{ "schemaVersion": 3, "id": "ayt-2026-fen", "kind": "exam", "title": { "tr": "2026 AYT · Fen Bilimleri", "en": "2026 AYT · Natural Sciences" }, "subtitle": { "tr": "Alan Yeterlilik Testi · Fen Bilimleri", "en": "Turkish university entrance exam · Natural Sciences" }, "category": { "tr": "Ulusal sınav", "en": "National exam" }, "description": { "tr": "Yapay zekâ modellerinin 2026 AYT Fen Bilimleri testindeki doğruluk, puan, maliyet ve token verimliliğini farklı akıl yürütme (reasoning effort) seviyelerinde karşılaştırır.", "en": "Compares AI models on the 2026 AYT Natural Sciences test across accuracy, score, cost, and token efficiency at different reasoning effort levels." }, "status": "published", "examDate": "2026-06-21", "runId": "2026-07-31T14-06-13Z", "runCompletedAt": "2026-08-12T04:37:33Z", "generatedAt": "2026-08-12T04:37:33Z", "itemCount": 40, "responseCount": 2960, "totalQuestions": 40, "score": { "label": { "tr": "Fen Puanı", "en": "Science Score" }, "min": 100.0, "max": 220.0, "formula": { "tr": "100 + Fen Bilimleri neti × 3", "en": "100 + Natural Sciences net × 3" } }, "netRule": { "tr": "net = doğru − yanlış/4; geçersiz yanıtlar yanlış sayılır.", "en": "net = correct − wrong/4; invalid responses count as wrong." }, "provenance": { "settingsSource": "run", "scoring": { "penaltyDivisor": 4.0, "base": 100.0, "coefficients": { "MAT": 3.0, "FEN": 3.0, "TDE-SB1": 3.0, "SB2": 3.0 } }, "invalidPolicy": "wrong" }, "sections": [ { "code": "FEN", "name": { "tr": "Fen Bilimleri", "en": "Natural Sciences" }, "questionCount": 40 } ], "models": [ { "modelId": "google/gemini-3.6-flash@minimal", "baseModelId": "google/gemini-3.6-flash", "modelSlug": "gemini-3-6-flash", "developerId": "google", "name": "Gemini 3.6 Flash (minimal)", "score": 216.25, "net": 38.75, "sections": { "FEN": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.1349, "known": true, "approximate": false }, "tokens": { "billed": 26469, "known": true, "approximate": false, "prompt": 10604, "completion": 15865, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "minimal", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 287.2498, "netPer1kTokens": 1.464 }, "rank": 1 }, { "modelId": "anthropic/claude-sonnet-5@medium", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (medium)", "score": 216.25, "net": 38.75, "sections": { "FEN": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.1875, "known": true, "approximate": false }, "tokens": { "billed": 33370, "known": true, "approximate": false, "prompt": 18280, "completion": 15090, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 206.6667, "netPer1kTokens": 1.1612 }, "rank": 2 }, { "modelId": "google/gemini-3.1-pro-preview@low", "baseModelId": "google/gemini-3.1-pro-preview", "modelSlug": "gemini-3-1-pro", "developerId": "google", "name": "Gemini 3.1 Pro (low)", "score": 216.25, "net": 38.75, "sections": { "FEN": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.3443, "known": true, "approximate": false }, "tokens": { "billed": 37465, "known": true, "approximate": false, "prompt": 10524, "completion": 26941, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 112.5472, "netPer1kTokens": 1.0343 }, "rank": 3 }, { "modelId": "qwen/qwen3.7-max", "baseModelId": "qwen/qwen3.7-max", "modelSlug": "qwen3-7-max", "developerId": "alibaba-qwen", "name": "Qwen3.7 Max", "score": 216.25, "net": 38.75, "sections": { "FEN": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.3497, "known": true, "approximate": false }, "tokens": { "billed": 86803, "known": true, "approximate": false, "prompt": 11654, "completion": 75149, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 110.8093, "netPer1kTokens": 0.4464 }, "rank": 4 }, { "modelId": "qwen/qwen3.6-27b", "baseModelId": "qwen/qwen3.6-27b", "modelSlug": "qwen3-6-27b", "developerId": "alibaba-qwen", "name": "Qwen3.6 27B", "score": 216.25, "net": 38.75, "sections": { "FEN": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.4574, "known": true, "approximate": true }, "tokens": { "billed": 136760, "known": true, "approximate": false, "prompt": 11654, "completion": 125106, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,8 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.8× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "io-net" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 84.718, "netPer1kTokens": 0.2833 }, "rank": 5 }, { "modelId": "google/gemini-3.1-pro-preview@medium", "baseModelId": "google/gemini-3.1-pro-preview", "modelSlug": "gemini-3-1-pro", "developerId": "google", "name": "Gemini 3.1 Pro (medium)", "score": 216.25, "net": 38.75, "sections": { "FEN": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.4861, "known": true, "approximate": false }, "tokens": { "billed": 49276, "known": true, "approximate": false, "prompt": 10524, "completion": 38752, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 79.7161, "netPer1kTokens": 0.7864 }, "rank": 6 }, { "modelId": "thinkingmachines/inkling@max", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (max)", "score": 216.25, "net": 38.75, "sections": { "FEN": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.522, "known": true, "approximate": false }, "tokens": { "billed": 137597, "known": true, "approximate": false, "prompt": 11563, "completion": 126034, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 74.2337, "netPer1kTokens": 0.2816 }, "rank": 7 }, { "modelId": "openai/gpt-5.6-luna@none", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (none)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0047, "known": true, "approximate": false }, "tokens": { "billed": 