{ "generatedAt": "2026-09-30T20:54:24.517Z", "method": { "url": "https://modelcap.ai/methodology", "indexMethodId": "modelcap-index-v4.12-fused-launch-placement", "evidenceAdmissionMethodId": "modelcap-score-v10-best-supported-configuration" }, "currentCount": 279, "allVersionsCount": 418, "pricedCount": 116, "medianPrice": 0.44375, "establishedCount": 78, "topTen": [ { "name": "Claude Sonnet 5.5", "slug": "anthropic-claude-sonnet-5-5", "rank": 1, "score": 91.9, "support": 20, "interval": { "lower": 72.4, "upper": 100 }, "state": "Measured", "access": "none", "price": 4.0, "endpoints": 8, "releasedAt": "2026-09-28T18:04:46.000Z", "families": { "general": 91.9, "coding": null, "agent": null, "reasoning": null }, "boards": [ "Artificial Analysis Intelligence Index" ] }, { "name": "Claude Opus 5.5", "slug": "anthropic-claude-opus-5-5", "rank": 2, "score": 88.4, "support": 65.4, "interval": { "lower": 79.9, "upper": 96.9 }, "state": "Measured", "access": "none", "price": 8.0, "endpoints": 11, "releasedAt": "2026-09-22T16:32:12.000Z", "families": { "general": 89.9, "coding": 83.9, "agent": 78.2, "reasoning": 84 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ] }, { "name": "GPT-6.1 Sol", "slug": "openai-gpt-6-1-sol", "rank": 3, "score": 88.2, "support": 23, "interval": { "lower": 68.9, "upper": 100 }, "state": "Measured", "access": "none", "price": 4.0, "endpoints": 6, "releasedAt": "2026-09-29T17:28:02.000Z", "families": { "general": 88.5, "coding": null, "agent": null, "reasoning": 78.2 }, "boards": [ "Artificial Analysis Intelligence Index", "ARC-AGI-2 Verified", "ARC-AGI-3 Verified" ] }, { "name": "Claude Fable 5.1", "slug": "anthropic-claude-fable-5-1", "rank": 4, "score": 86.9, "support": 83.2, "interval": { "lower": 80.7, "upper": 93.1 }, "state": "Measured", "access": "none", "price": 20.0, "endpoints": 4, "releasedAt": "2026-09-01T18:03:58.000Z", "families": { "general": 88.3, "coding": 81.8, "agent": 79.8, "reasoning": 80.8 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ] }, { "name": "Muse Spark 1.3", "slug": "meta-muse-spark-1-3", "rank": 5, "score": 84.7, "support": 82.4, "interval": { "lower": 78.4, "upper": 91 }, "state": "Measured", "access": "none", "price": 2.0, "endpoints": 1, "releasedAt": "2026-09-02T19:45:59.000Z", "families": { "general": 86.2, "coding": 84, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "GPT-6 Astra", "slug": "openai-gpt-6-astra", "rank": 6, "score": 83.6, "support": 81.1, "interval": { "lower": 77.1, "upper": 90.1 }, "state": "Measured", "access": "none", "price": 20.0, "endpoints": 7, "releasedAt": "2026-09-04T20:13:58.000Z", "families": { "general": 84, "coding": 84.2, "agent": 76.9, "reasoning": 82.2 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified", "ARC-AGI-3 Verified" ] }, { "name": "Qwen3.8 Max (0902)", "slug": "qwen-qwen3-8-max-0902", "rank": 7, "score": 83.3, "support": 20, "interval": { "lower": 63.8, "upper": 100 }, "state": "Measured", "access": "none", "price": 3.0, "endpoints": 1, "releasedAt": "2026-09-03T21:08:24.000Z", "families": { "general": 83.3, "coding": null, "agent": null, "reasoning": null }, "boards": [ "Artificial Analysis Intelligence Index" ] }, { "name": "MiMo-V2.6-Pro", "slug": "xiaomi-mimo-v2-6-pro", "rank": 8, "score": 83.2, "support": 70.8, "interval": { "lower": 75.7, "upper": 90.7 }, "state": "Measured", "access": "open", "price": 0.54375, "endpoints": 3, "releasedAt": "2026-09-21T20:07:39.000Z", "families": { "general": 83.1, "coding": 83.9, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, { "name": "Gemini 3.8 Flash", "slug": "google-gemini-3-8-flash", "rank": 9, "score": 82.7, "support": 88.6, "interval": { "lower": 77, "upper": 88.4 }, "state": "Measured", "access": "none", "price": 1.5, "endpoints": 6, "releasedAt": "2026-09-02T15:14:16.000Z", "families": { "general": 84.6, "coding": 82.9, "agent": 58.6, "reasoning": 68 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified", "ARC-AGI-3 Verified" ] }, { "name": "Kimi K3", "slug": "moonshotai-kimi-k3", "rank": 10, "score": 82, "support": 89, "interval": { "lower": 76.3, "upper": 87.7 }, "state": "Measured", "access": "restricted", "price": 6.0, "endpoints": 22, "releasedAt": "2026-07-16T15:30:58.000Z", "families": { "general": 83.6, "coding": 85, "agent": 62.2, "reasoning": 54.6 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "WildClawBench OpenClaw", "ARC-AGI-2 Verified" ] } ], "frontiers": { "all": [ { "name": "Dots3-Note Preview", "slug": "dots-studio-dots-3-note-preview", "rank": 73, "score": 60.1, "support": 2.1, "interval": { "lower": 37.5, "upper": 82.7 }, "state": "Measured · blended", "access": "none", "price": 0.0, "endpoints": 1, "releasedAt": "2026-08-14T04:06:01.000Z", "families": { "general": null, "coding": null, "agent": null, "reasoning": 65.5 }, "boards": [ "ARC-AGI-2 Verified" ] }, { "name": "Ling 3.0 Flash VL", "slug": "inclusionai-ling-3-0-flash-vl", "rank": 50, "score": 66.3, "support": 20, "interval": { "lower": 46.8, "upper": 85.8 }, "state": "Measured", "access": "open", "price": 0.031149999999999997, "endpoints": 2, "releasedAt": "2026-09-10T16:01:54.000Z", "families": { "general": 66.3, "coding": null, "agent": null, "reasoning": null }, "boards": [ "Artificial Analysis Intelligence Index" ] }, { "name": "Nex-N2.5-Mini", "slug": "nex-agi-nex-n2-5-mini", "rank": 32, "score": 71.4, "support": 23, "interval": { "lower": 62.6, "upper": 80.3 }, "state": "Modeled · peer benchmarks", "access": "open", "price": 0.04375, "endpoints": 1, "releasedAt": "2026-09-08T17:54:21.000Z", "families": null, "boards": [] }, { "name": "DeepSeek V4.1 Flash", "slug": "deepseek-deepseek-v4-1-flash", "rank": 12, "score": 79.9, "support": 78.2, "interval": { "lower": 73.2, "upper": 86.6 }, "state": "Measured", "access": "open", "price": 0.11385, "endpoints": 32, "releasedAt": "2026-09-10T06:21:25.000Z", "families": { "general": 80.8, "coding": 82.6, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "MiMo-V2.6-Pro", "slug": "xiaomi-mimo-v2-6-pro", "rank": 8, "score": 83.2, "support": 70.8, "interval": { "lower": 75.7, "upper": 90.7 }, "state": "Measured", "access": "open", "price": 0.54375, "endpoints": 3, "releasedAt": "2026-09-21T20:07:39.000Z", "families": { "general": 83.1, "coding": 83.9, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, { "name": "Muse Spark 1.3", "slug": "meta-muse-spark-1-3", "rank": 5, "score": 84.7, "support": 82.4, "interval": { "lower": 