{ "schemaVersion": 3, "id": "ayt-2026-tde-sb1", "kind": "exam", "title": { "tr": "2026 AYT · Türk Dili ve Edebiyatı - Sosyal Bilimler-1", "en": "2026 AYT · Turkish Language & Literature – Social Sciences-1" }, "subtitle": { "tr": "Alan Yeterlilik Testi · Türk Dili ve Edebiyatı - Sosyal Bilimler-1", "en": "Turkish university entrance exam · Turkish Language & Literature – Social Sciences-1" }, "category": { "tr": "Ulusal sınav", "en": "National exam" }, "description": { "tr": "Yapay zekâ modellerinin 2026 AYT Türk Dili ve Edebiyatı - Sosyal Bilimler-1 testindeki doğruluk, puan, maliyet ve token verimliliğini farklı akıl yürütme (reasoning effort) seviyelerinde karşılaştırır.", "en": "Compares AI models on the 2026 AYT Turkish Language & Literature – Social Sciences-1 test across accuracy, score, cost, and token efficiency at different reasoning effort levels." }, "status": "published", "examDate": "2026-06-21", "runId": "2026-07-31T12-50-01Z", "runCompletedAt": "2026-08-12T04:42:58Z", "generatedAt": "2026-08-12T04:42:58Z", "itemCount": 40, "responseCount": 2960, "totalQuestions": 40, "score": { "label": { "tr": "TDE-SB1 Puanı", "en": "TDE-SB1 Score" }, "min": 100.0, "max": 220.0, "formula": { "tr": "100 + Türk Dili ve Edebiyatı - Sosyal Bilimler-1 neti × 3", "en": "100 + Turkish Language & Literature – Social Sciences-1 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": "TDE-SB1", "name": { "tr": "Türk Dili ve Edebiyatı - Sosyal Bilimler-1", "en": "Turkish Language & Literature – Social Sciences-1" }, "questionCount": 40 } ], "models": [ { "modelId": "thinkingmachines/inkling@high", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (high)", "score": 220.0, "net": 40.0, "sections": { "TDE-SB1": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.1858, "known": true, "approximate": false }, "tokens": { "billed": 54882, "known": true, "approximate": false, "prompt": 11943, "completion": 42939, "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": 215.2853, "netPer1kTokens": 0.7288 }, "rank": 1 }, { "modelId": "anthropic/claude-opus-5@medium", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (medium)", "score": 220.0, "net": 40.0, "sections": { "TDE-SB1": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.2306, "known": true, "approximate": false }, "tokens": { "billed": 25139, "known": true, "approximate": false, "prompt": 19894, "completion": 5245, "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": 173.4605, "netPer1kTokens": 1.5912 }, "rank": 2 }, { "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": { "TDE-SB1": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.2864, "known": true, "approximate": false }, "tokens": { "billed": 19233, "known": true, "approximate": false, "prompt": 11623, "completion": 7610, "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": 139.6648, "netPer1kTokens": 2.0798 }, "rank": 3 }, { "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": { "TDE-SB1": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.373, "known": true, "approximate": false }, "tokens": { "billed": 40073, "known": true, "approximate": false, "prompt": 10788, "completion": 29285, "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": 107.2386, "netPer1kTokens": 0.9982 }, "rank": 4 }, { "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": { "TDE-SB1": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.5818, "known": true, "approximate": false }, "tokens": { "billed": 39188, "known": true, "approximate": false, "prompt": 19894, "completion": 19294, "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": 68.7521, "netPer1kTokens": 1.0207 }, "rank": 5 }, { "modelId": "anthropic/claude-opus-5@max", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (max)", "score": 220.0, "net": 40.0, "sections": { "TDE-SB1": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": 0.8955, "known": true, "approximate": false }, "tokens": { "billed": 51735, "known": true, "approximate": false, "prompt": 19894, "completion": 31841, "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": 44.6678, "netPer1kTokens": 0.7732 }, "rank": 6 }, { "modelId": "openrouter/fusion", "baseModelId": "openrouter/fusion", "modelSlug": "openrouter-fusion", "developerId": "openrouter", "name": "OpenRouter Fusion", "score": 220.0, "net": 40.0, "sections": { "TDE-SB1": { "correct": 40, "wrong": 0, "invalid": 0, "net": 40.0 } }, "cost": { "usd": null, "known": false, "approximate": false }, "tokens": { "billed": 87048, "known": true, "approximate": false, "prompt": 66198, "completion": 20850, "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": null, "netPer1kTokens": 0.4595 }, "rank": 7 }, { "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": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.0349, "known": true, "approximate": false }, "tokens": { "billed": 15500, "known": true, "approximate": false, "prompt": 11623, "completion": 3877, "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": 1110.3152, "netPer1kTokens": 2.5 }, "rank": 8 }, { "modelId": "deepseek/deepseek-v4-pro@xhigh", "baseModelId": "deepseek/deepseek-v4-pro", "modelSlug": "deepseek-v4-pro", "developerId": "deepseek", "name": "DeepSeek V4 Pro (xhigh)", "score": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.0548, "known": true, "approximate": true }, "tokens": { "billed": 72405, "known": true, "approximate": false, "prompt": 18786, "completion": 53619, "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": 707.1168, "netPer1kTokens": 0.5352 }, "rank": 9 }, { "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": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.1311, "known": true, "approximate": false }, "tokens": { "billed": 26109, "known": true, "approximate": false, "prompt": 10788, "completion": 15321, "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": 295.5759, "netPer1kTokens": 1.4842 }, "rank": 10 }, { "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": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.1487, "known": true, "approximate": false }, "tokens": { "billed": 14643, "known": true, "approximate": false, "prompt": 11623, "completion": 3020, "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": 260.5918, "netPer1kTokens": 2.6463 }, "rank": 11 }, { "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": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.1494, "known": true, "approximate": false }, "tokens": { "billed": 21890, "known": true, "approximate": false, "prompt": 19894, "completion": 1996, "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": 259.3708, "netPer1kTokens": 1.7702 }, "rank": 12 }, { "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": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.2074, "known": true, "approximate": false }, "tokens": { "billed": 16599, "known": true, "approximate": false, "prompt": 11623, "completion": 4976, "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": 186.837, "netPer1kTokens": 2.3345 }, "rank": 13 }, { "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": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.2357, "known": true, "approximate": false }, "tokens": { "billed": 17543, "known": true, "approximate": false, "prompt": 11623, "completion": 5920, "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": 164.4039, "netPer1kTokens": 2.2089 }, "rank": 14 }, { "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": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.2521, "known": true, "approximate": false }, "tokens": { "billed": 42239, "known": true, "approximate": false, "prompt": 10788, "completion": 31451, "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": 153.7088, "netPer1kTokens": 0.9174 }, "rank": 15 }, { "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": { "TDE-SB1": { "correct": 39, "wrong": 0, "invalid": 1, "net": 38.75 } }, "cost": { "usd": 0.2528, "known": true, "approximate": false }, "tokens": { "billed": 20971, "known": true, "approximate": false, "prompt": 19894, "completion": 1077, "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": 153.2832, "netPer1kTokens": 1.8478 }, "rank": 16 }, { "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": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.2584, "known": true, "approximate": false }, "tokens": { "billed": 30521, "known": true, "approximate": false, "prompt": 10788, "completion": 19733, "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": 149.9613, "netPer1kTokens": 1.2696 }, "rank": 17 }, { "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": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.2816, "known": true, "approximate": false }, "tokens": { "billed": 19074, "known": true, "approximate": false, "prompt": 11623, "completion": 7451, "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": 137.6065, "netPer1kTokens": 2.0316 }, "rank": 18 }, { "modelId": "thinkingmachines/inkling@max", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (max)", "score": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.2821, "known": true, "approximate": false }, "tokens": { "billed": 78646, "known": true, "approximate": false, "prompt": 11943, "completion": 66703, "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": 137.3626, "netPer1kTokens": 0.4927 }, "rank": 19 }, { "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": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 0.2964, "known": true, "approximate": false }, "tokens": { "billed": 48145, "known": true, "approximate": false, "prompt": 10788, "completion": 37357, "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": 130.7355, "netPer1kTokens": 0.8049 }, "rank": 20 }, { "modelId": "anthropic/claude-fable-5@medium", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (medium)", "score": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 0, "invalid": 1, "net": 38.75 } }, "cost": { "usd": 0.4935, "known": true, "approximate": false }, "tokens": { "billed": 25785, "known": true, "approximate": false, "prompt": 19894, "completion": 5891, "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": 78.5208, "netPer1kTokens": 1.5028 }, "rank": 21 }, { "modelId": "anthropic/claude-fable-5@high", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (high)", "score": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 0, "invalid": 1, "net": 38.75 } }, "cost": { "usd": 0.9146, "known": true, "approximate": false }, "tokens": { "billed": 34207, "known": true, "approximate": false, "prompt": 19894, "completion": 14313, "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": 42.3682, "netPer1kTokens": 1.1328 }, "rank": 22 }, { "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": 216.25, "net": 38.75, "sections": { "TDE-SB1": { "correct": 39, "wrong": 1, "invalid": 0, "net": 38.75 } }, "cost": { "usd": 1.3961, "known": true, "approximate": false }, "tokens": { "billed": 125331, "known": true, "approximate": false, "prompt": 10788, "completion": 114543, "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": 27.7559, "netPer1kTokens": 0.3092 }, "rank": 23 }, { "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": { "TDE-SB1": { "correct": 39, "wrong": 0, "invalid": 1, "net": 38.75 } }, "cost": { "usd": 1.4597, "known": true, "approximate": false }, "tokens": { "billed": 45110, "known": true, "approximate": false, "prompt": 19894, "completion": 25216, "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": 26.5466, "netPer1kTokens": 0.859 }, "rank": 24 }, { "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": { "TDE-SB1": { "correct": 39, "wrong": 0, "invalid": 1, "net": 38.75 } }, "cost": { "usd": 2.6862, "known": true, "approximate": false }, "tokens": { "billed": 69640, "known": true, "approximate": false, "prompt": 19894, "completion": 49746, "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": 14.4256, "netPer1kTokens": 0.5564 }, "rank": 25 }, { "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": 212.5, "net": 37.5, "sections": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0064, "known": true, "approximate": false }, "tokens": { "billed": 20290, "known": true, "approximate": false, "prompt": 11623, "completion": 8667, "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": 5859.375, "netPer1kTokens": 1.8482 }, "rank": 26 }, { "modelId": "deepseek/deepseek-v4-pro@high", "baseModelId": "deepseek/deepseek-v4-pro", "modelSlug": "deepseek-v4-pro", "developerId": "deepseek", "name": "DeepSeek V4 Pro (high)", "score": 212.5, "net": 37.5, "sections": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.039, "known": true, "approximate": true }, "tokens": { "billed": 52598, "known": true, "approximate": false, "prompt": 15626, "completion": 36972, "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": "high", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 961.5385, "netPer1kTokens": 0.713 }, "rank": 27 }, { "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": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0439, "known": true, "approximate": false }, "tokens": { "billed": 17010, "known": true, "approximate": false, "prompt": 11623, "completion": 5387, "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": 854.2141, "netPer1kTokens": 2.2046 }, "rank": 28 }, { "modelId": "tencent/hy3@high", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (high)", "score": 212.5, "net": 37.5, "sections": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0642, "known": true, "approximate": true }, "tokens": { "billed": 132488, "known": true, "approximate": false, "prompt": 14480, "completion": 118008, "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": 584.1121, "netPer1kTokens": 0.283 }, "rank": 29 }, { "modelId": "thinkingmachines/inkling@medium", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (medium)", "score": 212.5, "net": 37.5, "sections": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0662, "known": true, "approximate": false }, "tokens": { "billed": 25333, "known": true, "approximate": false, "prompt": 11943, "completion": 13390, "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": 566.4653, "netPer1kTokens": 1.4803 }, "rank": 30 }, { "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": 212.5, "net": 37.5, "sections": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.0784, "known": true, "approximate": false }, "tokens": { "billed": 19080, "known": true, "approximate": false, "prompt": 10788, "completion": 8292, "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": 478.3163, "netPer1kTokens": 1.9654 }, "rank": 31 }, { "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": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.1332, "known": true, "approximate": false }, "tokens": { "billed": 31887, "known": true, "approximate": false, "prompt": 11623, "completion": 20264, "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": 281.5315, "netPer1kTokens": 1.176 }, "rank": 32 }, { "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": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.3101, "known": true, "approximate": false }, "tokens": { "billed": 64526, "known": true, "approximate": false, "prompt": 19255, "completion": 45271, "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": 120.9287, "netPer1kTokens": 0.5812 }, "rank": 33 }, { "modelId": "qwen/qwen3.7-max", "baseModelId": "qwen/qwen3.7-max", "modelSlug": "qwen3-7-max", "developerId": "alibaba-qwen", "name": "Qwen3.7 Max", "score": 212.5, "net": 37.5, "sections": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.3379, "known": true, "approximate": false }, "tokens": { "billed": 84433, "known": true, "approximate": false, "prompt": 12100, "completion": 72333, "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": 110.9796, "netPer1kTokens": 0.4441 }, "rank": 34 }, { "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": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.4798, "known": true, "approximate": false }, "tokens": { "billed": 25679, "known": true, "approximate": false, "prompt": 11623, "completion": 14056, "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": 78.1576, "netPer1kTokens": 1.4603 }, "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": { "TDE-SB1": { "correct": 38, "wrong": 2, "invalid": 0, "net": 37.5 } }, "cost": { "usd": 0.5405, "known": true, "approximate": false }, "tokens": { "billed": 37535, "known": true, "approximate": false, "prompt": 19894, "completion": 17641, "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": 69.3802, "netPer1kTokens": 0.9991 }, "rank": 36 }, { "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": 208.75, "net": 36.25, "sections": { "TDE-SB1": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0054, "known": true, "approximate": false }, "tokens": { "billed": 18702, "known": true, "approximate": false, "prompt": 11623, "completion": 7079, "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": 6712.963, "netPer1kTokens": 1.9383 }, "rank": 37 }, { "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": 208.75, "net": 36.25, "sections": { "TDE-SB1": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0086, "known": true, "approximate": false }, "tokens": { "billed": 24027, "known": true, "approximate": false, "prompt": 11623, "completion": 12404, "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": 4215.1163, "netPer1kTokens": 1.5087 }, "rank": 38 }, { "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": 208.75, "net": 36.25, "sections": { "TDE-SB1": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0169, "known": true, "approximate": false }, "tokens": { "billed": 37930, "known": true, "approximate": false, "prompt": 11623, "completion": 26307, "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": 2144.9704, "netPer1kTokens": 0.9557 }, "rank": 39 }, { "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": 208.75, "net": 36.25, "sections": { "TDE-SB1": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0176, "known": true, "approximate": false }, "tokens": { "billed": 38943, "known": true, "approximate": false, "prompt": 11623, "completion": 27320, "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": 2059.6591, "netPer1kTokens": 0.9308 }, "rank": 40 }, { "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": 208.75, "net": 36.25, "sections": { "TDE-SB1": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.0487, "known": true, "approximate": false }, "tokens": { "billed": 17799, "known": true, "approximate": false, "prompt": 11623, "completion": 6176, "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": 744.3532, "netPer1kTokens": 2.0366 }, "rank": 41 }, { "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": { "TDE-SB1": { "correct": 37, "wrong": 2, "invalid": 1, "net": 36.25 } }, "cost": { "usd": 0.0495, "known": true, "approximate": false }, "tokens": { "billed": 184542, "known": true, "approximate": false, "prompt": 