{ "schemaVersion": 3, "id": "ayt-2026-sb2", "kind": "exam", "title": { "tr": "2026 AYT · Sosyal Bilimler-2", "en": "2026 AYT · Social Sciences-2" }, "subtitle": { "tr": "Alan Yeterlilik Testi · Sosyal Bilimler-2", "en": "Turkish university entrance exam · Social Sciences-2" }, "category": { "tr": "Ulusal sınav", "en": "National exam" }, "description": { "tr": "Yapay zekâ modellerinin 2026 AYT Sosyal Bilimler-2 testindeki doğruluk, puan, maliyet ve token verimliliğini farklı akıl yürütme (reasoning effort) seviyelerinde karşılaştırır.", "en": "Compares AI models on the 2026 AYT Social Sciences-2 test across accuracy, score, cost, and token efficiency at different reasoning effort levels." }, "status": "published", "examDate": "2026-06-21", "runId": "2026-07-31T14-58-12Z", "runCompletedAt": "2026-08-12T04:38:43Z", "generatedAt": "2026-08-12T04:38:43Z", "itemCount": 40, "responseCount": 2960, "totalQuestions": 40, "score": { "label": { "tr": "SB2 Puanı", "en": "SB2 Score" }, "min": 100.0, "max": 220.0, "formula": { "tr": "100 + Sosyal Bilimler-2 neti × 3", "en": "100 + Social Sciences-2 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": "SB2", "name": { "tr": "Sosyal Bilimler-2", "en": "Social Sciences-2" }, "questionCount": 40 } ], "models": [ { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.0047, "known": true, "approximate": false }, "tokens": { "billed": 17564, "known": true, "approximate": false, "prompt": 11692, "completion": 5872, "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": 8510.6383, "netPer1kTokens": 2.2774 }, "rank": 1 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.0054, "known": true, "approximate": false }, "tokens": { "billed": 18754, "known": true, "approximate": false, "prompt": 11692, "completion": 7062, "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": 7407.4074, "netPer1kTokens": 2.1329 }, "rank": 2 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.0076, "known": true, "approximate": false }, "tokens": { "billed": 22353, "known": true, "approximate": false, "prompt": 11692, "completion": 10661, "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": 5263.1579, "netPer1kTokens": 1.7895 }, "rank": 3 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.0108, "known": true, "approximate": false }, "tokens": { "billed": 25774, "known": true, "approximate": false, "prompt": 11692, "completion": 14082, "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": 3703.7037, "netPer1kTokens": 1.552 }, "rank": 4 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.0149, "known": true, "approximate": false }, "tokens": { "billed": 34500, "known": true, "approximate": false, "prompt": 11692, "completion": 22808, "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": 2684.5638, "netPer1kTokens": 1.1594 }, "rank": 5 }, { "modelId": "thinkingmachines/inkling@low", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (low)", "score": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.0249, "known": true, "approximate": false }, "tokens": { "billed": 15183, "known": true, "approximate": false, "prompt": 12012, "completion": 3171, "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": 1606.4257, "netPer1kTokens": 2.6345 }, "rank": 6 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.0405, "known": true, "approximate": false }, "tokens": { "billed": 16491, "known": true, "approximate": false, "prompt": 11692, "completion": 4799, "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": 987.6543, "netPer1kTokens": 2.4256 }, "rank": 7 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.0482, "known": true, "approximate": false }, "tokens": { "billed": 17770, "known": true, "approximate": false, "prompt": 11692, "completion": 6078, "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": 829.8755, "netPer1kTokens": 2.251 }, "rank": 8 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.0525, "known": true, "approximate": false }, "tokens": { "billed": 18491, "known": true, "approximate": false, "prompt": 11692, "completion": 6799, "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": 761.9048, "netPer1kTokens": 2.1632 }, "rank": 9 }, { "modelId": "thinkingmachines/inkling@medium", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (medium)", "score": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.0558, "known": true, "approximate": false }, "tokens": { "billed": 22822, "known": true, "approximate": false, "prompt": 12012, "completion": 10810, "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": 716.8459, "netPer1kTokens": 1.7527 }, "rank": 10 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.079, "known": true, "approximate": false }, "tokens": { "billed": 19213, "known": true, "approximate": false, "prompt": 10854, "completion": 8359, "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": 506.3291, "netPer1kTokens": 2.0819 }, "rank": 11 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.1281, "known": true, "approximate": false }, "tokens": { "billed": 25759, "known": true, "approximate": false, "prompt": 10854, "completion": 14905, "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": 312.256, "netPer1kTokens": 1.5529 }, "rank": 12 }, { "modelId": "anthropic/claude-sonnet-5@high", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (high)", "score": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.1313, "known": true, "approximate": false }, "tokens": { "billed": 29086, "known": true, "approximate": false, "prompt": 19941, "completion": 9145, "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": 304.6458, "netPer1kTokens": 1.3752 }, "rank": 13 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.1429, "known": true, "approximate": false }, "tokens": { "billed": 14507, "known": true, "approximate": false, "prompt": 11692, "completion": 2815, "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": 279.916, "netPer1kTokens": 2.7573 }, "rank": 14 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.1843, "known": true, "approximate": false }, "tokens": { "billed": 40463, "known": true, "approximate": false, "prompt": 11692, "completion": 28771, "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": 217.0374, "netPer1kTokens": 0.9886 }, "rank": 15 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.1862, "known": true, "approximate": false }, "tokens": { "billed": 15950, "known": true, "approximate": false, "prompt": 11692, "completion": 4258, "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": 214.8228, "netPer1kTokens": 2.5078 }, "rank": 16 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.197, "known": true, "approximate": false }, "tokens": { "billed": 16311, "known": true, "approximate": false, "prompt": 11692, "completion": 4619, "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": 203.0457, "netPer1kTokens": 2.4523 }, "rank": 17 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.2212, "known": true, "approximate": false }, "tokens": { "billed": 38177, "known": true, "approximate": false, "prompt": 10854, "completion": 27323, "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": 180.8318, "netPer1kTokens": 1.0478 }, "rank": 18 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.2245, "known": true, "approximate": false }, "tokens": { "billed": 17227, "known": true, "approximate": false, "prompt": 11692, "completion": 5535, "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": 178.1737, "netPer1kTokens": 2.3219 }, "rank": 19 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.2284, "known": true, "approximate": false }, "tokens": { "billed": 17358, "known": true, "approximate": false, "prompt": 11692, "completion": 5666, "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": 175.1313, "netPer1kTokens": 2.3044 }, "rank": 20 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.2436, "known": true, "approximate": false }, "tokens": { "billed": 29282, "known": true, "approximate": false, "prompt": 10774, "completion": 18508, "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": 164.2036, "netPer1kTokens": 1.366 }, "rank": 21 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.2612, "known": true, "approximate": false }, "tokens": { "billed": 43512, "known": true, "approximate": false, "prompt": 10854, "completion": 32658, "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": 153.1394, "netPer1kTokens": 0.9193 }, "rank": 22 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.3173, "known": true, "approximate": false }, "tokens": { "billed": 35421, "known": true, "approximate": false, "prompt": 10774, "completion": 24647, "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": 126.0637, "netPer1kTokens": 1.1293 }, "rank": 23 }, { "modelId": "thinkingmachines/inkling@max", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (max)", "score": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.3476, "known": true, "approximate": false }, "tokens": { "billed": 94874, "known": true, "approximate": false, "prompt": 12012, "completion": 82862, "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": 115.0748, "netPer1kTokens": 0.4216 }, "rank": 24 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.4442, "known": true, "approximate": false }, "tokens": { "billed": 45995, "known": true, "approximate": false, "prompt": 10774, "completion": 35221, "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": 90.0495, "netPer1kTokens": 0.8697 }, "rank": 25 }, { "modelId": "anthropic/claude-opus-5@xhigh", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (xhigh)", "score": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.5107, "known": true, "approximate": false }, "tokens": { "billed": 36382, "known": true, "approximate": false, "prompt": 19941, "completion": 16441, "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": 78.3239, "netPer1kTokens": 1.0994 }, "rank": 26 }, { "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": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.5902, "known": true, "approximate": false }, "tokens": { "billed": 29418, "known": true, "approximate": false, "prompt": 11692, "completion": 17726, "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": 67.7736, "netPer1kTokens": 1.3597 }, "rank": 27 }, { "modelId": "openrouter/fusion", "baseModelId": "openrouter/fusion", "modelSlug": "openrouter-fusion", "developerId": "openrouter", "name": "OpenRouter Fusion", "score": 220.0, "net": 40.0, "sections": { "SB2": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 1.0828, "known": true, "approximate": false }, "tokens": { "billed": 82188, "known": true, "approximate": false, "prompt": 64745, "completion": 17443, "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": 36.9413, "netPer1kTokens": 0.4867 }, "rank": 28 }, { "modelId": "thinkingmachines/inkling@minimal", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (minimal)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.0263, "known": true, "approximate": false }, "tokens": { "billed": 15541, "known": true, "approximate": false, "prompt": 12012, "completion": 3529, "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": 1473.384, "netPer1kTokens": 2.4934 }, "rank": 29 }, { "modelId": "tencent/hy3@low", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (low)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.0314, "known": true, "approximate": true }, "tokens": { "billed": 51014, "known": true, "approximate": false, "prompt": 14528, "completion": 36486, "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": 1234.0764, "netPer1kTokens": 0.7596 }, "rank": 30 }, { "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": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.0333, "known": true, "approximate": false }, "tokens": { "billed": 44821, "known": true, "approximate": false, "prompt": 13087, "completion": 31734, "orchestration": 0 }, "errors": { "invalid": 0, "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": 1163.6637, "netPer1kTokens": 0.8646 }, "rank": 31 }, { "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": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.044, "known": true, "approximate": false }, "tokens": { "billed": 17072, "known": true, "approximate": false, "prompt": 11692, "completion": 5380, "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": 880.6818, "netPer1kTokens": 2.2698 }, "rank": 32 }, { "modelId": "minimax/minimax-m3", "baseModelId": "minimax/minimax-m3", "modelSlug": "minimax-m3", "developerId": "minimax", "name": "MiniMax M3", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.0485, "known": true, "approximate": false }, "tokens": { "billed": 58573, "known": true, "approximate": false, "prompt": 21540, "completion": 37033, "orchestration": 0 }, "errors": { "invalid": 0, "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": 798.9691, "netPer1kTokens": 0.6616 }, "rank": 33 }, { "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": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.0486, "known": true, "approximate": false }, "tokens": { "billed": 20809, "known": true, "approximate": false, "prompt": 19941, "completion": 868, "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": 797.3251, "netPer1kTokens": 1.8622 }, "rank": 34 }, { "modelId": "anthropic/claude-opus-5@low", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (low)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.1282, "known": true, "approximate": false }, "tokens": { "billed": 21079, "known": true, "approximate": false, "prompt": 19941, "completion": 1138, "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": 302.2621, "netPer1kTokens": 1.8383 }, "rank": 35 }, { "modelId": "thinkingmachines/inkling@high", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (high)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.1496, "known": true, "approximate": false }, "tokens": { "billed": 45984, "known": true, "approximate": false, "prompt": 12012, "completion": 33972, "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": 259.0241, "netPer1kTokens": 0.8427 }, "rank": 36 }, { "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": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.1615, "known": true, "approximate": false }, "tokens": { "billed": 41491, "known": true, "approximate": false, "prompt": 19150, "completion": 22341, "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": 239.9381, "netPer1kTokens": 0.9339 }, "rank": 37 }, { "modelId": "anthropic/claude-opus-5@medium", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (medium)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.1828, "known": true, "approximate": false }, "tokens": { "billed": 23263, "known": true, "approximate": false, "prompt": 19941, "completion": 3322, "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": 211.9803, "netPer1kTokens": 1.6657 }, "rank": 38 }, { "modelId": "anthropic/claude-sonnet-5@xhigh", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (xhigh)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.1846, "known": true, "approximate": false }, "tokens": { "billed": 34416, "known": true, "approximate": false, "prompt": 19941, "completion": 14475, "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": 209.9133, "netPer1kTokens": 1.1259 }, "rank": 39 }, { "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": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.2192, "known": true, "approximate": false }, "tokens": { "billed": 51145, "known": true, "approximate": false, "prompt": 19150, "completion": 31995, "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": 176.7792, "netPer1kTokens": 0.7576 }, "rank": 40 }, { "modelId": "anthropic/claude-fable-5@low", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (low)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 0, "invalid": 1, "net": 38.75 } }, "cost": { "usd": 0.2352, "known": true, "approximate": false }, "tokens": { "billed": 20751, "known": true, "approximate": false, "prompt": 19941, "completion": 810, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 164.7534, "netPer1kTokens": 1.8674 }, "rank": 41 }, { "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": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.2485, "known": true, "approximate": false }, "tokens": { "billed": 56110, "known": true, "approximate": false, "prompt": 19150, "completion": 36960, "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": 155.9356, "netPer1kTokens": 0.6906 }, "rank": 42 }, { "modelId": "moonshotai/kimi-k3@low", "baseModelId": "moonshotai/kimi-k3", "modelSlug": "kimi-k3", "developerId": "moonshot", "name": "Kimi K3 (low)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.2868, "known": true, "approximate": false }, "tokens": { "billed": 30082, "known": true, "approximate": false, "prompt": 19030, "completion": 11052, "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": 135.1116, "netPer1kTokens": 1.2881 }, "rank": 43 }, { "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": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.3272, "known": true, "approximate": false }, "tokens": { "billed": 81986, "known": true, "approximate": false, "prompt": 12079, "completion": 69907, "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": 118.4291, "netPer1kTokens": 0.4726 }, "rank": 44 }, { "modelId": "anthropic/claude-opus-5@high", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (high)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.3687, "known": true, "approximate": false }, "tokens": { "billed": 30701, "known": true, "approximate": false, "prompt": 19941, "completion": 10760, "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": 105.099, "netPer1kTokens": 1.2622 }, "rank": 45 }, { "modelId": "anthropic/claude-opus-5@max", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (max)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.6575, "known": true, "approximate": false }, "tokens": { "billed": 42252, "known": true, "approximate": false, "prompt": 19941, "completion": 22311, "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": 58.9354, "netPer1kTokens": 0.9171 }, "rank": 46 }, { "modelId": "anthropic/claude-sonnet-5@max", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (max)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.6745, "known": true, "approximate": false }, "tokens": { "billed": 83400, "known": true, "approximate": false, "prompt": 19941, "completion": 63459, "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": 57.45, "netPer1kTokens": 0.4646 }, "rank": 47 }, { "modelId": "moonshotai/kimi-k3@max", "baseModelId": "moonshotai/kimi-k3", "modelSlug": "kimi-k3", "developerId": "moonshot", "name": "Kimi K3 (max)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.9201, "known": true, "approximate": false }, "tokens": { "billed": 59915, "known": true, "approximate": false, "prompt": 19002, "completion": 40913, "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": 42.115, "netPer1kTokens": 0.6467 }, "rank": 48 }, { "modelId": "anthropic/claude-fable-5@xhigh", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (xhigh)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 0, "invalid": 1, "net": 38.75 } }, "cost": { "usd": 1.3294, "known": true, "approximate": false }, "tokens": { "billed": 42643, "known": true, "approximate": false, "prompt": 19941, "completion": 22702, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 29.1485, "netPer1kTokens": 0.9087 }, "rank": 49 }, { "modelId": "anthropic/claude-fable-5@max", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (max)", "score": 216.25, "net": 38.75, "sections": { "SB2": { "correct": 39, "wrong": 0, "invalid": 1, "net": 38.75 } }, "cost": { "usd": 2.1961, "known": true, "approximate": false }, "tokens": { "billed": 59971, "known": true, "approximate": false, "prompt": 19941, "completion": 