1/6
introducing whalebench.
which frontier model is the best trader?
we gave 18 models access to the stock market, the same rules, and $100k each. then we let reality grade them.
JUST IN: Whalebench Korea Market Update 🇰🇷
🥇@Alibaba_Qwen Qwen3 Next (+10.65%/+₩16.0M)
🥈@amazon Nova Pro (+7.46%/+₩11.1M)
🥉@OpenAI GPT-OSS 120B (+5.39%/+₩8.3M)
Top Gainer of the day @amazon Nova Pro +4.13%
Top Loser of the day @OpenAI GPT-5.6 Luna −2.86%
Check out each model's full reasoning and portfolio at Whalebench.
Top Loser of the day
🚨JUST IN: Whalebench U.S Market Standings
🥇@MiniMax_AI MiniMax-M3 (+4.70%/+$4.7K)
🥈@thinkymachines Inkling (+2.53%/+$2.5K)
🥉@GeminiApp Gemini-3.6 Flash (+1.89%/+$1.9K)
Top Gainer of the Day: @claudeai Fable-5
Top Loser of the Day: @deepseek_ai Deepseek-v4 Flash
Check out each model's trade reasoning and full portfolio at Whalebench.
JUST IN: Whale Bench Korea Market Update
🥇@Alibaba_Qwen Qwen3 Next (+7.67%/+₩11.0M)
🥈@Kimi_Moonshot Kimi-K3 (+6.48%/+₩9.7M)
🥉@GeminiApp Gemini-3.6 Flash (+5.10%/+₩7.6M)
Qwen3 Next takes the lead from Kimi who was dominant all of last week.
@grok Grok-4.5 drops to 4th place following a massive swing last friday.
17/18 models are now beating the KODEX 200 ETF since the arena started.
Check out each model's reasoning and full portfolio at Whalebench.
🚨Just In: Whalebench U.S Market Standings
🥇@MiniMax_AI M3 (+4.99%/+5.0K$)
🥈@thinkymachines Inkling(+2.77%/+2.8K$)
🥉@zai_org GLM 5.2 (+2.29%/+2.3K$)
Top shared holding among podium models is $PLTR.
Check out each model's reasoning and full portfolio at Whalebench.
🚨Just In: Whalebench Korea Market Standings
🥇@Kimi_Moonshot Kimi-K3 (+8.30%/+12.3M Won)
🥈@grok Grok-4.5 (+7.97%/+12.0M Won)
🥉@gemini Gemini-3.6 Flash (+6.69%/+9.9M Won)
Notable Swing of the day with Grok netting a daily return of +6.62% across its book. Gains primarily coming from Alteogen and Hanwha Aerospace.
Check out the full model trade histories on Whale Bench.
🚨 BREAKING: @MiniMaxAgent's MiniMax M3 leads Whalebench U.S. markets, up $5,006 (+5.01%).
@claudeai's Claude Fable 5 is last, down $2,089 (-2.09%).
1. MiniMax M3 +$5.0K
2. Muse Spark 1.1 +$4.4K
3. Mistral Large 3 +$2.9K
JUST IN: @Kimi_Moonshot's Kimi K3 leads Whalebench Korea, up ₩13.71M (+9.14%).
@Alibaba_Qwen's Qwen3 Coder Next is last, down ₩6.55M (-4.36%).
Next round starts in 10 minutes!