15626, "completion": 168916, "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": 732.3232, "netPer1kTokens": 0.1964 }, "rank": 42 }, { "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": { "TDE-SB1": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.1159, "known": true, "approximate": false }, "tokens": { "billed": 27510, "known": true, "approximate": false, "prompt": 19894, "completion": 7616, "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": 312.7696, "netPer1kTokens": 1.3177 }, "rank": 43 }, { "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": 208.75, "net": 36.25, "sections": { "TDE-SB1": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.3036, "known": true, "approximate": false }, "tokens": { "billed": 63441, "known": true, "approximate": false, "prompt": 19255, "completion": 44186, "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": 119.4005, "netPer1kTokens": 0.5714 }, "rank": 44 }, { "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": { "TDE-SB1": { "correct": 37, "wrong": 2, "invalid": 1, "net": 36.25 } }, "cost": { "usd": 0.3056, "known": true, "approximate": false }, "tokens": { "billed": 35353, "known": true, "approximate": false, "prompt": 18727, "completion": 16626, "orchestration": 0 }, "errors": { "invalid": 1, "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": 118.6191, "netPer1kTokens": 1.0254 }, "rank": 45 }, { "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": { "TDE-SB1": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.6533, "known": true, "approximate": false }, "tokens": { "billed": 58749, "known": true, "approximate": false, "prompt": 18995, "completion": 39754, "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": 55.4875, "netPer1kTokens": 0.617 }, "rank": 46 }, { "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": { "TDE-SB1": { "correct": 37, "wrong": 3, "invalid": 0, "net": 36.25 } }, "cost": { "usd": 0.755, "known": true, "approximate": false }, "tokens": { "billed": 65484, "known": true, "approximate": false, "prompt": 18936, "completion": 46548, "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": 48.0132, "netPer1kTokens": 0.5536 }, "rank": 47 }, { "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": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.0033, "known": true, "approximate": false }, "tokens": { "billed": 15111, "known": true, "approximate": false, "prompt": 11623, "completion": 3488, "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": 10606.0606, "netPer1kTokens": 2.3162 }, "rank": 48 }, { "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": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.0175, "known": true, "approximate": false }, "tokens": { "billed": 12602, "known": true, "approximate": false, "prompt": 11623, "completion": 979, "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": 2000.0, "netPer1kTokens": 2.7773 }, "rank": 49 }, { "modelId": "tencent/hy3@low", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (low)", "score": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.025, "known": true, "approximate": true }, "tokens": { "billed": 58234, "known": true, "approximate": false, "prompt": 14480, "completion": 43754, "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": 1400.0, "netPer1kTokens": 0.601 }, "rank": 50 }, { "modelId": "thinkingmachines/inkling@low", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (low)", "score": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.0251, "known": true, "approximate": false }, "tokens": { "billed": 15190, "known": true, "approximate": false, "prompt": 11943, "completion": 3247, "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": 1394.4223, "netPer1kTokens": 2.3041 }, "rank": 51 }, { "modelId": "minimax/minimax-m3", "baseModelId": "minimax/minimax-m3", "modelSlug": "minimax-m3", "developerId": "minimax", "name": "MiniMax M3", "score": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.0567, "known": true, "approximate": false }, "tokens": { "billed": 63373, "known": true, "approximate": false, "prompt": 21449, "completion": 41924, "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": 617.284, "netPer1kTokens": 0.5523 }, "rank": 52 }, { "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": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.0787, "known": true, "approximate": false }, "tokens": { "billed": 22808, "known": true, "approximate": false, "prompt": 11623, "completion": 11185, "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": 444.7268, "netPer1kTokens": 1.5345 }, "rank": 53 }, { "modelId": "qwen/qwen3.7-plus", "baseModelId": "qwen/qwen3.7-plus", "modelSlug": "qwen3-7-plus", "developerId": "alibaba-qwen", "name": "Qwen3.7 