78.4, "upper": 91 }, "state": "Measured", "access": "none", "price": 2.0, "endpoints": 1, "releasedAt": "2026-09-02T19:45:59.000Z", "families": { "general": 86.2, "coding": 84, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "Claude Sonnet 5.5", "slug": "anthropic-claude-sonnet-5-5", "rank": 1, "score": 91.9, "support": 20, "interval": { "lower": 72.4, "upper": 100 }, "state": "Measured", "access": "none", "price": 4.0, "endpoints": 8, "releasedAt": "2026-09-28T18:04:46.000Z", "families": { "general": 91.9, "coding": null, "agent": null, "reasoning": null }, "boards": [ "Artificial Analysis Intelligence Index" ] } ], "support25": [ { "name": "gpt-oss-20b", "slug": "openai-gpt-oss-20b", "rank": 164, "score": 37.9, "support": 80.3, "interval": { "lower": 31.4, "upper": 44.4 }, "state": "Measured", "access": "open", "price": 0.036, "endpoints": 12, "releasedAt": "2025-08-05T17:17:09.000Z", "families": { "general": 38.2, "coding": 35.2, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, { "name": "gpt-oss-120b", "slug": "openai-gpt-oss-120b", "rank": 131, "score": 44.7, "support": 83.2, "interval": { "lower": 38.6, "upper": 50.8 }, "state": "Measured", "access": "open", "price": 0.07024999999999999, "endpoints": 22, "releasedAt": "2025-08-05T17:17:11.000Z", "families": { "general": 45.3, "coding": 40.5, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, { "name": "Qwen3 30B A3B Instruct 2507", "slug": "qwen-qwen3-30b-a3b-instruct-2507", "rank": 119, "score": 50.6, "support": 63, "interval": { "lower": 45.6, "upper": 55.6 }, "state": "Measured", "access": "open", "price": 0.084375, "endpoints": 5, "releasedAt": "2025-07-29T16:36:05.000Z", "families": { "general": 50.6, "coding": 50.9, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Arena Coding" ] }, { "name": "DeepSeek V4.1 Flash", "slug": "deepseek-deepseek-v4-1-flash", "rank": 12, "score": 79.9, "support": 78.2, "interval": { "lower": 73.2, "upper": 86.6 }, "state": "Measured", "access": "open", "price": 0.11385, "endpoints": 32, "releasedAt": "2026-09-10T06:21:25.000Z", "families": { "general": 80.8, "coding": 82.6, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "MiMo-V2.6-Pro", "slug": "xiaomi-mimo-v2-6-pro", "rank": 8, "score": 83.2, "support": 70.8, "interval": { "lower": 75.7, "upper": 90.7 }, "state": "Measured", "access": "open", "price": 0.54375, "endpoints": 3, "releasedAt": "2026-09-21T20:07:39.000Z", "families": { "general": 83.1, "coding": 83.9, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, { "name": "Muse Spark 1.3", "slug": "meta-muse-spark-1-3", "rank": 5, "score": 84.7, "support": 82.4, "interval": { "lower": 78.4, "upper": 91 }, "state": "Measured", "access": "none", "price": 2.0, "endpoints": 1, "releasedAt": "2026-09-02T19:45:59.000Z", "families": { "general": 86.2, "coding": 84, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "Claude Opus 5.5", "slug": "anthropic-claude-opus-5-5", "rank": 2, "score": 88.4, "support": 65.4, "interval": { "lower": 79.9, "upper": 96.9 }, "state": "Measured", "access": "none", "price": 8.0, "endpoints": 11, "releasedAt": "2026-09-22T16:32:12.000Z", "families": { "general": 89.9, "coding": 83.9, "agent": 78.2, "reasoning": 84 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ] } ], "measured60": [ { "name": "gpt-oss-20b", "slug": "openai-gpt-oss-20b", "rank": 164, "score": 37.9, "support": 80.3, "interval": { "lower": 31.4, "upper": 44.4 }, "state": "Measured", "access": "open", "price": 0.036, "endpoints": 12, "releasedAt": "2025-08-05T17:17:09.000Z", "families": { "general": 38.2, "coding": 35.2, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, { "name": "gpt-oss-120b", "slug": "openai-gpt-oss-120b", "rank": 131, "score": 44.7, "support": 83.2, "interval": { "lower": 38.6, "upper": 50.8 }, "state": "Measured", "access": "open", "price": 0.07024999999999999, "endpoints": 22, "releasedAt": "2025-08-05T17:17:11.000Z", "families": { "general": 45.3, "coding": 40.5, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, { "name": "Qwen3 30B A3B Instruct 2507", "slug": "qwen-qwen3-30b-a3b-instruct-2507", "rank": 119, "score": 50.6, "support": 63, "interval": { "lower": 45.6, "upper": 55.6 }, "state": "Measured", "access": "open", "price": 0.084375, "endpoints": 5, "releasedAt": "2025-07-29T16:36:05.000Z", "families": { "general": 50.6, "coding": 50.9, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Arena Coding" ] }, { "name": "DeepSeek V4.1 Flash", "slug": "deepseek-deepseek-v4-1-flash", "rank": 12, "score": 79.9, "support": 78.2, "interval": { "lower": 73.2, "upper": 86.6 }, "state": "Measured", "access": "open", "price": 0.11385, "endpoints": 32, "releasedAt": "2026-09-10T06:21:25.000Z", "families": { "general": 80.8, "coding": 82.6, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "MiMo-V2.6-Pro", "slug": "xiaomi-mimo-v2-6-pro", "rank": 8, "score": 83.2, "support": 70.8, "interval": { "lower": 75.7, "upper": 90.7 }, "state": "Measured", "access": "open", "price": 0.54375, "endpoints": 3, "releasedAt": "2026-09-21T20:07:39.000Z", "families": { "general": 83.1, "coding": 83.9, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, { "name": "Muse Spark 1.3", "slug": "meta-muse-spark-1-3", "rank": 5, "score": 84.7, "support": 82.4, "interval": { "lower": 78.4, "upper": 91 }, "state": "Measured", "access": "none", "price": 2.0, "endpoints": 1, "releasedAt": "2026-09-02T19:45:59.000Z", "families": { "general": 86.2, "coding": 84, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "Claude Opus 5.5", "slug": "anthropic-claude-opus-5-5", "rank": 2, "score": 88.4, "support": 65.4, "interval": { "lower": 79.9, "upper": 96.9 }, "state": "Measured", "access": "none", "price": 8.0, "endpoints": 11, "releasedAt": "2026-09-22T16:32:12.000Z", "families": { "general": 89.9, "coding": 83.9, "agent": 78.2, "reasoning": 84 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ] } ] }, "states": { "Measured": { "count": 124, "supportMedian": 62.75, "intervalWidthMedian": 11.400000000000002, "top25": 20 }, "Measured · blended": { "count": 7, "supportMedian": 2.1, "intervalWidthMedian": 45.2, "top25": 0 }, "Modeled · lineage": { "count": 90, "supportMedian": 17.049999999999997, "intervalWidthMedian": 33.8, "top25": 0 }, "Modeled · peer benchmarks": { "count": 54, "supportMedian": 29.0, "intervalWidthMedian": 22.049999999999997, "top25": 5 }, "Modeled · prior": { "count": 2, "supportMedian": 25.9, "intervalWidthMedian": 43.75, "top25": 