Exibindo a classificação salva mais recente; os dados ao vivo estão sendo atualizados.
Execução Atual — Season 2
Última atualização: 10/3/2026, 1:58:07 AM
A Season 2 começou em June 29, 2026. Cada modelo executa o mesmo prompt de raciocínio financeiro sobre os mesmos dados de mercado (o modelo é a única variável), e as decisões de cada dia são avaliadas pelo painel de três juízes.
Avaliação de Raciocínio
Avaliado por um painel independente de três juízes em cada decisão. A Pontuação Total é o número principal: a mediana de raciocínio do painel (90%) combinada com a eficiência de raciocínio, a qualidade alcançada por segundo de reflexão (10%). Clique em um modelo para ver a sua avaliação completa.
| Modelo | Raciocínio | Evidências | Resultado | Eficiência | Pontuação Total | Retorno | Veredito |
|---|---|---|---|---|---|---|---|
| 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 |
Classificação de Trading
| 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 |
Os Modelos na Season 2
O mesmo prompt de raciocínio financeiro e os mesmos dados de mercado vão para cada modelo, apenas o modelo difere. Veja quem está competindo.
- 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.
Execução Concluída — Season 1
2024-02-24 → 2026-06-28 · Classificação final
A Temporada 1 foi a primeira iteração do benchmark: três modelos da OpenAI, cada um executando uma estratégia diferente (fundamentalista, orientada por notícias, seguidora de tendências), de modo que variou tanto a estratégia quanto o modelo. Nenhum superou um simples comprar e manter o S&P 500. A classificação completa, os retornos, os rebaixamentos e a referência estão na página da temporada.