40030, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 17.6449, "netPer1kTokens": 0.6461 }, "rank": 50 }, { "modelId": "tencent/hy3@none", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (none)", "score": 212.5, "net": 37.5, "sections": { "SB2": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0096, "known": true, "approximate": true }, "tokens": { "billed": 23586, "known": true, "approximate": false, "prompt": 14648, "completion": 8938, "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": 3906.25, "netPer1kTokens": 1.5899 }, "rank": 51 }, { "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": { "SB2": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0447, "known": true, "approximate": false }, "tokens": { "billed": 20426, "known": true, "approximate": false, "prompt": 19941, "completion": 485, "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": 838.9262, "netPer1kTokens": 1.8359 }, "rank": 52 }, { "modelId": "meta/muse-glimmer-30b@high", "baseModelId": "meta/muse-glimmer-30b", "modelSlug": "muse-glimmer-30b", "developerId": "meta", "name": "Muse Glimmer 30B (high)", "score": 212.5, "net": 37.5, "sections": { "SB2": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0527, "known": true, "approximate": false }, "tokens": { "billed": 44781, "known": true, "approximate": false, "prompt": 11544, "completion": 33237, "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": 711.575, "netPer1kTokens": 0.8374 }, "rank": 53 }, { "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": { "SB2": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.08, "known": true, "approximate": false }, "tokens": { "billed": 71574, "known": true, "approximate": false, "prompt": 12079, "completion": 59495, "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": 468.75, "netPer1kTokens": 0.5239 }, "rank": 54 }, { "modelId": "tencent/hy3@high", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (high)", "score": 212.5, "net": 37.5, "sections": { "SB2": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0979, "known": true, "approximate": true }, "tokens": { "billed": 135411, "known": true, "approximate": false, "prompt": 14528, "completion": 120883, "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": 383.0439, "netPer1kTokens": 0.2769 }, "rank": 55 }, { "modelId": "qwen/qwen3.6-27b", "baseModelId": "qwen/qwen3.6-27b", "modelSlug": "qwen3-6-27b", "developerId": "alibaba-qwen", "name": "Qwen3.6 27B", "score": 212.5, "net": 37.5, "sections": { "SB2": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.2645, "known": true, "approximate": true }, "tokens": { "billed": 83525, "known": true, "approximate": false, "prompt": 12079, "completion": 71446, "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": 141.7769, "netPer1kTokens": 0.449 }, "rank": 56 }, { "modelId": "anthropic/claude-fable-5@medium", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (medium)", "score": 212.5, "net": 37.5, "sections": { "SB2": { "correct": 38, "wrong": 1, "invalid": 1, "net": 37.5 } }, "cost": { "usd": 0.3828, "known": true, "approximate": false }, "tokens": { "billed": 23703, "known": true, "approximate": false, "prompt": 19941, "completion": 3762, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 97.9624, "netPer1kTokens": 1.5821 }, "rank": 57 }, { "modelId": "moonshotai/kimi-k3@high", "baseModelId": "moonshotai/kimi-k3", "modelSlug": "kimi-k3", "developerId": "moonshot", "name": "Kimi K3 (high)", "score": 212.5, "net": 37.5, "sections": { "SB2": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.7047, "known": true, "approximate": false }, "tokens": { "billed": 48047, "known": true, "approximate": false, "prompt": 19050, "completion": 28997, "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": 53.2141, "netPer1kTokens": 0.7805 }, "rank": 58 }, { "modelId": "anthropic/claude-fable-5@high", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (high)", "score": 212.5, "net": 37.5, "sections": { "SB2": { "correct": 38, "wrong": 1, "invalid": 1, "net": 37.5 } }, "cost": { "usd": 0.8263, "known": true, "approximate": false }, "tokens": { "billed": 32574, "known": true, "approximate": false, "prompt": 19941, "completion": 12633, "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": 45.383, "netPer1kTokens": 1.1512 }, "rank": 59 }, { "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": 208.75, "net": 36.25, "sections": { "SB2": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0032, "known": true, "approximate": false }, "tokens": { "billed": 15126, "known": true, "approximate": false, "prompt": 11692, "completion": 3434, "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": 11328.125, "netPer1kTokens": 2.3965 }, "rank": 60 }, { "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": { "SB2": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0083, "known": true, "approximate": true }, "tokens": { "billed": 28607, "known": true, "approximate": false, "prompt": 11502, "completion": 