0 }, "Modeled · succession": { "count": 2, "supportMedian": 24.049999999999997, "intervalWidthMedian": 21.5, "top25": 0 } }, "access": { "gated": { "count": 20, "measured": 14, "supportMedian": 58.85, "priced": 10, "priceMedian": 0.135, "served": 10, "servedMeasured": 10 }, "none": { "count": 39, "measured": 38, "supportMedian": 63.2, "priced": 39, "priceMedian": 1.9, "served": 39, "servedMeasured": 38 }, "open": { "count": 153, "measured": 59, "supportMedian": 28.6, "priced": 53, "priceMedian": 0.31, "served": 53, "servedMeasured": 46 }, "restricted": { "count": 64, "measured": 20, "supportMedian": 31.0, "priced": 13, "priceMedian": 0.37, "served": 13, "servedMeasured": 11 }, "unknown": { "count": 3, "measured": 0, "supportMedian": 15.2, "priced": 1, "priceMedian": 0.42500000000000004, "served": 1, "servedMeasured": 0 } }, "establishedTop": [ { "name": "Claude Opus 5.5", "slug": "anthropic-claude-opus-5-5", "rank": 2, "score": 88.4, "support": 65.4, "interval": { "lower": 79.9, "upper": 96.9 }, "state": "Measured", "access": "none", "price": 8.0, "endpoints": 11, "releasedAt": "2026-09-22T16:32:12.000Z", "families": { "general": 89.9, "coding": 83.9, "agent": 78.2, "reasoning": 84 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ] }, { "name": "Claude Fable 5.1", "slug": "anthropic-claude-fable-5-1", "rank": 4, "score": 86.9, "support": 83.2, "interval": { "lower": 80.7, "upper": 93.1 }, "state": "Measured", "access": "none", "price": 20.0, "endpoints": 4, "releasedAt": "2026-09-01T18:03:58.000Z", "families": { "general": 88.3, "coding": 81.8, "agent": 79.8, "reasoning": 80.8 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ] }, { "name": "Muse Spark 1.3", "slug": "meta-muse-spark-1-3", "rank": 5, "score": 84.7, "support": 82.4, "interval": { "lower": 78.4, "upper": 91 }, "state": "Measured", "access": "none", "price": 2.0, "endpoints": 1, "releasedAt": "2026-09-02T19:45:59.000Z", "families": { "general": 86.2, "coding": 84, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "GPT-6 Astra", "slug": "openai-gpt-6-astra", "rank": 6, "score": 83.6, "support": 81.1, "interval": { "lower": 77.1, "upper": 90.1 }, "state": "Measured", "access": "none", "price": 20.0, "endpoints": 7, "releasedAt": "2026-09-04T20:13:58.000Z", "families": { "general": 84, "coding": 84.2, "agent": 76.9, "reasoning": 82.2 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified", "ARC-AGI-3 Verified" ] }, { "name": "MiMo-V2.6-Pro", "slug": "xiaomi-mimo-v2-6-pro", "rank": 8, "score": 83.2, "support": 70.8, "interval": { "lower": 75.7, "upper": 90.7 }, "state": "Measured", "access": "open", "price": 0.54375, "endpoints": 3, "releasedAt": "2026-09-21T20:07:39.000Z", "families": { "general": 83.1, "coding": 83.9, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, { "name": "Gemini 3.8 Flash", "slug": "google-gemini-3-8-flash", "rank": 9, "score": 82.7, "support": 88.6, "interval": { "lower": 77, "upper": 88.4 }, "state": "Measured", "access": "none", "price": 1.5, "endpoints": 6, "releasedAt": "2026-09-02T15:14:16.000Z", "families": { "general": 84.6, "coding": 82.9, "agent": 58.6, "reasoning": 68 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified", "ARC-AGI-3 Verified" ] }, { "name": "Kimi K3", "slug": "moonshotai-kimi-k3", "rank": 10, "score": 82, "support": 89, "interval": { "lower": 76.3, "upper": 87.7 }, "state": "Measured", "access": "restricted", "price": 6.0, "endpoints": 22, "releasedAt": "2026-07-16T15:30:58.000Z", "families": { "general": 83.6, "coding": 85, "agent": 62.2, "reasoning": 54.6 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "WildClawBench OpenClaw", "ARC-AGI-2 Verified" ] }, { "name": "GLM 5.3", "slug": "z-ai-glm-5-3", "rank": 11, "score": 81, "support": 85.4, "interval": { "lower": 75, "upper": 87 }, "state": "Measured", "access": "restricted", "price": 2.15, "endpoints": 36, "releasedAt": "2026-08-18T20:57:35.000Z", "families": { "general": 82.5, "coding": 80.2, "agent": 56.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "DeepSeek V4.1 Flash", "slug": "deepseek-deepseek-v4-1-flash", "rank": 12, "score": 79.9, "support": 78.2, "interval": { "lower": 73.2, "upper": 86.6 }, "state": "Measured", "access": "open", "price": 0.11385, "endpoints": 32, "releasedAt": "2026-09-10T06:21:25.000Z", "families": { "general": 80.8, "coding": 82.6, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "GPT-5.5", "slug": "openai-gpt-5-5", "rank": 13, "score": 79.8, "support": 90.3, "interval": { "lower": 74.2, "upper": 85.4 }, "state": "Measured", "access": "none", "price": 11.25, "endpoints": 7, "releasedAt": "2026-04-24T17:31:33.000Z", "families": { "general": 81.6, "coding": 79.4, "agent": 62.5, "reasoning": 62 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "WildClawBench OpenClaw", "ARC-AGI-2 Verified", "ARC-AGI-3 Verified" ] }, { "name": "Grok 4.20 Multi-Agent", "slug": "x-ai-grok-4-20-multi-agent", "rank": 14, "score": 79.8, "support": 62.3, "interval": { "lower": 74.8, "upper": 84.8 }, "state": "Measured", "access": "none", "price": 1.5625, "endpoints": 4, "releasedAt": "2026-03-31T17:45:58.000Z", "families": { "general": 80.3, "coding": 76, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Arena Coding" ] }, { "name": "Gemini 3.1 Pro Preview", "slug": "google-gemini-3-1-pro-preview", "rank": 15, "score": 79.4, "support": 86.1, "interval": { "lower": 73.5, "upper": 85.3 }, "state": "Measured", "access": "none", "price": 4.5, "endpoints": 6, "releasedAt": "2026-02-19T14:00:27.000Z", "families": { "general": 80.7, "coding": 80.4, "agent": 31.1, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] } ], "establishedAccessLeaders": { "none": { "name": "Claude Opus 5.5", "slug": "anthropic-claude-opus-5-5", "rank": 2, "score": 88.4, "support": 65.4, "interval": { "lower": 79.9, "upper": 96.9 }, "state": "Measured", "access": "none", "price": 8.0, "endpoints": 11, "releasedAt": "2026-09-22T16:32:12.000Z", "families": { "general": 89.9, "coding": 83.9, "agent": 78.2, "reasoning": 84 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ] }, "open": { "name": "MiMo-V2.6-Pro", "slug": "xiaomi-mimo-v2-6-pro", "rank": 8, "score": 83.2, "support": 70.8, "interval": { "lower": 75.7, "upper": 90.7 }, "state": "Measured", "access": "open", "price": 0.54375, "endpoints": 3, "releasedAt": "2026-09-21T20:07:39.000Z", "families": { "general": 83.1, "coding": 83.9, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, "restricted": { "name": "Kimi