17269, "known": true, "approximate": false, "prompt": 11243, "completion": 6026, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "none", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 7978.7234, "netPer1kTokens": 2.1715 }, "rank": 8 }, { "modelId": "openai/gpt-5.6-luna@low", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (low)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0068, "known": true, "approximate": false }, "tokens": { "billed": 20659, "known": true, "approximate": false, "prompt": 11243, "completion": 9416, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 5514.7059, "netPer1kTokens": 1.8152 }, "rank": 9 }, { "modelId": "openai/gpt-5.6-luna@high", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (high)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0086, "known": true, "approximate": false }, "tokens": { "billed": 23643, "known": true, "approximate": false, "prompt": 11243, "completion": 12400, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 4360.4651, "netPer1kTokens": 1.5861 }, "rank": 10 }, { "modelId": "openai/gpt-5.6-luna@xhigh", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (xhigh)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0111, "known": true, "approximate": false }, "tokens": { "billed": 27879, "known": true, "approximate": false, "prompt": 11243, "completion": 16636, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 3378.3784, "netPer1kTokens": 1.3451 }, "rank": 11 }, { "modelId": "openai/gpt-5.6-luna@max", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (max)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0274, "known": true, "approximate": false }, "tokens": { "billed": 55035, "known": true, "approximate": false, "prompt": 11243, "completion": 43792, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 1368.6131, "netPer1kTokens": 0.6814 }, "rank": 12 }, { "modelId": "tencent/hy3@low", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (low)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0394, "known": true, "approximate": true }, "tokens": { "billed": 61855, "known": true, "approximate": false, "prompt": 13313, "completion": 48542, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,5 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.5× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 951.7766, "netPer1kTokens": 0.6063 }, "rank": 13 }, { "modelId": "openai/gpt-5.6-terra@low", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (low)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0537, "known": true, "approximate": false }, "tokens": { "billed": 18318, "known": true, "approximate": false, "prompt": 11243, "completion": 7075, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 698.324, "netPer1kTokens": 2.0472 }, "rank": 14 }, { "modelId": "openai/gpt-5.6-terra@medium", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (medium)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.057, "known": true, "approximate": false }, "tokens": { "billed": 18870, "known": true, "approximate": false, "prompt": 11243, "completion": 7627, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 657.8947, "netPer1kTokens": 1.9873 }, "rank": 15 }, { "modelId": "openai/gpt-5.6-terra@high", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (high)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0654, "known": true, "approximate": false }, "tokens": { "billed": 20270, "known": true, "approximate": false, "prompt": 11243, "completion": 9027, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 573.3945, "netPer1kTokens": 1.85 }, "rank": 16 }, { "modelId": "anthropic/claude-sonnet-5@low", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (low)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0706, "known": true, "approximate": false }, "tokens": { "billed": 21680, "known": true, "approximate": false, "prompt": 18280, "completion": 3400, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 531.1615, "netPer1kTokens": 1.7297 }, "rank": 17 }, { "modelId": "openai/gpt-5.6-terra@xhigh", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (xhigh)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0752, "known": true, "approximate": false }, "tokens": { "billed": 21904, "known": true, "approximate": false, "prompt": 11243, "completion": 10661, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 498.6702, "netPer1kTokens": 1.712 }, "rank": 18 }, { "modelId": "qwen/qwen3.7-plus", "baseModelId": "qwen/qwen3.7-plus", "modelSlug": "qwen3-7-plus", "developerId": "alibaba-qwen", "name": "Qwen3.7 Plus", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0916, "known": true, "approximate": false }, "tokens": { "billed": 80273, "known": true, "approximate": false, "prompt": 11654, "completion": 68619, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 409.3886, "netPer1kTokens": 0.4672 }, "rank": 19 }, { "modelId": "tencent/hy3@high", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (high)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.1166, "known": true, "approximate": true }, "tokens": { "billed": 162505, "known": true, "approximate": false, "prompt": 13313, "completion": 149192, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,5 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.5× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 321.6123, "netPer1kTokens": 0.2308 }, "rank": 20 }, { "modelId": "anthropic/claude-opus-5@low", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (low)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.1701, "known": true, "approximate": false }, "tokens": { "billed": 