Plus", "score": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.1335, "known": true, "approximate": false }, "tokens": { "billed": 113402, "known": true, "approximate": false, "prompt": 12100, "completion": 101302, "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": 262.1723, "netPer1kTokens": 0.3086 }, "rank": 54 }, { "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": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 2, "invalid": 2, "net": 35.0 } }, "cost": { "usd": 0.2128, "known": true, "approximate": false }, "tokens": { "billed": 48304, "known": true, "approximate": false, "prompt": 19255, "completion": 29049, "orchestration": 0 }, "errors": { "invalid": 2, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 164.4737, "netPer1kTokens": 0.7246 }, "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": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.2367, "known": true, "approximate": false }, "tokens": { "billed": 39590, "known": true, "approximate": false, "prompt": 19894, "completion": 19696, "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": 147.8665, "netPer1kTokens": 0.8841 }, "rank": 56 }, { "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": { "TDE-SB1": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.6366, "known": true, "approximate": true }, "tokens": { "billed": 189916, "known": true, "approximate": false, "prompt": 13279, "completion": 176637, "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": 54.9796, "netPer1kTokens": 0.1843 }, "rank": 57 }, { "modelId": "anthropic/claude-sonnet-5@max", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (max)", "score": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 4, "invalid": 0, "net": 35.0 } }, "cost": { "usd": 0.6691, "known": true, "approximate": false }, "tokens": { "billed": 82824, "known": true, "approximate": false, "prompt": 19894, "completion": 62930, "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": 52.3091, "netPer1kTokens": 0.4226 }, "rank": 58 }, { "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": 205.0, "net": 35.0, "sections": { "TDE-SB1": { "correct": 36, "wrong": 3, "invalid": 1, "net": 35.0 } }, "cost": { "usd": 0.8879, "known": true, "approximate": true }, "tokens": { "billed": 261300, "known": true, "approximate": false, "prompt": 13279, "completion": 248021, "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": "high", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "akashml", "ambient" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 39.4189, "netPer1kTokens": 0.1339 }, "rank": 59 }, { "modelId": "google/gemma-4-31b-it", "baseModelId": "google/gemma-4-31b-it", "modelSlug": "gemma-4-31b", "developerId": "google", "name": "Gemma 4 31B", "score": 201.25, "net": 33.75, "sections": { "TDE-SB1": { "correct": 35, "wrong": 5, "invalid": 0, "net": 33.75 } }, "cost": { "usd": 0.0069, "known": true, "approximate": true }, "tokens": { "billed": 28440, "known": true, "approximate": false, "prompt": 11511, "completion": 16929, "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": 4891.3043, "netPer1kTokens": 1.1867 }, "rank": 60 }, { "modelId": "thinkingmachines/inkling@minimal", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (minimal)", "score": 201.25, "net": 33.75, "sections": { "TDE-SB1": { "correct": 35, "wrong": 4, "invalid": 1, "net": 33.75 } }, "cost": { "usd": 0.0247, "known": true, "approximate": false }, "tokens": { "billed": 15097, "known": true, "approximate": false, "prompt": 11943, "completion": 3154, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "minimal", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 1366.3968, "netPer1kTokens": 2.2355 }, "rank": 61 }, { "modelId": "thinkingmachines/inkling@none", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (none)", "score": 201.25, "net": 33.75, "sections": { "TDE-SB1": { "correct": 35, "wrong": 5, "invalid": 0, "net": 33.75 } }, "cost": { "usd": 0.0345, "known": true, "approximate": false }, "tokens": { "billed": 17498, "known": true, "approximate": false, "prompt": 11931, "completion": 5567, "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": 978.2609, "netPer1kTokens": 1.9288 }, "rank": 62 }, { "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": { "TDE-SB1": { "correct": 35, "wrong": 4, "invalid": 1, "net": 33.75 } }, "cost": { "usd": 0.1718, "known": true, "approximate": false }, "tokens": { "billed": 203953, "known": true, "approximate": false, "prompt": 12972, "completion": 190981, "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": 196.4494, "netPer1kTokens": 0.1655 }, "rank": 63 }, { "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": 201.25, "net": 33.75, "sections": { "TDE-SB1": { "correct": 