K3", "slug": "moonshotai-kimi-k3", "rank": 10, "score": 82, "support": 89, "interval": { "lower": 76.3, "upper": 87.7 }, "state": "Measured", "access": "restricted", "price": 6.0, "endpoints": 22, "releasedAt": "2026-07-16T15:30:58.000Z", "families": { "general": 83.6, "coding": 85, "agent": 62.2, "reasoning": 54.6 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "WildClawBench OpenClaw", "ARC-AGI-2 Verified" ] }, "gated": { "name": "Gemma 3 27B", "slug": "google-gemma-3-27b-it", "rank": 132, "score": 44.5, "support": 62.7, "interval": { "lower": 39.5, "upper": 49.5 }, "state": "Measured", "access": "gated", "price": 0.1725, "endpoints": 4, "releasedAt": "2025-03-12T05:12:39.000Z", "families": { "general": 46.2, "coding": 32, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Arena Coding" ] } }, "codingCount": 105, "codingTop": [ { "name": "Kimi K3", "slug": "moonshotai-kimi-k3", "rank": 10, "score": 82, "support": 89, "interval": { "lower": 76.3, "upper": 87.7 }, "state": "Measured", "access": "restricted", "price": 6.0, "endpoints": 22, "releasedAt": "2026-07-16T15:30:58.000Z", "families": { "general": 83.6, "coding": 85, "agent": 62.2, "reasoning": 54.6 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "WildClawBench OpenClaw", "ARC-AGI-2 Verified" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Max", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 85, "confidence": 88.4, "share": 0.3 } }, { "name": "GPT-6 Astra", "slug": "openai-gpt-6-astra", "rank": 6, "score": 83.6, "support": 81.1, "interval": { "lower": 77.1, "upper": 90.1 }, "state": "Measured", "access": "none", "price": 20.0, "endpoints": 7, "releasedAt": "2026-09-04T20:13:58.000Z", "families": { "general": 84, "coding": 84.2, "agent": 76.9, "reasoning": 82.2 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified", "ARC-AGI-3 Verified" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Max", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 84.2, "confidence": 58.1, "share": 0.3 } }, { "name": "Muse Spark 1.3", "slug": "meta-muse-spark-1-3", "rank": 5, "score": 84.7, "support": 82.4, "interval": { "lower": 78.4, "upper": 91 }, "state": "Measured", "access": "none", "price": 2.0, "endpoints": 1, "releasedAt": "2026-09-02T19:45:59.000Z", "families": { "general": 86.2, "coding": 84, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Max", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 84, "confidence": 70.1, "share": 0.3 } }, { "name": "Claude Opus 5.5", "slug": "anthropic-claude-opus-5-5", "rank": 2, "score": 88.4, "support": 65.4, "interval": { "lower": 79.9, "upper": 96.9 }, "state": "Measured", "access": "none", "price": 8.0, "endpoints": 11, "releasedAt": "2026-09-22T16:32:12.000Z", "families": { "general": 89.9, "coding": 83.9, "agent": 78.2, "reasoning": 84 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "High", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 83.9, "confidence": 39.7, "share": 0.3 } }, { "name": "MiMo-V2.6-Pro", "slug": "xiaomi-mimo-v2-6-pro", "rank": 8, "score": 83.2, "support": 70.8, "interval": { "lower": 75.7, "upper": 90.7 }, "state": "Measured", "access": "open", "price": 0.54375, "endpoints": 3, "releasedAt": "2026-09-21T20:07:39.000Z", "families": { "general": 83.1, "coding": 83.9, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Default", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 83.9, "confidence": 48.1, "share": 0.3 } }, { "name": "Gemini 3.8 Flash", "slug": "google-gemini-3-8-flash", "rank": 9, "score": 82.7, "support": 88.6, "interval": { "lower": 77, "upper": 88.4 }, "state": "Measured", "access": "none", "price": 1.5, "endpoints": 6, "releasedAt": "2026-09-02T15:14:16.000Z", "families": { "general": 84.6, "coding": 82.9, "agent": 58.6, "reasoning": 68 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified", "ARC-AGI-3 Verified" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "High", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 82.9, "confidence": 85.8, "share": 0.3 } }, { "name": "DeepSeek V4.1 Flash", "slug": "deepseek-deepseek-v4-1-flash", "rank": 12, "score": 79.9, "support": 78.2, "interval": { "lower": 73.2, "upper": 86.6 }, "state": "Measured", "access": "open", "price": 0.11385, "endpoints": 32, "releasedAt": "2026-09-10T06:21:25.000Z", "families": { "general": 80.8, "coding": 82.6, "agent": 58.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Max", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 82.6, "confidence": 60.9, "share": 0.3 } }, { "name": "Claude Fable 5.1", "slug": "anthropic-claude-fable-5-1", "rank": 4, "score": 86.9, "support": 83.2, "interval": { "lower": 80.7, "upper": 93.1 }, "state": "Measured", "access": "none", "price": 20.0, "endpoints": 4, "releasedAt": "2026-09-01T18:03:58.000Z", "families": { "general": 88.3, "coding": 81.8, "agent": 79.8, "reasoning": 80.8 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Max", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 81.8, "confidence": 61.9, "share": 0.3 } }, { "name": "GLM 5.3 Flash", "slug": "z-ai-glm-5-3-flash", "rank": 17, "score": 78.9, "support": 87.6, "interval": { "lower": 73.1, "upper": 84.7 }, "state": "Measured", "access": "open", "price": 0.2375, "endpoints": 32, "releasedAt": "2026-08-26T13:59:01.000Z", "families": { "general": 80.8, "coding": 80.8, "agent": 44.7, "reasoning": 60.8 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Default", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 80.8, "confidence": 83.6, "share": 0.3 } }, { "name": "Gemini 3.1 Pro Preview", "slug": "google-gemini-3-1-pro-preview", "rank": 15, "score": 79.4, "support": 86.1, "interval": { "lower": 73.5, "upper": 85.3 }, "state": "Measured", "access": "none", "price": 4.5, "endpoints": 6, "releasedAt": "2026-02-19T14:00:27.000Z", "families": { "general": 80.7, "coding": 80.4, "agent": 31.1, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Default", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 80.4, "confidence": 100, "share": 0.3 } }, { "name": "GLM 5.3", "slug": "z-ai-glm-5-3", "rank": 11, "score": 81, "support": 85.4, "interval": { "lower": 75, "upper": 87 }, "state": "Measured", "access": "restricted", "price": 2.15, "endpoints": 36, "releasedAt": "2026-08-18T20:57:35.000Z", "families": { "general": 82.5, "coding": 80.2, "agent": 56.7, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Max", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 80.2, "confidence": 