21427, "known": true, "approximate": false, "prompt": 18280, "completion": 3147, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 220.4586, "netPer1kTokens": 1.7501 }, "rank": 21 }, { "modelId": "openai/gpt-5.6-terra@max", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (max)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.1736, "known": true, "approximate": false }, "tokens": { "billed": 38309, "known": true, "approximate": false, "prompt": 11243, "completion": 27066, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 216.0138, "netPer1kTokens": 0.9789 }, "rank": 22 }, { "modelId": "x-ai/grok-4.5@low", "baseModelId": "x-ai/grok-4.5", "modelSlug": "grok-4-5", "developerId": "xai", "name": "Grok 4.5 (low)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.1778, "known": true, "approximate": false }, "tokens": { "billed": 43763, "known": true, "approximate": false, "prompt": 18845, "completion": 24918, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 210.9111, "netPer1kTokens": 0.8569 }, "rank": 23 }, { "modelId": "openai/gpt-5.6-sol@none", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (none)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.2012, "known": true, "approximate": false }, "tokens": { "billed": 16076, "known": true, "approximate": false, "prompt": 11243, "completion": 4833, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "none", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 186.3817, "netPer1kTokens": 2.3327 }, "rank": 24 }, { "modelId": "google/gemini-3.6-flash@low", "baseModelId": "google/gemini-3.6-flash", "modelSlug": "gemini-3-6-flash", "developerId": "google", "name": "Gemini 3.6 Flash (low)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.2037, "known": true, "approximate": false }, "tokens": { "billed": 35638, "known": true, "approximate": false, "prompt": 10604, "completion": 25034, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 184.0943, "netPer1kTokens": 1.0522 }, "rank": 25 }, { "modelId": "openai/gpt-5.6-sol@low", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (low)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.2819, "known": true, "approximate": false }, "tokens": { "billed": 18765, "known": true, "approximate": false, "prompt": 11243, "completion": 7522, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 133.0259, "netPer1kTokens": 1.9984 }, "rank": 26 }, { "modelId": "z-ai/glm-5.2@high", "baseModelId": "z-ai/glm-5.2", "modelSlug": "glm-5-2", "developerId": "zai", "name": "GLM 5.2 (high)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.2957, "known": true, "approximate": true }, "tokens": { "billed": 53096, "known": true, "approximate": false, "prompt": 12492, "completion": 40604, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,3 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.3× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "akashml", "ambient" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 126.8177, "netPer1kTokens": 0.7063 }, "rank": 27 }, { "modelId": "anthropic/claude-opus-5@medium", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (medium)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.3016, "known": true, "approximate": false }, "tokens": { "billed": 26689, "known": true, "approximate": false, "prompt": 18280, "completion": 8409, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 124.3369, "netPer1kTokens": 1.4051 }, "rank": 28 }, { "modelId": "openai/gpt-5.6-sol@medium", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (medium)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.3051, "known": true, "approximate": false }, "tokens": { "billed": 19540, "known": true, "approximate": false, "prompt": 11243, "completion": 8297, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 122.9105, "netPer1kTokens": 1.9191 }, "rank": 29 }, { "modelId": "google/gemini-3.6-flash@medium", "baseModelId": "google/gemini-3.6-flash", "modelSlug": "gemini-3-6-flash", "developerId": "google", "name": "Gemini 3.6 Flash (medium)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.345, "known": true, "approximate": false }, "tokens": { "billed": 54483, "known": true, "approximate": false, "prompt": 10604, "completion": 43879, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 108.6957, "netPer1kTokens": 0.6883 }, "rank": 30 }, { "modelId": "x-ai/grok-4.5@medium", "baseModelId": "x-ai/grok-4.5", "modelSlug": "grok-4-5", "developerId": "xai", "name": "Grok 4.5 (medium)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.3497, "known": true, "approximate": false }, "tokens": { "billed": 72339, "known": true, "approximate": false, "prompt": 18845, "completion": 53494, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 107.2348, "netPer1kTokens": 0.5184 }, "rank": 31 }, { "modelId": "x-ai/grok-4.5@high", "baseModelId": "x-ai/grok-4.5", "modelSlug": "grok-4-5", "developerId": "xai", "name": "Grok 4.5 (high)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.3503, "known": true, "approximate": false }, "tokens": { "billed": 72507, "known": true, "approximate": false, "prompt": 18845, "completion": 53662, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 107.0511, "netPer1kTokens": 0.5172 }, "rank": 32 }, { "modelId": "openai/gpt-5.6-sol@high", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (high)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.3628, "known": true, "approximate": false }, "tokens": { "billed": 21464, "known": true, "approximate": false, "prompt": 11243, "completion": 10221, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 103.3627, "netPer1kTokens": 1.7471 }, "rank": 33 }, { "modelId": "openai/gpt-5.6-sol@xhigh", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (xhigh)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.3673, "known": true, "approximate": false }, "tokens": { "billed": 21611, "known": true, "approximate": false, "prompt": 11243, "completion": 10368, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 102.0964, "netPer1kTokens": 1.7352 }, "rank": 34 }, { "modelId": "google/gemini-3.6-flash@high", "baseModelId": "google/gemini-3.6-flash", "modelSlug": "gemini-3-6-flash", "developerId": "google", "name": "Gemini 3.6 Flash (high)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.4173, "known": true, "approximate": false }, "tokens": { "billed": 64122, "known": true, "approximate": false, "prompt": 10604, "completion": 53518, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 89.8634, "netPer1kTokens": 0.5848 }, "rank": 35 }, { "modelId": "anthropic/claude-opus-5@high", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (high)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.481, "known": true, "approximate": false }, "tokens": { "billed": 33866, "known": true, "approximate": false, "prompt": 18280, "completion": 15586, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 77.9626, "netPer1kTokens": 1.1073 }, "rank": 36 }, { "modelId": "openai/gpt-5.6-sol@max", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (max)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.4878, "known": true, "approximate": false }, "tokens": { "billed": 25629, "known": true, "approximate": false, "prompt": 11243, "completion": 14386, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 76.8758, "netPer1kTokens": 1.4632 }, "rank": 37 }, { "modelId": "anthropic/claude-opus-5@xhigh", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (xhigh)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.5386, "known": true, "approximate": false }, "tokens": { "billed": 36168, "known": true, "approximate": false, "prompt": 18280, "completion": 17888, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 69.625, "netPer1kTokens": 1.0368 }, "rank": 38 }, { "modelId": "openrouter/fusion", "baseModelId": "openrouter/fusion", "modelSlug": "openrouter-fusion", "developerId": "openrouter", "name": "OpenRouter Fusion", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.6609, "known": true, "approximate": false }, "tokens": { "billed": 68322, "known": true, "approximate": false, "prompt": 52360, "completion": 15962, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 56.7408, "netPer1kTokens": 0.5489 }, "rank": 39 }, { "modelId": "anthropic/claude-opus-5@max", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (max)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.677, "known": true, "approximate": false }, "tokens": { "billed": 41704, "known": true, "approximate": false, "prompt": 18280, "completion": 23424, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 55.3914, "netPer1kTokens": 0.8992 }, "rank": 40 }, { "modelId": "google/gemini-3.1-pro-preview@high", "baseModelId": "google/gemini-3.1-pro-preview", "modelSlug": "gemini-3-1-pro", "developerId": "google", "name": "Gemini 3.1 Pro (high)", "score": 212.5, "net": 37.5, "sections": { "FEN": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.7446, "known": true, "approximate": false }, "tokens": { "billed": 70816, "known": true, "approximate": false, "prompt": 10524, "completion": 60292, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 50.3626, "netPer1kTokens": 0.5295 }, "rank": 41 }, { "modelId": "openai/gpt-5.6-luna@medium", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (medium)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0074, "known": true, "approximate": false }, "tokens": { "billed": 21692, "known": true, "approximate": false, "prompt": 11243, "completion": 10449, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 4898.6486, "netPer1kTokens": 1.6711 }, "rank": 42 }, { "modelId": "google/gemma-4-31b-it", "baseModelId": "google/gemma-4-31b-it", "modelSlug": "gemma-4-31b", "developerId": "google", "name": "Gemma 4 31B", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0094, "known": true, "approximate": true }, "tokens": { "billed": 31213, "known": true, "approximate": false, "prompt": 11246, "completion": 19967, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,2 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.2× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 3856.383, "netPer1kTokens": 1.1614 }, "rank": 43 }, { "modelId": "tencent/hy3@none", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (none)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0096, "known": true, "approximate": true }, "tokens": { "billed": 23803, "known": true, "approximate": false, "prompt": 13433, "completion": 10370, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,5 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.5× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "none", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 3776.0417, "netPer1kTokens": 1.5229 }, "rank": 44 }, { "modelId": "deepseek/deepseek-v4-flash@high", "baseModelId": "deepseek/deepseek-v4-flash", "modelSlug": "deepseek-v4-flash", "developerId": "deepseek", "name": "DeepSeek V4 Flash (high)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 2, "invalid": 1, "net": 36.25 } }, "cost": { "usd": 0.016, "known": true, "approximate": false }, "tokens": { "billed": 64104, "known": true, "approximate": false, "prompt": 14225, "completion": 49879, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "akashml", "io-net" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 