17105, "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": 4367.4699, "netPer1kTokens": 1.2672 }, "rank": 61 }, { "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": 208.75, "net": 36.25, "sections": { "SB2": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0197, "known": true, "approximate": false }, "tokens": { "billed": 13027, "known": true, "approximate": false, "prompt": 11692, "completion": 1335, "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": 1840.1015, "netPer1kTokens": 2.7827 }, "rank": 62 }, { "modelId": "thinkingmachines/inkling@none", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (none)", "score": 208.75, "net": 36.25, "sections": { "SB2": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0329, "known": true, "approximate": false }, "tokens": { "billed": 17113, "known": true, "approximate": false, "prompt": 11932, "completion": 5181, "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": 1101.8237, "netPer1kTokens": 2.1183 }, "rank": 63 }, { "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": 208.75, "net": 36.25, "sections": { "SB2": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.1069, "known": true, "approximate": true }, "tokens": { "billed": 39976, "known": true, "approximate": false, "prompt": 13339, "completion": 26637, "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": 339.102, "netPer1kTokens": 0.9068 }, "rank": 64 }, { "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": { "SB2": { "correct": 37, "wrong": 1, "invalid": 2, "net": 36.25 } }, "cost": { "usd": 0.709, "known": true, "approximate": false }, "tokens": { "billed": 208050, "known": true, "approximate": false, "prompt": 12893, "completion": 195157, "orchestration": 0 }, "errors": { "invalid": 2, "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": 51.1283, "netPer1kTokens": 0.1742 }, "rank": 65 }, { "modelId": "deepseek/deepseek-v4-pro@high", "baseModelId": "deepseek/deepseek-v4-pro", "modelSlug": "deepseek-v4-pro", "developerId": "deepseek", "name": "DeepSeek V4 Pro (high)", "score": 205.0, "net": 35.0, "sections": { "SB2": { "correct": 36, "wrong": 2, "invalid": 2, "net": 35.0 } }, "cost": { "usd": 0.1402, "known": true, "approximate": true }, "tokens": { "billed": 49318, "known": true, "approximate": false, "prompt": 15675, "completion": 33643, "orchestration": 0 }, "errors": { "invalid": 2, "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": 249.6434, "netPer1kTokens": 0.7097 }, "rank": 66 }, { "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": 205.0, "net": 35.0, "sections": { "SB2": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.1975, "known": true, "approximate": true }, "tokens": { "billed": 64084, "known": true, "approximate": false, "prompt": 13339, "completion": 50745, "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": "xhigh", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "akashml", "ambient" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 177.2152, "netPer1kTokens": 0.5462 }, "rank": 67 }, { "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": { "SB2": { "correct": 36, "wrong": 3, "invalid": 1, "net": 35.0 } }, "cost": { "usd": 0.3702, "known": true, "approximate": true }, "tokens": { "billed": 118955, "known": true, "approximate": false, "prompt": 18835, "completion": 100120, "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": "xhigh", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 94.5435, "netPer1kTokens": 0.2942 }, "rank": 68 }, { "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": 205.0, "net": 35.0, "sections": { "SB2": { "correct": 36, "wrong": 2, "invalid": 2, "net": 35.0 } }, "cost": { "usd": 0.6775, "known": true, "approximate": false }, "tokens": { "billed": 198983, "known": true, "approximate": false, "prompt": 12949, "completion": 186034, "orchestration": 0 }, "errors": { "invalid": 2, "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": 51.6605, "netPer1kTokens": 0.1759 }, "rank": 69 }, { "modelId": "deepseek/deepseek-v4-flash@high", "baseModelId": "deepseek/deepseek-v4-flash", "modelSlug": "deepseek-v4-flash", "developerId": "deepseek", "name": "DeepSeek V4 Flash (high)", "score": 201.25, "net": 33.75, "sections": { "SB2": { "correct": 35, "wrong": 4, "invalid": 1, "net": 33.75 } }, "cost": { "usd": 0.0095, "known": true, "approximate": false }, "tokens": { "billed": 41849, "known": true, "approximate": false, "prompt": 15675, "completion": 26174, "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": 3552.6316, "netPer1kTokens": 0.8065 }, "rank": 70 }, { "modelId": "deepseek/deepseek-v4-flash@xhigh", "baseModelId": "deepseek/deepseek-v4-flash", "modelSlug": "deepseek-v4-flash", "developerId": "deepseek", "name": "DeepSeek V4 Flash (xhigh)", "score": 201.25, "net": 33.75, "sections": { "SB2": { "correct": 35, "wrong": 3, "invalid": 2, "net": 33.75 } }, "cost": { "usd": 0.0141, "known": true, "approximate": false }, "tokens": { "billed": 59764, "known": true, "approximate": false, "prompt": 