Recent releases have reshaped Code Arena WebDev leaderboards, the key crowdsourced Elo benchmark for agentic coding tasks involving multi-step reasoning and tool use. OpenAI’s GPT-6 Astra, launched September 3, quickly claimed the top spot at roughly 1,797–1,800 points, edging Anthropic’s Claude Fable 5.1 (around 1,758–1,765) which had led after its early September debut. These gains reflect iterative improvements in frontend workflows, data handling, and iterative debugging over prior models like Claude Opus 5. With three-plus months left in 2026, additional frontier updates from OpenAI, Anthropic, or competitors could extend the climb, though progress depends on sustained gains in real-user voting rather than static benchmarks. Trader sentiment hinges on whether the current ~1,800 ceiling breaks before year-end.
基於Polymarket數據的AI實驗性摘要。這不是交易建議,也不影響該市場的結算方式。 · 更新於$215,292 交易量
1560
22%
1580
19%
1600
9%
$215,292 交易量
1560
22%
1580
19%
1600
9%
Results from the "Score" column under the "Text Arena | Coding" Leaderboard tab at https://arena.ai/leaderboard/text/coding-no-style-control with style control off will be used to resolve this market.
The resolution source for this market is the Chatbot Arena LLM Leaderboard found at arena.ai/leaderboard/text. If this resolution source is unavailable at check time, this market will remain open until the leaderboard comes back online and will resolve based on the first check after it becomes available. If permanently unavailable, this market will resolve to "No".
市場開放時間: Apr 2, 2026, 6:09 PM ET
Results from the "Score" column under the "Text Arena | Coding" Leaderboard tab at https://arena.ai/leaderboard/text/coding-no-style-control with style control off will be used to resolve this market.
The resolution source for this market is the Chatbot Arena LLM Leaderboard found at arena.ai/leaderboard/text. If this resolution source is unavailable at check time, this market will remain open until the leaderboard comes back online and will resolve based on the first check after it becomes available. If permanently unavailable, this market will resolve to "No".
Recent releases have reshaped Code Arena WebDev leaderboards, the key crowdsourced Elo benchmark for agentic coding tasks involving multi-step reasoning and tool use. OpenAI’s GPT-6 Astra, launched September 3, quickly claimed the top spot at roughly 1,797–1,800 points, edging Anthropic’s Claude Fable 5.1 (around 1,758–1,765) which had led after its early September debut. These gains reflect iterative improvements in frontend workflows, data handling, and iterative debugging over prior models like Claude Opus 5. With three-plus months left in 2026, additional frontier updates from OpenAI, Anthropic, or competitors could extend the climb, though progress depends on sustained gains in real-user voting rather than static benchmarks. Trader sentiment hinges on whether the current ~1,800 ceiling breaks before year-end.
基於Polymarket數據的AI實驗性摘要。這不是交易建議,也不影響該市場的結算方式。 · 更新於



警惕外部連結哦。
警惕外部連結哦。
Frequently Asked Questions