80.6, "share": 0.3 } }, { "name": "MiMo-V2.6-Flash", "slug": "xiaomi-mimo-v2-6-flash", "rank": 25, "score": 75.3, "support": 70.9, "interval": { "lower": 67.8, "upper": 82.8 }, "state": "Measured", "access": "open", "price": 0.17500000000000002, "endpoints": 5, "releasedAt": "2026-09-21T20:07:44.000Z", "families": { "general": 74.7, "coding": 79.8, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ], "coding": { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Default", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 79.8, "confidence": 47.5, "share": 0.3 } } ], "noCodingTop20": [ "Claude Sonnet 5.5", "GPT-6.1 Sol", "Qwen3.8 Max (0902)", "LongCat 2.0", "Qwen3.8 Flash", "Nex N2.5 Max", "Nex-N2.5-Pro" ], "pairs": [ { "name": "Claude Opus 5.5", "rank": 2, "arena": 88.3, "aa": 93.3, "difference": 5.0, "arenaConfiguration": "High", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "Claude Fable 5.1", "rank": 4, "arena": 87.8, "aa": 89.8, "difference": 2.0, "arenaConfiguration": "Max", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "Muse Spark 1.3", "rank": 5, "arena": 86.4, "aa": 85.5, "difference": -0.9, "arenaConfiguration": "Max", "aaConfiguration": "intelligence-index:v4.3:max" }, { "name": "GPT-6 Astra", "rank": 6, "arena": 82, "aa": 89.2, "difference": 7.2, "arenaConfiguration": "Max", "aaConfiguration": "intelligence-index:v4.3:max" }, { "name": "MiMo-V2.6-Pro", "rank": 8, "arena": 82.8, "aa": 84, "difference": 1.2, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "Gemini 3.8 Flash", "rank": 9, "arena": 86.3, "aa": 79.6, "difference": -6.7, "arenaConfiguration": "High", "aaConfiguration": "intelligence-index:v4.3:high" }, { "name": "Kimi K3", "rank": 10, "arena": 84.2, "aa": 81.8, "difference": -2.4, "arenaConfiguration": "Max", "aaConfiguration": "intelligence-index:v4.3:max" }, { "name": "GLM 5.3", "rank": 11, "arena": 82.5, "aa": 82.8, "difference": 0.3, "arenaConfiguration": "Max", "aaConfiguration": "intelligence-index:v4.3:max" }, { "name": "DeepSeek V4.1 Flash", "rank": 12, "arena": 81.7, "aa": 78.4, "difference": -3.3, "arenaConfiguration": "Max", "aaConfiguration": "intelligence-index:v4.3:max" }, { "name": "GPT-5.5", "rank": 13, "arena": 82.9, "aa": 77.5, "difference": -5.4, "arenaConfiguration": "High", "aaConfiguration": "intelligence-index:v4.3:xhigh" }, { "name": "Gemini 3.1 Pro Preview", "rank": 15, "arena": 84.2, "aa": 70.5, "difference": -13.7, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "GLM 5.3 Flash", "rank": 17, "arena": 80.9, "aa": 80.4, "difference": -0.5, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "GPT-5.6 Terra", "rank": 23, "arena": 78.2, "aa": 80.6, "difference": 2.4, "arenaConfiguration": "X-High", "aaConfiguration": "intelligence-index:v4.3:max" }, { "name": "DeepSeek V4 Pro 0813", "rank": 24, "arena": 78.1, "aa": 75.6, "difference": -2.5, "arenaConfiguration": "High", "aaConfiguration": "intelligence-index:v4.3:max" }, { "name": "MiMo-V2.6-Flash", "rank": 25, "arena": 73.8, "aa": 77.1, "difference": 3.3, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "Gemini 3.5 Flash Lite", "rank": 29, "arena": 76, "aa": 64.3, "difference": -11.7, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "Grok 4.7", "rank": 30, "arena": 69.4, "aa": 84.1, "difference": 14.7, "arenaConfiguration": "X-High", "aaConfiguration": "intelligence-index:v4.3:xhigh" }, { "name": "Qwen3.7 Plus", "rank": 31, "arena": 75.7, "aa": 66.8, "difference": -8.9, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "GPT-6 Luna", "rank": 33, "arena": 70.5, "aa": 76.6, "difference": 6.1, "arenaConfiguration": "Max", "aaConfiguration": "intelligence-index:v4.3:max" }, { "name": "GPT-5.4 Mini", "rank": 36, "arena": 72.1, "aa": 65.9, "difference": -6.2, "arenaConfiguration": "High", "aaConfiguration": "intelligence-index:v4.3:xhigh" }, { "name": "Qwen3.8 27B", "rank": 37, "arena": 68.4, "aa": 73.7, "difference": 5.3, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:xhigh" }, { "name": "MiniMax M3", "rank": 40, "arena": 70.1, "aa": 70.1, "difference": 0.0, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "Qwen3.5 397B A17B", "rank": 42, "arena": 70.5, "aa": 61.3, "difference": -9.2, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "Inkling", "rank": 46, "arena": 70.5, "aa": 66.6, "difference": -3.9, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:xhigh" }, { "name": "Mistral Medium 3.5", "rank": 64, "arena": 64.5, "aa": 57.8, "difference": -6.7, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "DeepSeek V3.1 Terminus", "rank": 72, "arena": 61.4, "aa": 58.3, "difference": -3.1, "arenaConfiguration": "Thinking", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "Qwen3.5-122B-A10B", "rank": 74, "arena": 60.7, "aa": 58.9, "difference": -1.8, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "Inkling Small", "rank": 82, "arena": 56.6, "aa": 67.2, "difference": 10.6, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "GPT-5.4 Nano", "rank": 85, "arena": 55.4, "aa": 63.1, "difference": 7.7, "arenaConfiguration": "High", "aaConfiguration": "intelligence-index:v4.3:xhigh" }, { "name": "Trinity Large Thinking", "rank": 123, "arena": 46.7, "aa": 55.1, "difference": 8.4, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "gpt-oss-120b", "rank": 131, "arena": 41.8, "aa": 55.7, "difference": 13.9, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:high" }, { "name": "gpt-oss-20b", "rank": 164, "arena": 32.9, "aa": 53.6, "difference": 20.7, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:high" }, { "name": "granite 4.2 3b", "rank": 171, "arena": 26.3, "aa": 53.6, "difference": 27.3, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" }, { "name": "Granite 4.2 8B", "rank": 172, "arena": 25.8, "aa": 55.3, "difference": 29.5, "arenaConfiguration": "Default", "aaConfiguration": "intelligence-index:v4.3:reasoning-unspecified" } ], "pairSummary": { "n": 34, "medianAbsoluteGap": 5.75, "meanAbsoluteGap": 7.426470588235294, "over10": 8, "within3": 10, "medianSignedGap": 0.15, "aaHigher": 17, "arenaHigher": 16 }, "caps": [ { "name": "Muse Spark 1.3", "family": "agent", "delta": -27.5 }, { "name": "Gemini 3.8 Flash", "family": "agent", "delta": -25.999999999999993 }, { "name": "Kimi K3", "family": "reasoning", "delta": -28.999999999999993 }, { "name": "GLM 5.3", "family": "agent", "delta": -25.799999999999997 }, { "name": "Gemini 3.1 Pro Preview", "family": "agent", "delta": -49.6 }, { "name": "GLM 5.3 