35, "wrong": 4, "invalid": 1, "net": 33.75 } }, "cost": { "usd": 0.3641, "known": true, "approximate": false }, "tokens": { "billed": 111882, "known": true, "approximate": false, "prompt": 12878, "completion": 99004, "orchestration": 0 }, "errors": { "invalid": 1, "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": 92.6943, "netPer1kTokens": 0.3017 }, "rank": 64 }, { "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": 201.25, "net": 33.75, "sections": { "TDE-SB1": { "correct": 35, "wrong": 5, "invalid": 0, "net": 33.75 } }, "cost": { "usd": 0.3927, "known": true, "approximate": false }, "tokens": { "billed": 119825, "known": true, "approximate": false, "prompt": 12878, "completion": 106947, "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": 85.9435, "netPer1kTokens": 0.2817 }, "rank": 65 }, { "modelId": "tencent/hy3@none", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (none)", "score": 197.5, "net": 32.5, "sections": { "TDE-SB1": { "correct": 34, "wrong": 6, "invalid": 0, "net": 32.5 } }, "cost": { "usd": 0.007, "known": true, "approximate": true }, "tokens": { "billed": 24127, "known": true, "approximate": false, "prompt": 14600, "completion": 9527, "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": 4642.8571, "netPer1kTokens": 1.347 }, "rank": 66 }, { "modelId": "deepseek/deepseek-v4-flash@xhigh", "baseModelId": "deepseek/deepseek-v4-flash", "modelSlug": "deepseek-v4-flash", "developerId": "deepseek", "name": "DeepSeek V4 Flash (xhigh)", "score": 197.5, "net": 32.5, "sections": { "TDE-SB1": { "correct": 34, "wrong": 1, "invalid": 5, "net": 32.5 } }, "cost": { "usd": 0.1998, "known": true, "approximate": false }, "tokens": { "billed": 723130, "known": true, "approximate": false, "prompt": 18786, "completion": 704344, "orchestration": 0 }, "errors": { "invalid": 5, "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": 162.6627, "netPer1kTokens": 0.0449 }, "rank": 67 }, { "modelId": "meta/muse-glimmer-30b@high", "baseModelId": "meta/muse-glimmer-30b", "modelSlug": "muse-glimmer-30b", "developerId": "meta", "name": "Muse Glimmer 30B (high)", "score": 193.75, "net": 31.25, "sections": { "TDE-SB1": { "correct": 33, "wrong": 7, "invalid": 0, "net": 31.25 } }, "cost": { "usd": 0.0733, "known": true, "approximate": false }, "tokens": { "billed": 58523, "known": true, "approximate": false, "prompt": 11583, "completion": 46940, "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": 426.3302, "netPer1kTokens": 0.534 }, "rank": 68 }, { "modelId": "anthropic/claude-sonnet-5@low", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (low)", "score": 190.0, "net": 30.0, "sections": { "TDE-SB1": { "correct": 32, "wrong": 8, "invalid": 0, "net": 30.0 } }, "cost": { "usd": 0.0458, "known": true, "approximate": false }, "tokens": { "billed": 20492, "known": true, "approximate": false, "prompt": 19894, "completion": 598, "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": 655.0218, "netPer1kTokens": 1.464 }, "rank": 69 }, { "modelId": "anthropic/claude-sonnet-5@medium", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (medium)", "score": 190.0, "net": 30.0, "sections": { "TDE-SB1": { "correct": 32, "wrong": 8, "invalid": 0, "net": 30.0 } }, "cost": { "usd": 0.0613, "known": true, "approximate": false }, "tokens": { "billed": 22047, "known": true, "approximate": false, "prompt": 19894, "completion": 2153, "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": 489.3964, "netPer1kTokens": 1.3607 }, "rank": 70 }, { "modelId": "qwen/qwen3.6-27b", "baseModelId": "qwen/qwen3.6-27b", "modelSlug": "qwen3-6-27b", "developerId": "alibaba-qwen", "name": "Qwen3.6 27B", "score": 190.0, "net": 30.0, "sections": { "TDE-SB1": { "correct": 32, "wrong": 8, "invalid": 0, "net": 30.0 } }, "cost": { "usd": 0.1967, "known": true, "approximate": true }, "tokens": { "billed": 108630, "known": true, "approximate": false, "prompt": 12100, "completion": 96530, "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": 152.5165, "netPer1kTokens": 0.2762 }, "rank": 71 }, { "modelId": "meituan/longcat-2.0", "baseModelId": "meituan/longcat-2.0", "modelSlug": "longcat-2-0", "developerId": "meituan", "name": "LongCat 2.0", "score": 186.25, "net": 28.75, "sections": { "TDE-SB1": { "correct": 31, "wrong": 6, "invalid": 3, "net": 28.75 } }, "cost": { "usd": 0.5142, "known": true, "approximate": false }, "tokens": { "billed": 440056, "known": true, "approximate": false, "prompt": 15381, "completion": 424675, "orchestration": 0 }, "errors": { "invalid": 3, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 55.9121, "netPer1kTokens": 0.0653 }, "rank": 72 }, { "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": 182.5, "net": 