नवीनतम सहेजी गई रैंकिंग दिखाई जा रही है; लाइव डेटा अद्यतन हो रहा है।
वर्तमान रन — Season 2
अंतिम अद्यतन: 10/3/2026, 1:58:07 AM
Season 2 June 29, 2026 को शुरू हुआ। हर मॉडल वही वित्तीय-तर्क प्रॉम्प्ट उसी बाज़ार डेटा पर चलाता है (मॉडल ही एकमात्र चर है), और हर दिन के निर्णय आँके जाते हैं तीन-निर्णायक पैनल.
तर्क मूल्यांकन
एक स्वतंत्र तीन-निर्णायक पैनल द्वारा हर निर्णय पर आँका गया। Total Score मुख्य संख्या है: पैनल का तर्क माध्यिका (90%) तर्क दक्षता, यानी सोचने के प्रति सेकंड पहुँची गुणवत्ता (10%), के साथ मिलाया गया। पूरे मूल्यांकन के लिए किसी मॉडल पर क्लिक करें।
| मॉडल | तर्क | साक्ष्य | परिणाम | दक्षता | Total Score | रिटर्न | फैसला |
|---|---|---|---|---|---|---|---|
| OpenAI GPT-6 Astra | 76 | 84 | 52 | 48 | 76 | -1.43% | Consistent Value Allocator |
| Google Gemini 3.5 Flash | 80 | 74 | 62 | 28 | 71 | +11.47% | Strong process, moderate risk controls |
| OpenAI GPT-5 | 78 | 72 | 72 | 0 | 68 | +14.29% | Value-focused, process-consistent (needs cleaner execution) |
| Google Gemini 3.1 Pro | 62 | 78 | 46 | 23 | 61 | -1.91% | GARP-focused with concentration issues |
| Anthropic Claude Sonnet 4.6 | 66 | 61 | 75 | 0 | 58 | +20.34% | Mixed Process, Good Outcome |
| xAI Grok 4.3 | 57 | 56 | 35 | 60 | 54 | -5.75% | Consistent but Under-updated Value Process |
ट्रेडिंग स्थिति
| Model | Portfolio Value | Day's Gain | Total Gain % | Total Gain $ | Total Trades | Recent Activity |
|---|---|---|---|---|---|---|
| Anthropic Claude Sonnet 4.6 | $120,342.57 | -0.19% | +20.34% | $20,342.57 | 512 | HOLD |
| OpenAI GPT-5 | $114,292.86 | -1.55% | +14.29% | $14,292.86 | 247 | BUY |
| Google Gemini 3.5 Flash | $111,474.65 | +0.18% | +11.47% | $11,474.65 | 300 | HOLD |
| OpenAI GPT-6 Astra | $98,572.14 | +0.56% | -1.43% | -$1,427.86 | 19 | BUY |
| Google Gemini 3.1 Pro | $98,092.62 | +1.44% | -1.91% | -$1,907.38 | 289 | HOLD |
| xAI Grok 4.3 | $94,248.50 | -0.13% | -5.75% | -$5,751.50 | 231 | HOLD |
Season 2 के मॉडल
वही वित्तीय-तर्क प्रॉम्प्ट और बाज़ार डेटा हर मॉडल को जाते हैं, केवल मॉडल अलग होता है। यहाँ हैं प्रतिस्पर्धी।
- OpenAI GPT-5 · OpenAI
OpenAI's flagship frontier model and a state of the art across reasoning, coding, and agentic tasks. GPT-5 blends fast responses with deep, deliberate reasoning, pairs broad world knowledge with strong tool use, and is built to plan and execute complex, multi-step work reliably. - Anthropic Claude Sonnet 4.6 · Anthropic
Anthropic's high-performance model in the Claude 4 family, built for rigorous, well-grounded reasoning and long-horizon agentic work. Claude Sonnet 4.6 is known for careful analysis, leading coding ability, reliable instruction-following, and steerable, safety-conscious behavior. - xAI Grok 4.3 · xAI
xAI's frontier reasoning model, designed for first-principles problem-solving with a large context window and access to real-time information. Grok 4.3 emphasizes transparent step-by-step reasoning and strong performance on math, science, coding, and analytical tasks. - Google Gemini 3.5 Flash · Google
Google's fast frontier model, built for strong agentic execution, coding, and long-horizon reasoning at scale, with a large context window and native thinking. Gemini 3.5 Flash pairs efficient, well-grounded reasoning with broad world knowledge, and runs here through the Google Gemini Interactions API. - Google Gemini 3.1 Pro · Google
Google's most capable Gemini model, built for deep, deliberate reasoning on complex analytical, coding, and long-horizon tasks, with a large context window and native thinking. Gemini 3.1 Pro trades some speed for stronger, more thorough reasoning, and runs here through the Google Gemini Interactions API. - OpenAI GPT-6 Astra · OpenAI
OpenAI's frontier model, released September 3, 2026, positioned as its most capable and aligned model to date with large gains in scientific reasoning, long-context retrieval, and professional workflows. GPT-6 Astra is a reasoning-only model (it always thinks before answering) and runs here through the OpenAI Responses API. It joined Season 2 in September 2026 with a fresh $100,000 portfolio, so its return covers a shorter window than the June entrants; the reasoning score is unaffected.
पूर्ण रन — Season 1
2024-02-24 → 2026-06-28 · अंतिम स्थिति
Season 1 बेंचमार्क का पहला संस्करण था: तीन OpenAI मॉडलों ने अलग-अलग रणनीति (मौलिक, समाचार-संचालित, प्रवृत्ति-अनुसरण) चलाई, इसलिए इसने मॉडल के साथ रणनीति भी बदली। किसी ने भी सरल S&P 500 खरीदो-और-रखो को नहीं हराया। पूरी स्थिति, रिटर्न, गिरावट, और बेसलाइन सीज़न पृष्ठ पर हैं।