": 74.259797, "label": "ci95", "low": 69.052004, "product": "claude-fable-5-high", "q": 71.655901, "rank": 10, "status": "active" }, { "cost": null, "cost_confirmed": false, "high": 74.474676, "label": "ci95", "low": 66.897367, "product": "ernie-5-1", "q": 70.686021, "rank": 11, "status": "active" }, { "cost": 25, "cost_confirmed": true, "high": 69.991619, "label": "ci95", "low": 66.412208, "product": "claude-sonnet-4-6-search", "q": 68.201914, "rank": 12, "status": "active" }, { "cost": null, "cost_confirmed": false, "high": 68.717836, "label": "ci95", "low": 64.03183, "product": "grok-4-5", "q": 66.374833, "rank": 13, "status": "active" }, { "cost": 30, "cost_confirmed": true, "high": 67.936781, "label": "ci95", "low": 62.265239, "product": "claude-opus-5-5-medium", "q": 65.10101, "rank": 14, "status": "active" }, { "cost": 24.8, "cost_confirmed": true, "high": 65.393348, "label": "ci95", "low": 61.651188, "product": "gemini-3-1-pro-grounding", "q": 63.522268, "rank": 15, "status": "active" }, { "cost": 17.75, "cost_confirmed": true, "high": 61.956356, "label": "ci95", "low": 57.805024, "product": "gpt-5-1-search", "q": 59.88069, "rank": 16, "status": "active" }, { "cost": 16.7, "cost_confirmed": true, "high": 59.998335, "label": "ci95", "low": 56.557482, "product": "gemini-3-flash-grounding", "q": 58.277909, "rank": 17, "status": "active" }, { "cost": 20, "cost_confirmed": true, "high": 58.840615, "label": "ci95", "low": 53.911469, "product": "claude-sonnet-5-search", "q": 56.376042, "rank": 18, "status": "active" }, { "cost": 10.5, "cost_confirmed": true, "high": 46.713265, "label": "ci95", "low": 40.93223, "product": "gpt-6-luna-max", "q": 43.822748, "rank": 19, "status": "active" } ], "schema": 1, "seed": null, "site": { "board_note": "Arena data: Arena (lmarena-ai), dataset lmarena-ai/leaderboard-dataset, https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset, licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Changes by Inadviser: the Arena scores (the Agent Arena IPS score and the text, text-to-image, text-to-video and search ratings) are rescaled to a 0-100 utility against fixed anchors set by Inadviser; the confidence interval and the observation or vote count are carried over; only a selection of models is shown on each board; where a board shows Arena's cost per task, it is shown unchanged. The dataset maintainers' method applies (https://arena.ai/blog/agent-arena-methodology/ for Agent Arena). Arena does not endorse Inadviser. Only one origin (Arena) measures quality, so no board entry can be named Best until a second independent origin or an own test exists. Arena's rating comes from blind votes by its own users on their own prompts; it is not a test of your job. Prices are read from each vendor's own API price page and are list prices for the stated workload; a model whose price could not be read on the vendor's page is listed with unknown cost and is not plotted. Selection: the 15 models with the highest Arena rank in the overall category, plus every model ranked 16 to 42 whose price was read on vendor pages and whose output price is at most $10 per million tokens.", "first_paragraph": "As of 5 November 2026, no option is proven ahead. 6 options of 19 cannot be separated with the evidence in hand: gpt-5.6-sol-xhigh, qwen3.8-max, claude-opus-4-6-search, gpt-5.5-search, claude-fable-5.1-high and 1 more.", "interval_labels": { "ci95": "built from 95% confidence intervals given by the source" }, "ranking_is_quality_order": true, "ranking_note": "Order is by score point estimate; intervals may overlap.", "sources_used": [ { "checked_on": "2026-11-03", "id": "arena-leaderboard-dataset", "name": "Arena Leaderboard Dataset (lmarena-ai/leaderboard-dataset)", "status": "attribution", "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset" }, { "checked_on": "2026-11-05", "id": "pricing-anthropic-api", "name": "Claude API pricing", "status": "facts-only", "url": "https://platform.claude.com/docs/en/about-claude/pricing" }, { "checked_on": "2026-11-05", "id": "pricing-google-gemini-api", "name": "Gemini API pricing", "status": "facts-only", "url": "https://aistudio.google.com/docs/pricing" }, { "checked_on": "2026-11-05", "id": "pricing-openai-api", "name": "OpenAI API pricing", "status": "facts-only", "url": "https://developers.openai.com/api/docs/pricing" } ], "verdict_label": "Shortlist" }, "tariff_ids": [], "weights": { "arena_search_rating": 1 } }