2265.625, "netPer1kTokens": 0.5655 }, "rank": 45 }, { "modelId": "inclusionai/ring-2.6-1t@high", "baseModelId": "inclusionai/ring-2.6-1t", "modelSlug": "ring-2-6-1t", "developerId": "inclusion-ai", "name": "Ring-2.6-1T (high)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 2, "invalid": 1, "net": 36.25 } }, "cost": { "usd": 0.0215, "known": true, "approximate": false }, "tokens": { "billed": 47308, "known": true, "approximate": false, "prompt": 12950, "completion": 34358, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 1686.0465, "netPer1kTokens": 0.7663 }, "rank": 46 }, { "modelId": "thinkingmachines/inkling@minimal", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (minimal)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.037, "known": true, "approximate": false }, "tokens": { "billed": 17848, "known": true, "approximate": false, "prompt": 11563, "completion": 6285, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "minimal", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 979.7297, "netPer1kTokens": 2.031 }, "rank": 47 }, { "modelId": "thinkingmachines/inkling@medium", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (medium)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0877, "known": true, "approximate": false }, "tokens": { "billed": 30362, "known": true, "approximate": false, "prompt": 11563, "completion": 18799, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 413.3409, "netPer1kTokens": 1.1939 }, "rank": 48 }, { "modelId": "meituan/longcat-2.0", "baseModelId": "meituan/longcat-2.0", "modelSlug": "longcat-2-0", "developerId": "meituan", "name": "LongCat 2.0", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.1655, "known": true, "approximate": false }, "tokens": { "billed": 148737, "known": true, "approximate": false, "prompt": 14443, "completion": 134294, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 219.0332, "netPer1kTokens": 0.2437 }, "rank": 49 }, { "modelId": "nvidia/nemotron-3-ultra-550b-a55b@medium", "baseModelId": "nvidia/nemotron-3-ultra-550b-a55b", "modelSlug": "nemotron-3-ultra", "developerId": "nvidia", "name": "Nemotron 3 Ultra (medium)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.2622, "known": true, "approximate": false }, "tokens": { "billed": 83072, "known": true, "approximate": false, "prompt": 12271, "completion": 70801, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": { "ignore": [ "baseten" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 138.2532, "netPer1kTokens": 0.4364 }, "rank": 50 }, { "modelId": "deepseek/deepseek-v4-pro@high", "baseModelId": "deepseek/deepseek-v4-pro", "modelSlug": "deepseek-v4-pro", "developerId": "deepseek", "name": "DeepSeek V4 Pro (high)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 2, "invalid": 1, "net": 36.25 } }, "cost": { "usd": 0.2655, "known": true, "approximate": true }, "tokens": { "billed": 85664, "known": true, "approximate": false, "prompt": 14225, "completion": 71439, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; model allow_fallbacks ile koşuldu ve 2026-07-31 usage.cost ölçümünde gerçek fatura hesaplananın ≈3,9 katına ulaştı.", "en": "Cost computed from the list price; the model ran with allow_fallbacks and a 2026-07-31 usage.cost probe measured the actual bill at ≈3.9× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 136.5348, "netPer1kTokens": 0.4232 }, "rank": 51 }, { "modelId": "thinkingmachines/inkling@high", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (high)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.2943, "known": true, "approximate": false }, "tokens": { "billed": 81384, "known": true, "approximate": false, "prompt": 11563, "completion": 69821, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 123.1736, "netPer1kTokens": 0.4454 }, "rank": 52 }, { "modelId": "nvidia/nemotron-3-ultra-550b-a55b@high", "baseModelId": "nvidia/nemotron-3-ultra-550b-a55b", "modelSlug": "nemotron-3-ultra", "developerId": "nvidia", "name": "Nemotron 3 Ultra (high)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.3182, "known": true, "approximate": false }, "tokens": { "billed": 98620, "known": true, "approximate": false, "prompt": 12271, "completion": 86349, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "ignore": [ "baseten" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 113.9221, "netPer1kTokens": 0.3676 }, "rank": 53 }, { "modelId": "anthropic/claude-sonnet-5@high", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (high)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.3297, "known": true, "approximate": false }, "tokens": { "billed": 47598, "known": true, "approximate": false, "prompt": 18280, "completion": 29318, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 109.9484, "netPer1kTokens": 0.7616 }, "rank": 54 }, { "modelId": "moonshotai/kimi-k3@low", "baseModelId": "moonshotai/kimi-k3", "modelSlug": "kimi-k3", "developerId": "moonshot", "name": "Kimi K3 (low)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.4432, "known": true, "approximate": false }, "tokens": { "billed": 36445, "known": true, "approximate": false, "prompt": 17759, "completion": 18686, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": { "ignore": [ "nebius", "moonshotai" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 81.7915, "netPer1kTokens": 0.9946 }, "rank": 55 }, { "modelId": "anthropic/claude-sonnet-5@xhigh", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (xhigh)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.5252, "known": true, "approximate": false }, "tokens": { "billed": 67145, "known": true, "approximate": false, "prompt": 