Flash", "family": "agent", "delta": -36.099999999999994 }, { "name": "GPT-5.6 Terra", "family": "agent", "delta": -33.4 }, { "name": "Gemma 4 31B", "family": "agent", "delta": -36.8 }, { "name": "Gemini 3.5 Flash Lite", "family": "reasoning", "delta": -41.199999999999996 }, { "name": "Qwen3.7 Plus", "family": "agent", "delta": -43.4 }, { "name": "GPT-6 Luna", "family": "reasoning", "delta": -27.9 }, { "name": "GPT-5.4 Mini", "family": "reasoning", "delta": -34.3 }, { "name": "MiniMax M3", "family": "agent", "delta": -38.599999999999994 }, { "name": "Inkling", "family": "agent", "delta": -48.2 }, { "name": "Inkling", "family": "reasoning", "delta": -27.299999999999997 }, { "name": "o3", "family": "reasoning", "delta": -38.0 }, { "name": "Mistral Medium 3.5", "family": "agent", "delta": -42.099999999999994 }, { "name": "DeepSeek V3.2", "family": "agent", "delta": -31.0 }, { "name": "DeepSeek V3.2", "family": "reasoning", "delta": -44.5 }, { "name": "Claude Haiku 4.5", "family": "reasoning", "delta": -40.7 }, { "name": "Inkling Small", "family": "agent", "delta": -35.900000000000006 }, { "name": "GPT-5.4 Nano", "family": "reasoning", "delta": -33.9 }, { "name": "o4 Mini", "family": "reasoning", "delta": -26.200000000000003 }, { "name": "Solar Pro 4", "family": "agent", "delta": -36.8 } ], "supportCheck": { "n": 230, "maximumAbsoluteError": 0.07319999999999993 }, "boardCoverage": [ { "board": "Arena Overall", "family": "general", "share": 0.75, "currentCount": 105, "newestAdmittedPublication": "2026-09-25T00:00:00.000Z" }, { "board": "Artificial Analysis Intelligence Index", "family": "general", "share": 0.25, "currentCount": 53, "newestAdmittedPublication": "2026-09-30T20:36:10.452Z" }, { "board": "SWE-bench bash-only", "family": "coding", "share": 0.45, "currentCount": 0, "newestAdmittedPublication": null }, { "board": "Arena Coding", "family": "coding", "share": 0.3, "currentCount": 105, "newestAdmittedPublication": "2026-09-25T00:00:00.000Z" }, { "board": "Terminal-Bench 2.1 Terminus 2", "family": "coding", "share": 0.25, "currentCount": 0, "newestAdmittedPublication": null }, { "board": "LMArena Agent", "family": "agent", "share": 0.5, "currentCount": 21, "newestAdmittedPublication": "2026-09-28T00:00:00.000Z" }, { "board": "BFCL V4", "family": "agent", "share": 0.25, "currentCount": 0, "newestAdmittedPublication": null }, { "board": "WildClawBench OpenClaw", "family": "agent", "share": 0.25, "currentCount": 9, "newestAdmittedPublication": "2026-07-20T00:00:00.000Z" }, { "board": "ARC-AGI-2 Verified", "family": "reasoning", "share": 0.7, "currentCount": 25, "newestAdmittedPublication": "2026-09-30T19:42:43.000Z" }, { "board": "ARC-AGI-3 Verified", "family": "reasoning", "share": 0.3, "currentCount": 6, "newestAdmittedPublication": "2026-09-30T19:42:43.000Z" } ], "publisherNames": { "count": 52, "note": "Exact exported publisher display names; no unverified alias merging.", "rows": [ { "name": "Qwen", "count": 37, "top25": 2, "top50": 5 }, { "name": "NVIDIA", "count": 24, "top25": 0, "top50": 0 }, { "name": "Microsoft", "count": 22, "top25": 0, "top50": 0 }, { "name": "Google", "count": 19, "top25": 2, "top50": 5 }, { "name": "OpenAI", "count": 17, "top25": 4, "top50": 6 }, { "name": "Allen Institute", "count": 16, "top25": 0, "top50": 0 }, { "name": "IBM Granite", "count": 11, "top25": 0, "top50": 0 }, { "name": "Meta", "count": 8, "top25": 1, "top50": 2 }, { "name": "Arcee AI", "count": 7, "top25": 0, "top50": 0 }, { "name": "InclusionAI", "count": 7, "top25": 0, "top50": 1 }, { "name": "Tencent", "count": 7, "top25": 0, "top50": 1 }, { "name": "Amazon", "count": 6, "top25": 0, "top50": 0 }, { "name": "DeepSeek", "count": 6, "top25": 2, "top50": 3 }, { "name": "Mistral AI", "count": 6, "top25": 0, "top50": 0 }, { "name": "Z.ai", "count": 6, "top25": 2, "top50": 3 }, { "name": "Aisingapore", "count": 5, "top25": 0, "top50": 0 }, { "name": "Cohere", "count": 5, "top25": 0, "top50": 0 }, { "name": "Tiiuae", "count": 5, "top25": 0, "top50": 0 }, { "name": "Amd", "count": 4, "top25": 0, "top50": 0 }, { "name": "Anthropic", "count": 4, "top25": 3, "top50": 3 }, { "name": "IFM", "count": 4, "top25": 0, "top50": 1 }, { "name": "Poolside", "count": 4, "top25": 0, "top50": 2 }, { "name": "Deep Cogito", "count": 3, "top25": 0, "top50": 0 }, { "name": "Moonshot AI", "count": 3, "top25": 1, "top50": 2 }, { "name": "Nex AGI", "count": 3, "top25": 2, "top50": 3 }, { "name": "Ornith Ai", "count": 3, "top25": 1, "top50": 3 }, { "name": "Upstage", "count": 3, "top25": 0, "top50": 0 }, { "name": "Xiaomi", "count": 3, "top25": 2, "top50": 3 }, { "name": "CohereLabs", "count": 2, "top25": 0, "top50": 0 }, { "name": "Internlm", "count": 2, "top25": 0, "top50": 0 }, { "name": "Meituan Longcat", "count": 2, "top25": 0, "top50": 0 }, { "name": "MiniMax", "count": 2, "top25": 0, "top50": 1 }, { "name": "Thinking Machines", "count": 2, "top25": 0, "top50": 1 }, { "name": "XiaomiMiMo", "count": 2, "top25": 0, "top50": 0 }, { "name": "xAI", "count": 2, "top25": 1, "top50": 2 }, { "name": "Apple", "count": 1, "top25": 0, "top50": 0 }, { "name": "ByteDance Seed", "count": 1, "top25": 1, "top50": 1 }, { "name": "Deepseek Ai", "count": 1, "top25": 0, "top50": 0 }, { "name": "Dots Studio", "count": 1, "top25": 0, "top50": 0 }, { "name": "Inception Labs", "count": 1, "top25": 0, "top50": 0 }, { "name": "Kwaipilot", "count": 1, "top25": 0, "top50": 1 }, { "name": "LGAI EXAONE", "count": 1, "top25": 0, "top50": 0 }, { "name": "LiquidAI", "count": 1, "top25": 0, "top50": 0 }, { "name": "Meituan", "count": 1, "top25": 1, "top50": 1 }, { "name": "MiniMaxAI", "count": 1, "top25": 0, "top50": 0 }, { "name": "Nanbeige", "count": 1, "top25": 0, "top50": 0 }, { "name": "Openbmb", "count": 1, "top25": 0, "top50": 0 }, { "name": "StepFun", "count": 1, "top25": 0, "top50": 0 }, { "name": "Stepfun Ai", "count": 1, "top25": 0, "top50": 0 }, { "name": "Togethercomputer", "count": 1, "top25": 0, "top50": 0 }, { "name": "XingChen AGI", "count": 1, "top25": 0, "top50": 0 }, { "name": "Zai Org", "count": 1, "top25": 0, "top50": 0 } ] }, "canonicalPublishers": { "count": 45, "note": "Recorded publisher keys joined by exact reviewed site/Hub authority aliases, not fuzzy names or legal entity ownership.", "rows": [ { "publisherId": "qwen", "count": 37, "top25": 2, "top50": 5 }, { "publisherId": "nvidia", "count": 24, "top25": 0, "top50": 0 }, { "publisherId": "microsoft", "count": 22, "top25": 0, "top50": 