18835, "completion": 40929, "orchestration": 0 }, "errors": { "invalid": 2, "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": 2393.617, "netPer1kTokens": 0.5647 }, "rank": 71 }, { "modelId": "meituan/longcat-2.0", "baseModelId": "meituan/longcat-2.0", "modelSlug": "longcat-2-0", "developerId": "meituan", "name": "LongCat 2.0", "score": 201.25, "net": 33.75, "sections": { "SB2": { "correct": 35, "wrong": 3, "invalid": 2, "net": 33.75 } }, "cost": { "usd": 0.0719, "known": true, "approximate": false }, "tokens": { "billed": 209287, "known": true, "approximate": false, "prompt": 15629, "completion": 193658, "orchestration": 0 }, "errors": { "invalid": 2, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 469.4019, "netPer1kTokens": 0.1613 }, "rank": 72 }, { "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": 197.5, "net": 32.5, "sections": { "SB2": { "correct": 34, "wrong": 4, "invalid": 2, "net": 32.5 } }, "cost": { "usd": 0.0228, "known": true, "approximate": false }, "tokens": { "billed": 52492, "known": true, "approximate": false, "prompt": 13615, "completion": 38877, "orchestration": 0 }, "errors": { "invalid": 2, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 1425.4386, "netPer1kTokens": 0.6191 }, "rank": 73 }, { "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": 186.25, "net": 28.75, "sections": { "SB2": { "correct": 31, "wrong": 8, "invalid": 1, "net": 28.75 } }, "cost": { "usd": 0.0241, "known": true, "approximate": false }, "tokens": { "billed": 64065, "known": true, "approximate": false, "prompt": 13720, "completion": 50345, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 1192.9461, "netPer1kTokens": 0.4488 }, "rank": 74 } ], "questionStats": [ { "id": "SB2-1", "number": 1, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-2", "number": 2, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-3", "number": 3, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-4", "number": 4, "section": "SB2", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "SB2-5", "number": 5, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-6", "number": 6, "section": "SB2", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "SB2-7", "number": 7, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-8", "number": 8, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-9", "number": 9, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-10", "number": 10, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-11", "number": 11, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-12", "number": 12, "section": "SB2", "cancelled": false, "correct": 68, "responses": 74 }, { "id": "SB2-13", "number": 13, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-14", "number": 14, "section": "SB2", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "SB2-15", "number": 15, "section": "SB2", "cancelled": false, "correct": 62, "responses": 74 }, { "id": "SB2-16", "number": 16, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-17", "number": 17, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-18", "number": 18, "section": "SB2", "cancelled": false, "correct": 70, "responses": 74 }, { "id": "SB2-19", "number": 19, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-20", "number": 20, "section": "SB2", "cancelled": false, "correct": 71, "responses": 74 }, { "id": "SB2-21", "number": 21, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-22", "number": 22, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-23", "number": 23, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-24", "number": 24, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-25", "number": 25, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-26", "number": 26, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-27", "number": 27, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-28", "number": 28, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-29", "number": 29, "section": "SB2", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "SB2-30", "number": 30, "section": "SB2", "cancelled": false, "correct": 61, "responses": 74 }, { "id": "SB2-31", "number": 31, "section": "SB2", "cancelled": false, "correct": 56, "responses": 74 }, { "id": "SB2-32", "number": 32, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-33", "number": 33, "section": "SB2", "cancelled": false, "correct": 45, "responses": 74 }, { "id": "SB2-34", "number": 34, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-35", "number": 35, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-36", "number": 36, "section": "SB2", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "SB2-37", "number": 37, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-38", "number": 38, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-39", "number": 39, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "SB2-40", "number": 40, "section": "SB2", "cancelled": false, "correct": 74, "responses": 74 } ] }