18280, "completion": 48865, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 69.0213, "netPer1kTokens": 0.5399 }, "rank": 56 }, { "modelId": "moonshotai/kimi-k3@high", "baseModelId": "moonshotai/kimi-k3", "modelSlug": "kimi-k3", "developerId": "moonshot", "name": "Kimi K3 (high)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.8087, "known": true, "approximate": false }, "tokens": { "billed": 55096, "known": true, "approximate": false, "prompt": 17735, "completion": 37361, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "ignore": [ "nebius", "moonshotai" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 44.825, "netPer1kTokens": 0.6579 }, "rank": 57 }, { "modelId": "moonshotai/kimi-k3@max", "baseModelId": "moonshotai/kimi-k3", "modelSlug": "kimi-k3", "developerId": "moonshot", "name": "Kimi K3 (max)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 1.2075, "known": true, "approximate": false }, "tokens": { "billed": 70615, "known": true, "approximate": false, "prompt": 17687, "completion": 52928, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": { "ignore": [ "nebius", "moonshotai" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 30.0207, "netPer1kTokens": 0.5133 }, "rank": 58 }, { "modelId": "z-ai/glm-5.2@xhigh", "baseModelId": "z-ai/glm-5.2", "modelSlug": "glm-5-2", "developerId": "zai", "name": "GLM 5.2 (xhigh)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 2, "invalid": 1, "net": 36.25 } }, "cost": { "usd": 1.5508, "known": true, "approximate": true }, "tokens": { "billed": 247730, "known": true, "approximate": false, "prompt": 12492, "completion": 235238, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,3 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.3× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "akashml", "ambient" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 23.375, "netPer1kTokens": 0.1463 }, "rank": 59 }, { "modelId": "anthropic/claude-sonnet-5@max", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (max)", "score": 208.75, "net": 36.25, "sections": { "FEN": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 1.7001, "known": true, "approximate": false }, "tokens": { "billed": 184631, "known": true, "approximate": false, "prompt": 18280, "completion": 166351, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 21.3223, "netPer1kTokens": 0.1963 }, "rank": 60 }, { "modelId": "deepseek/deepseek-v4-flash@xhigh", "baseModelId": "deepseek/deepseek-v4-flash", "modelSlug": "deepseek-v4-flash", "developerId": "deepseek", "name": "DeepSeek V4 Flash (xhigh)", "score": 205.0, "net": 35.0, "sections": { "FEN": { "correct": 36, "wrong": 3, "invalid": 1, "net": 35.0 } }, "cost": { "usd": 0.0368, "known": true, "approximate": false }, "tokens": { "billed": 140227, "known": true, "approximate": false, "prompt": 17385, "completion": 122842, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "akashml", "io-net" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 951.087, "netPer1kTokens": 0.2496 }, "rank": 61 }, { "modelId": "thinkingmachines/inkling@low", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (low)", "score": 205.0, "net": 35.0, "sections": { "FEN": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.0369, "known": true, "approximate": false }, "tokens": { "billed": 17822, "known": true, "approximate": false, "prompt": 11563, "completion": 6259, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 948.5095, "netPer1kTokens": 1.9639 }, "rank": 62 }, { "modelId": "inclusionai/ring-2.6-1t@xhigh", "baseModelId": "inclusionai/ring-2.6-1t", "modelSlug": "ring-2-6-1t", "developerId": "inclusion-ai", "name": "Ring-2.6-1T (xhigh)", "score": 205.0, "net": 35.0, "sections": { "FEN": { "correct": 36, "wrong": 1, "invalid": 3, "net": 35.0 } }, "cost": { "usd": 0.0606, "known": true, "approximate": false }, "tokens": { "billed": 139254, "known": true, "approximate": false, "prompt": 12788, "completion": 126466, "orchestration": 0 }, "errors": { "invalid": 3, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 577.5578, "netPer1kTokens": 0.2513 }, "rank": 63 }, { "modelId": "meta/muse-glimmer-30b@high", "baseModelId": "meta/muse-glimmer-30b", "modelSlug": "muse-glimmer-30b", "developerId": "meta", "name": "Muse Glimmer 30B (high)", "score": 205.0, "net": 35.0, "sections": { "FEN": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.0712, "known": true, "approximate": false }, "tokens": { "billed": 56822, "known": true, "approximate": false, "prompt": 11147, "completion": 45675, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Koşu ayarlarındaki 65.536 istenen tavandır; efektif tavan 32.768'dir. Diğer varyantlar 131.072 ile koştu, ancak modelin bağlamı 131.072 olduğundan genel tavan gönderilemezdi. Pinlenen Together endpoint'i ilan etmediği bir sınırla talebi 32.768'de kesiyor. 