0 }, { "publisherId": "google", "count": 19, "top25": 2, "top50": 5 }, { "publisherId": "openai", "count": 17, "top25": 4, "top50": 6 }, { "publisherId": "allenai", "count": 16, "top25": 0, "top50": 0 }, { "publisherId": "ibm-granite", "count": 11, "top25": 0, "top50": 0 }, { "publisherId": "meta-llama", "count": 8, "top25": 1, "top50": 2 }, { "publisherId": "arcee-ai", "count": 7, "top25": 0, "top50": 0 }, { "publisherId": "cohere", "count": 7, "top25": 0, "top50": 0 }, { "publisherId": "deepseek", "count": 7, "top25": 2, "top50": 3 }, { "publisherId": "inclusionai", "count": 7, "top25": 0, "top50": 1 }, { "publisherId": "tencent", "count": 7, "top25": 0, "top50": 1 }, { "publisherId": "z-ai", "count": 7, "top25": 2, "top50": 3 }, { "publisherId": "amazon", "count": 6, "top25": 0, "top50": 0 }, { "publisherId": "mistralai", "count": 6, "top25": 0, "top50": 0 }, { "publisherId": "aisingapore", "count": 5, "top25": 0, "top50": 0 }, { "publisherId": "tiiuae", "count": 5, "top25": 0, "top50": 0 }, { "publisherId": "xiaomi", "count": 5, "top25": 2, "top50": 3 }, { "publisherId": "IFM", "count": 4, "top25": 0, "top50": 1 }, { "publisherId": "amd", "count": 4, "top25": 0, "top50": 0 }, { "publisherId": "anthropic", "count": 4, "top25": 3, "top50": 3 }, { "publisherId": "poolside", "count": 4, "top25": 0, "top50": 2 }, { "publisherId": "deepcogito", "count": 3, "top25": 0, "top50": 0 }, { "publisherId": "meituan", "count": 3, "top25": 1, "top50": 1 }, { "publisherId": "minimax", "count": 3, "top25": 0, "top50": 1 }, { "publisherId": "moonshotai", "count": 3, "top25": 1, "top50": 2 }, { "publisherId": "nex-agi", "count": 3, "top25": 2, "top50": 3 }, { "publisherId": "ornith-ai", "count": 3, "top25": 1, "top50": 3 }, { "publisherId": "upstage", "count": 3, "top25": 0, "top50": 0 }, { "publisherId": "internlm", "count": 2, "top25": 0, "top50": 0 }, { "publisherId": "stepfun", "count": 2, "top25": 0, "top50": 0 }, { "publisherId": "thinkingmachines", "count": 2, "top25": 0, "top50": 1 }, { "publisherId": "x-ai", "count": 2, "top25": 1, "top50": 2 }, { "publisherId": "Nanbeige", "count": 1, "top25": 0, "top50": 0 }, { "publisherId": "XingChen-AGI", "count": 1, "top25": 0, "top50": 0 }, { "publisherId": "apple", "count": 1, "top25": 0, "top50": 0 }, { "publisherId": "bytedance-seed", "count": 1, "top25": 1, "top50": 1 }, { "publisherId": "dots-studio", "count": 1, "top25": 0, "top50": 0 }, { "publisherId": "inception", "count": 1, "top25": 0, "top50": 0 }, { "publisherId": "kwaipilot", "count": 1, "top25": 0, "top50": 1 }, { "publisherId": "lgai-exaone", "count": 1, "top25": 0, "top50": 0 }, { "publisherId": "liquid", "count": 1, "top25": 0, "top50": 0 }, { "publisherId": "openbmb", "count": 1, "top25": 0, "top50": 0 }, { "publisherId": "togethercomputer", "count": 1, "top25": 0, "top50": 0 } ] }, "septemberCount": 33, "context": { "count": 121, "median": 262144, "atLeastMillion": 36, "maximum": 2000000 }, "launches": [ { "name": "Claude Opus 5.5", "slug": "anthropic-claude-opus-5-5", "rank": 2, "score": 88.4, "support": 65.4, "interval": { "lower": 79.9, "upper": 96.9 }, "state": "Measured", "access": "none", "price": 8.0, "endpoints": 11, "releasedAt": "2026-09-22T16:32:12.000Z", "families": { "general": 89.9, "coding": 83.9, "agent": 78.2, "reasoning": 84 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified" ] }, { "name": "MiMo-V2.6-Pro", "slug": "xiaomi-mimo-v2-6-pro", "rank": 8, "score": 83.2, "support": 70.8, "interval": { "lower": 75.7, "upper": 90.7 }, "state": "Measured", "access": "open", "price": 0.54375, "endpoints": 3, "releasedAt": "2026-09-21T20:07:39.000Z", "families": { "general": 83.1, "coding": 83.9, "agent": null, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding" ] }, { "name": "Grok 4.7", "slug": "x-ai-grok-4-7", "rank": 30, "score": 72.2, "support": 73.2, "interval": { "lower": 64.9, "upper": 79.5 }, "state": "Measured", "access": "none", "price": 3.0, "endpoints": 5, "releasedAt": "2026-09-21T16:19:01.000Z", "families": { "general": 73.7, "coding": 68, "agent": 58.9, "reasoning": null }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent" ] }, { "name": "GPT-6 Sol", "slug": "openai-gpt-6-sol", "rank": null, "score": 77.7, "support": 77.2, "interval": { "lower": 70.7, "upper": 84.7 }, "state": "Measured", "access": "none", "price": 4.0, "endpoints": 7, "releasedAt": "2026-09-22T18:12:55.000Z", "families": { "general": 78.1, "coding": 79.5, "agent": 74.2, "reasoning": 67.6 }, "boards": [ "Arena Overall", "Artificial Analysis Intelligence Index", "Arena Coding", "LMArena Agent", "ARC-AGI-2 Verified", "ARC-AGI-3 Verified" ] } ], "estimatesTop25": [ { "name": "LongCat 2.0", "slug": "meituan-longcat-2-0", "rank": 16, "score": 79.1, "support": 32, "interval": { "lower": 71.4, "upper": 86.9 }, "state": "Modeled · peer benchmarks", "access": "open", "price": 0.5249999999999999, "endpoints": 1, "releasedAt": "2026-07-20T13:37:38.000Z", "families": null, "boards": [], "fusion": { "mean": 79.1, "sigma": 6.1, "terms": [ { "channel": "corpus-prior", "mean": 56.3, "sigma": 23, "share": 0.07, "disclosure": "global-corpus-prior over 230 held-out anchors" }, { "channel": "within-card", "mean": 80.8, "sigma": 6.3, "share": 0.93, "disclosure": "launch card against 4 resolved peers on 6 rows" } ] } }, { "name": "Nex N2.5 Max", "slug": "nex-agi-nex-n2-5-max", "rank": 19, "score": 78.4, "support": 27, "interval": { "lower": 70.9, "upper": 85.8 }, "state": "Modeled · peer benchmarks", "access": "open", "price": null, "endpoints": 0, "releasedAt": "2026-09-07T03:03:49.000Z", "families": null, "boards": [], "fusion": { "mean": 78.4, "sigma": 5.8, "terms": [ { "channel": "corpus-prior", "mean": 55.3, "sigma": 23, "share": 0.064, "disclosure": "global-corpus-prior over 230 held-out anchors" }, { "channel": "within-card", "mean": 79.9, "sigma": 6, "share": 0.936, "disclosure": "launch card against 6 resolved peers on 5 rows" } ] } }, { "name": "Nex-N2.5-Pro", "slug": "nex-agi-nex-n2-5-pro", "rank": 20, "score": 77.8, "support": 38, "interval": { "lower": 69.5, "upper": 86.1 }, "state": "Modeled · peer benchmarks", "access": "open", "price": 0.11875, "endpoints": 1, "releasedAt": "2026-09-08T17:54:10.000Z", "families": null, "boards": [], "fusion": { "mean": 77.8, "sigma": 6.5, "terms": [ { "channel": "corpus-prior", "mean": 56.3, "sigma": 23, "share": 0.079, "disclosure": "global-corpus-prior over 230 held-out anchors" }, { "channel": "within-card", "mean": 