27.5, "sections": { "TDE-SB1": { "correct": 30, "wrong": 7, "invalid": 3, "net": 27.5 } }, "cost": { "usd": 0.1087, "known": true, "approximate": false }, "tokens": { "billed": 185984, "known": true, "approximate": false, "prompt": 13686, "completion": 172298, "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": 252.9899, "netPer1kTokens": 0.1479 }, "rank": 73 }, { "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": 178.75, "net": 26.25, "sections": { "TDE-SB1": { "correct": 29, "wrong": 9, "invalid": 2, "net": 26.25 } }, "cost": { "usd": 0.0777, "known": true, "approximate": false }, "tokens": { "billed": 136358, "known": true, "approximate": false, "prompt": 13721, "completion": 122637, "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": 337.8378, "netPer1kTokens": 0.1925 }, "rank": 74 } ], "questionStats": [ { "id": "TDE-SB1-1", "number": 1, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-2", "number": 2, "section": "TDE-SB1", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "TDE-SB1-3", "number": 3, "section": "TDE-SB1", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "TDE-SB1-4", "number": 4, "section": "TDE-SB1", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "TDE-SB1-5", "number": 5, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-6", "number": 6, "section": "TDE-SB1", "cancelled": false, "correct": 25, "responses": 74 }, { "id": "TDE-SB1-7", "number": 7, "section": "TDE-SB1", "cancelled": false, "correct": 71, "responses": 74 }, { "id": "TDE-SB1-8", "number": 8, "section": "TDE-SB1", "cancelled": false, "correct": 65, "responses": 74 }, { "id": "TDE-SB1-9", "number": 9, "section": "TDE-SB1", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "TDE-SB1-10", "number": 10, "section": "TDE-SB1", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "TDE-SB1-11", "number": 11, "section": "TDE-SB1", "cancelled": false, "correct": 71, "responses": 74 }, { "id": "TDE-SB1-12", "number": 12, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-13", "number": 13, "section": "TDE-SB1", "cancelled": false, "correct": 59, "responses": 74 }, { "id": "TDE-SB1-14", "number": 14, "section": "TDE-SB1", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "TDE-SB1-15", "number": 15, "section": "TDE-SB1", "cancelled": false, "correct": 61, "responses": 74 }, { "id": "TDE-SB1-16", "number": 16, "section": "TDE-SB1", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "TDE-SB1-17", "number": 17, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-18", "number": 18, "section": "TDE-SB1", "cancelled": false, "correct": 67, "responses": 74 }, { "id": "TDE-SB1-19", "number": 19, "section": "TDE-SB1", "cancelled": false, "correct": 52, "responses": 74 }, { "id": "TDE-SB1-20", "number": 20, "section": "TDE-SB1", "cancelled": false, "correct": 49, "responses": 74 }, { "id": "TDE-SB1-21", "number": 21, "section": "TDE-SB1", "cancelled": false, "correct": 71, "responses": 74 }, { "id": "TDE-SB1-22", "number": 22, "section": "TDE-SB1", "cancelled": false, "correct": 70, "responses": 74 }, { "id": "TDE-SB1-23", "number": 23, "section": "TDE-SB1", "cancelled": false, "correct": 42, "responses": 74 }, { "id": "TDE-SB1-24", "number": 24, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-25", "number": 25, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-26", "number": 26, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-27", "number": 27, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-28", "number": 28, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-29", "number": 29, "section": "TDE-SB1", "cancelled": false, "correct": 70, "responses": 74 }, { "id": "TDE-SB1-30", "number": 30, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-31", "number": 31, "section": "TDE-SB1", "cancelled": false, "correct": 68, "responses": 74 }, { "id": "TDE-SB1-32", "number": 32, "section": "TDE-SB1", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "TDE-SB1-33", "number": 33, "section": "TDE-SB1", "cancelled": false, "correct": 71, "responses": 74 }, { "id": "TDE-SB1-34", "number": 34, "section": "TDE-SB1", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "TDE-SB1-35", "number": 35, "section": "TDE-SB1", "cancelled": false, "correct": 69, "responses": 74 }, { "id": "TDE-SB1-36", "number": 36, "section": "TDE-SB1", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "TDE-SB1-37", "number": 37, "section": "TDE-SB1", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "TDE-SB1-38", "number": 38, "section": "TDE-SB1", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "TDE-SB1-39", "number": 39, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "TDE-SB1-40", "number": 40, "section": "TDE-SB1", "cancelled": false, "correct": 74, "responses": 74 } ] }