2026-08-12 ölçümünde bu tavanda hiçbir yanıt kesilmedi.", "en": "The 65,536 shown in run settings is the requested ceiling; the effective ceiling is 32,768. The other variants ran with 131,072, but the run-wide ceiling could not be sent because this model's context is 131,072. The pinned Together endpoint enforces an undeclared limit that caps the request at 32,768. In the 2026-08-12 measurement no response was truncated at this ceiling." }, "settings": { "temperature": 0.0, "maxTokens": 65536, "reasoningEffort": "high", "requestedRoutingPolicy": { "ignore": [ "deepinfra" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 491.573, "netPer1kTokens": 0.616 }, "rank": 64 }, { "modelId": "deepseek/deepseek-v4-pro@xhigh", "baseModelId": "deepseek/deepseek-v4-pro", "modelSlug": "deepseek-v4-pro", "developerId": "deepseek", "name": "DeepSeek V4 Pro (xhigh)", "score": 205.0, "net": 35.0, "sections": { "FEN": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.5034, "known": true, "approximate": true }, "tokens": { "billed": 157629, "known": true, "approximate": false, "prompt": 17385, "completion": 140244, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; model allow_fallbacks ile koşuldu ve 2026-07-31 usage.cost ölçümünde gerçek fatura hesaplananın ≈3,9 katına ulaştı.", "en": "Cost computed from the list price; the model ran with allow_fallbacks and a 2026-07-31 usage.cost probe measured the actual bill at ≈3.9× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 69.5272, "netPer1kTokens": 0.222 }, "rank": 65 }, { "modelId": "openai/gpt-5.6-terra@none", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (none)", "score": 201.25, "net": 33.75, "sections": { "FEN": { "correct": 35, "wrong": 5, "invalid": 0, "net": 33.75 } }, "cost": { "usd": 0.0313, "known": true, "approximate": false }, "tokens": { "billed": 14583, "known": true, "approximate": false, "prompt": 11243, "completion": 3340, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "none", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 1078.2748, "netPer1kTokens": 2.3143 }, "rank": 66 }, { "modelId": "xiaomi/mimo-v2.5-pro", "baseModelId": "xiaomi/mimo-v2.5-pro", "modelSlug": "mimo-v2-5-pro", "developerId": "xiaomi", "name": "MiMo V2.5 Pro", "score": 201.25, "net": 33.75, "sections": { "FEN": { "correct": 35, "wrong": 3, "invalid": 2, "net": 33.75 } }, "cost": { "usd": 0.0467, "known": true, "approximate": false }, "tokens": { "billed": 59996, "known": true, "approximate": false, "prompt": 12306, "completion": 47690, "orchestration": 0 }, "errors": { "invalid": 2, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 722.6981, "netPer1kTokens": 0.5625 }, "rank": 67 }, { "modelId": "thinkingmachines/inkling@none", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (none)", "score": 197.5, "net": 32.5, "sections": { "FEN": { "correct": 34, "wrong": 6, "invalid": 0, "net": 32.5 } }, "cost": { "usd": 0.0416, "known": true, "approximate": false }, "tokens": { "billed": 18908, "known": true, "approximate": false, "prompt": 11483, "completion": 7425, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "none", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 781.25, "netPer1kTokens": 1.7188 }, "rank": 68 }, { "modelId": "minimax/minimax-m3", "baseModelId": "minimax/minimax-m3", "modelSlug": "minimax-m3", "developerId": "minimax", "name": "MiniMax M3", "score": 190.0, "net": 30.0, "sections": { "FEN": { "correct": 32, "wrong": 7, "invalid": 1, "net": 30.0 } }, "cost": { "usd": 0.0164, "known": true, "approximate": false }, "tokens": { "billed": 96652, "known": true, "approximate": false, "prompt": 19825, "completion": 76827, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 1829.2683, "netPer1kTokens": 0.3104 }, "rank": 69 }, { "modelId": "anthropic/claude-fable-5@low", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (low)", "score": 171.25, "net": 23.75, "sections": { "FEN": { "correct": 27, "wrong": 0, "invalid": 13, "net": 23.75 } }, "cost": { "usd": 0.2166, "known": true, "approximate": false }, "tokens": { "billed": 19926, "known": true, "approximate": false, "prompt": 18280, "completion": 1646, "orchestration": 0 }, "errors": { "invalid": 13, "apiFailures": 0 }, "note": { "tr": "40 sorunun 13'ünde (biyoloji) sağlayıcı içerik filtresi yanıtı engelledi; model bu sorulara hiç girmedi. Bu bölümdeki net, modelin bilgisini değil filtre davranışını yansıtır.", "en": "On 13 of 40 questions (biology) the provider's content filter blocked the response; the model never engaged with them. The section net reflects filter behaviour, not knowledge." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 109.6491, "netPer1kTokens": 1.1919 }, "rank": 70 }, { "modelId": "anthropic/claude-fable-5@medium", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (medium)", "score": 171.25, "net": 23.75, "sections": { "FEN": { "correct": 27, "wrong": 0, "invalid": 13, "net": 23.75 } }, "cost": { "usd": 0.4393, "known": true, "approximate": false }, "tokens": { "billed": 24388, "known": true, "approximate": false, "prompt": 18280, "completion": 6108, "orchestration": 0 }, "errors": { "invalid": 13, "apiFailures": 0 }, "note": { "tr": "40 sorunun 13'ünde (biyoloji) sağlayıcı içerik filtresi yanıtı engelledi; model bu sorulara hiç girmedi. Bu bölümdeki net, modelin bilgisini değil filtre davranışını yansıtır.", "en": "On 13 of 40 questions (biology) the provider's content filter blocked the response; the model never engaged with them. The section net reflects filter behaviour, not knowledge." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 54.0633, "netPer1kTokens": 0.9738 }, "rank": 71 }, { "modelId": "anthropic/claude-fable-5@high", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (high)", "score": 171.25, "net": 23.75, "sections": { "FEN": { "correct": 27, "wrong": 0, "invalid": 13, "net": 23.75 } }, "cost": { "usd": 0.6513, "known": true, "approximate": false }, "tokens": { "billed": 28648, "known": true, "approximate": false, "prompt": 18280, "completion": 10368, "orchestration": 0 }, "errors": { "invalid": 13, "apiFailures": 0 }, "note": { "tr": "40 sorunun 13'ünde (biyoloji) sağlayıcı içerik filtresi yanıtı engelledi; model bu sorulara hiç girmedi. Bu bölümdeki net, modelin bilgisini değil filtre davranışını yansıtır.", "en": "On 13 of 40 questions (biology) the provider's content filter blocked the response; the model never engaged with them. The section net reflects filter behaviour, not knowledge." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 36.4655, "netPer1kTokens": 0.829 }, "rank": 72 }, { "modelId": "anthropic/claude-fable-5@xhigh", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (xhigh)", "score": 171.25, "net": 23.75, "sections": { "FEN": { "correct": 27, "wrong": 0, "invalid": 13, "net": 23.75 } }, "cost": { "usd": 0.9811, "known": true, "approximate": false }, "tokens": { "billed": 35249, "known": true, "approximate": false, "prompt": 18280, "completion": 16969, "orchestration": 0 }, "errors": { "invalid": 13, "apiFailures": 0 }, "note": { "tr": "40 sorunun 13'ünde (biyoloji) sağlayıcı içerik filtresi yanıtı engelledi; model bu sorulara hiç girmedi. Bu bölümdeki net, modelin bilgisini değil filtre davranışını yansıtır.", "en": "On 13 of 40 questions (biology) the provider's content filter blocked the response; the model never engaged with them. The section net reflects filter behaviour, not knowledge." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 24.2075, "netPer1kTokens": 0.6738 }, "rank": 73 }, { "modelId": "anthropic/claude-fable-5@max", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (max)", "score": 171.25, "net": 23.75, "sections": { "FEN": { "correct": 27, "wrong": 0, "invalid": 13, "net": 23.75 } }, "cost": { "usd": 1.3733, "known": true, "approximate": false }, "tokens": { "billed": 43087, "known": true, "approximate": false, "prompt": 18280, "completion": 24807, "orchestration": 0 }, "errors": { "invalid": 13, "apiFailures": 0 }, "note": { "tr": "40 sorunun 13'ünde (biyoloji) sağlayıcı içerik filtresi yanıtı engelledi; model bu sorulara hiç girmedi. Bu bölümdeki net, modelin bilgisini değil filtre davranışını yansıtır.", "en": "On 13 of 40 questions (biology) the provider's content filter blocked the response; the model never engaged with them. The section net reflects filter behaviour, not knowledge." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 17.2941, "netPer1kTokens": 0.5512 }, "rank": 74 } ], "questionStats": [ { "id": "FEN-1", "number": 1, "section": "FEN", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "FEN-2", "number": 2, "section": "FEN", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "FEN-3", "number": 3, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-4", "number": 4, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-5", "number": 5, "section": "FEN", "cancelled": false, "correct": 71, "responses": 74 }, { "id": "FEN-6", "number": 6, "section": "FEN", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "FEN-7", "number": 7, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-8", "number": 8, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-9", "number": 9, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-10", "number": 10, "section": "FEN", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "FEN-11", "number": 11, "section": "FEN", "cancelled": false, "correct": 70, "responses": 74 }, { "id": "FEN-12", "number": 12, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-13", "number": 13, "section": "FEN", "cancelled": false, "correct": 71, "responses": 74 }, { "id": "FEN-14", "number": 14, "section": "FEN", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "FEN-15", "number": 15, "section": "FEN", "cancelled": false, "correct": 71, "responses": 74 }, { "id": "FEN-16", "number": 16, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-17", "number": 17, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-18", "number": 18, "section": "FEN", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "FEN-19", "number": 19, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-20", "number": 20, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-21", "number": 21, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-22", "number": 22, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-23", "number": 23, "section": "FEN", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "FEN-24", "number": 24, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-25", "number": 25, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-26", "number": 26, "section": "FEN", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "FEN-27", "number": 27, "section": "FEN", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "FEN-28", "number": 28, "section": "FEN", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "FEN-29", "number": 29, "section": "FEN", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "FEN-30", "number": 30, "section": "FEN", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "FEN-31", "number": 31, "section": "FEN", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "FEN-32", "number": 32, "section": "FEN", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "FEN-33", "number": 33, "section": "FEN", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "FEN-34", "number": 34, "section": "FEN", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "FEN-35", "number": 35, "section": "FEN", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "FEN-36", "number": 36, "section": "FEN", "cancelled": false, "correct": 8, "responses": 74 }, { "id": "FEN-37", "number": 37, "section": "FEN", "cancelled": false, "correct": 1, "responses": 74 }, { "id": "FEN-38", "number": 38, "section": "FEN", "cancelled": false, "correct": 68, "responses": 74 }, { "id": "FEN-39", "number": 39, "section": "FEN", "cancelled": false, "correct": 68, "responses": 74 }, { "id": "FEN-40", "number": 40, "section": "FEN", "cancelled": false, "correct": 49, "responses": 74 } ] }