79.7, "sigma": 6.8, "share": 0.921, "disclosure": "launch card against 8 resolved peers on 8 rows" } ] } }, { "name": "Ornith 1.5 397B", "slug": "ornith-ai-ornith-1-5-397b", "rank": 21, "score": 77.4, "support": 40, "interval": { "lower": 68.1, "upper": 86.7 }, "state": "Modeled · peer benchmarks", "access": "open", "price": null, "endpoints": 0, "releasedAt": "2026-08-18T06:34:59.000Z", "families": null, "boards": [], "fusion": { "mean": 77.4, "sigma": 7.2, "terms": [ { "channel": "corpus-prior", "mean": 56.3, "sigma": 23, "share": 0.099, "disclosure": "global-corpus-prior over 230 held-out anchors" }, { "channel": "within-card", "mean": 79.7, "sigma": 7.6, "share": 0.901, "disclosure": "launch card against 2 resolved peers on 11 rows" } ] } }, { "name": "Seed 2.1 Turbo", "slug": "bytedance-seed-seed-2-1-turbo", "rank": 22, "score": 77, "support": 23, "interval": { "lower": 66.9, "upper": 87.1 }, "state": "Modeled · peer benchmarks", "access": "none", "price": 1.0, "endpoints": 1, "releasedAt": "2026-08-12T16:29:36.000Z", "families": null, "boards": [], "fusion": { "mean": 77, "sigma": 7.9, "terms": [ { "channel": "corpus-prior", "mean": 56.3, "sigma": 23, "share": 0.118, "disclosure": "global-corpus-prior over 230 held-out anchors" }, { "channel": "within-card", "mean": 79.8, "sigma": 8.4, "share": 0.882, "disclosure": "launch card against 2 resolved peers on 3 rows" } ] } } ], "workedExamples": { "Claude Fable 5.1": { "modelId": "anthropic/claude-fable-5.1", "slug": "anthropic-claude-fable-5-1", "name": "Claude Fable 5.1", "publisher": "Anthropic", "modelUrl": "https://modelcap.ai/model/anthropic-claude-fable-5-1", "identityKey": "anthropic::anthropic/claude-fable-5.1::default::catalogue-product", "rank": { "current": 4, "allVersions": 4 }, "score": 86.9, "basis": "measured", "confidence": 83.2, "interval": { "lower": 80.7, "upper": 93.1 }, "families": { "general": 88.3, "coding": 81.8, "agent": 79.8, "reasoning": 80.8 }, "observations": [ { "family": "general", "board": "Arena Overall", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Max", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 87.8, "confidence": 94.3, "share": 0.75 }, { "family": "general", "board": "Artificial Analysis Intelligence Index", "source": "Artificial Analysis", "sourceUrl": "https://artificialanalysis.ai/models/claude-fable-5-1", "configuration": "intelligence-index:v4.3:reasoning-unspecified", "publishedAt": "2026-09-19T22:46:47.972Z", "sourceScore": 89.8, "confidence": 100, "share": 0.25 }, { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Max", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 81.8, "confidence": 61.9, "share": 0.3 }, { "family": "agent", "board": "LMArena Agent", "source": "LMArena Agent", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Max", "publishedAt": "2026-09-28T00:00:00.000Z", "sourceScore": 79.8, "confidence": 92.5, "share": 0.5 }, { "family": "reasoning", "board": "ARC-AGI-2 Verified", "source": "ARC Prize", "sourceUrl": "https://arcprize.org/results/anthropic-claude-fable-5-1", "configuration": "Max", "publishedAt": "2026-09-30T19:42:43.000Z", "sourceScore": 80.8, "confidence": 100, "share": 0.7 } ], "fusion": null }, "Gemini 3.1 Pro Preview": { "modelId": "google/gemini-3.1-pro-preview", "slug": "google-gemini-3-1-pro-preview", "name": "Gemini 3.1 Pro Preview", "publisher": "Google", "modelUrl": "https://modelcap.ai/model/google-gemini-3-1-pro-preview", "identityKey": "google::google/gemini-3.1-pro-preview::default::catalogue-product", "rank": { "current": 15, "allVersions": 29 }, "score": 79.4, "basis": "measured", "confidence": 86.1, "interval": { "lower": 73.5, "upper": 85.3 }, "families": { "general": 80.7, "coding": 80.4, "agent": 31.1, "reasoning": null }, "observations": [ { "family": "general", "board": "Arena Overall", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Default", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 84.2, "confidence": 100, "share": 0.75 }, { "family": "general", "board": "Artificial Analysis Intelligence Index", "source": "Artificial Analysis", "sourceUrl": "https://artificialanalysis.ai/models/gemini-3-1-pro-preview", "configuration": "intelligence-index:v4.3:reasoning-unspecified", "publishedAt": "2026-09-19T22:46:47.972Z", "sourceScore": 70.5, "confidence": 100, "share": 0.25 }, { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Default", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 80.4, "confidence": 100, "share": 0.3 }, { "family": "agent", "board": "LMArena Agent", "source": "LMArena Agent", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Default", "publishedAt": "2026-09-28T00:00:00.000Z", "sourceScore": 31.1, "confidence": 100, "share": 0.5 } ], "fusion": null }, "MiMo-V2.6-Flash": { "modelId": "xiaomi/mimo-v2.6-flash", "slug": "xiaomi-mimo-v2-6-flash", "name": "MiMo-V2.6-Flash", "publisher": "Xiaomi", "modelUrl": "https://modelcap.ai/model/xiaomi-mimo-v2-6-flash", "identityKey": "xiaomi::xiaomi/mimo-v2.6-flash::fp8::catalogue-product", "rank": { "current": 25, "allVersions": 51 }, "score": 75.3, "basis": "measured", "confidence": 70.9, "interval": { "lower": 67.8, "upper": 82.8 }, "families": { "general": 74.7, "coding": 79.8, "agent": null, "reasoning": null }, "observations": [ { "family": "general", "board": "Arena Overall", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Default", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 73.8, "confidence": 82, "share": 0.75 }, { "family": "general", "board": "Artificial Analysis Intelligence Index", "source": "Artificial Analysis", "sourceUrl": "https://artificialanalysis.ai/models/mimo-v2-6-flash", "configuration": "intelligence-index:v4.3:reasoning-unspecified", "publishedAt": "2026-09-27T16:12:27.866Z", "sourceScore": 77.1, "confidence": 100, "share": 0.25 }, { "family": "coding", "board": "Arena Coding", "source": "Arena", "sourceUrl": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset", "configuration": "Default", "publishedAt": "2026-09-25T00:00:00.000Z", "sourceScore": 79.8, "confidence": 47.5, "share": 0.3 } ], "fusion": null } }, "launchValidationCheck": { "prospective": { "n": 12, "mae": 9.3, "signedMedian": -1.9, "p80": 13.9, "coverage80": 0.75 }, "correctPairs": 30, "totalPairs": 46, "chronology": "All retained revision-zero prediction times precede settled ledger outcome times. This does not prove the upstream result was unavailable at prediction time." } }