Recent LiveBench mathematics results reflect a tight race among frontier large language models, with market odds clustering near 45-50% for leading contenders as traders weigh incremental gains from reasoning-focused releases. OpenAI's GPT-5 series and o-series variants currently post the strongest scores on the contamination-resistant math tasks, driven by scaled test-time compute and targeted fine-tuning, while Anthropic's Claude 5 Opus and related reasoning modes trail but remain competitive on proof-style and competition problems. Chinese labs like DeepSeek and Moonshot trail further due to narrower gaps in advanced contest math, though rapid iteration keeps them in play. With resolution at month-end, any new model drops, efficiency updates, or benchmark refreshes in August could shift the narrow margins, underscoring how live evaluation rewards timely capability jumps over static leaderboards.
Résumé expérimental généré par IA à partir des données Polymarket. Ceci n'est pas un conseil de trading et ne joue aucun rôle dans la résolution de ce marché. · Mis à jourOpenAI 45%
Anthropic 38%
Alibaba 28%
DeepSeek 22%
NOUVEAU
NOUVEAU
31 août 2026
OpenAI
43%
Anthropic
34%
Alibaba
28%
DeepSeek
22%
Xiaomi
22%
SpaceXAI
22%
22%
Meta
21%
ByteDance
21%
Moonshot
20%
Baidu
9%
Z.ai
9%
Thinky
6%
Nvidia
7%
Mistral
3%
Tencent
3%
Amazon
3%
Meituan
1%
StepFun
1%
Microsoft
1%
MiniMax
-
OpenAI 45%
Anthropic 38%
Alibaba 28%
DeepSeek 22%
NOUVEAU
NOUVEAU
31 août 2026
OpenAI
$128 Vol.
43%
Anthropic
$106 Vol.
34%
Alibaba
$67 Vol.
28%
DeepSeek
$67 Vol.
22%
Xiaomi
$67 Vol.
22%
SpaceXAI
$67 Vol.
22%
$67 Vol.
22%
Meta
$67 Vol.
21%
ByteDance
$89 Vol.
21%
Moonshot
$67 Vol.
20%
Baidu
$67 Vol.
9%
Z.ai
$67 Vol.
9%
Thinky
$138 Vol.
6%
Nvidia
$138 Vol.
7%
Mistral
$75 Vol.
3%
Tencent
$75 Vol.
3%
Amazon
$81 Vol.
3%
Meituan
$83 Vol.
1%
StepFun
$85 Vol.
1%
Microsoft
$92 Vol.
1%
MiniMax
$92 Vol.
-
This market will resolve according to the company which owns the model with the highest Mathematics score on LiveBench.ai when the leaderboard is checked on August 31, 2026, at 12:00 PM ET.
Results from the “Mathematics” column of the leaderboard at https://livebench.ai/#/?cats=Mathematics, with the latest available LiveBench release selected and the category set to “Mathematics,” will be used to resolve this market.
Models will be ranked according to the specified score, with higher scores ranked ahead of lower scores. If two or more models have exactly the same score as displayed on the leaderboard, the model with the lower listed "cost per successful task" will be ranked ahead. If a tie still remains, alphabetical order of company names as listed in this market group will be used as a final tiebreaker (e.g., if the two models are tied by exact score and cost per successful task, “Google” would be ranked ahead of “SpaceXAI”). This market will resolve based on the company that occupies first place under this ranking.
The resolution source for this market is the LiveBench leaderboard. 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 it becomes permanently unavailable, this market will resolve to “Other.”Recent LiveBench mathematics results reflect a tight race among frontier large language models, with market odds clustering near 45-50% for leading contenders as traders weigh incremental gains from reasoning-focused releases. OpenAI's GPT-5 series and o-series variants currently post the strongest scores on the contamination-resistant math tasks, driven by scaled test-time compute and targeted fine-tuning, while Anthropic's Claude 5 Opus and related reasoning modes trail but remain competitive on proof-style and competition problems. Chinese labs like DeepSeek and Moonshot trail further due to narrower gaps in advanced contest math, though rapid iteration keeps them in play. With resolution at month-end, any new model drops, efficiency updates, or benchmark refreshes in August could shift the narrow margins, underscoring how live evaluation rewards timely capability jumps over static leaderboards.
This market will resolve according to the company which owns the model with the highest Mathematics score on LiveBench.ai when the leaderboard is checked on August 31, 2026, at 12:00 PM ET.
Results from the “Mathematics” column of the leaderboard at https://livebench.ai/#/?cats=Mathematics, with the latest available LiveBench release selected and the category set to “Mathematics,” will be used to resolve this market.
Models will be ranked according to the specified score, with higher scores ranked ahead of lower scores. If two or more models have exactly the same score as displayed on the leaderboard, the model with the lower listed "cost per successful task" will be ranked ahead. If a tie still remains, alphabetical order of company names as listed in this market group will be used as a final tiebreaker (e.g., if the two models are tied by exact score and cost per successful task, “Google” would be ranked ahead of “SpaceXAI”). This market will resolve based on the company that occupies first place under this ranking.
The resolution source for this market is the LiveBench leaderboard. 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 it becomes permanently unavailable, this market will resolve to “Other.”
Results from the “Mathematics” column of the leaderboard at https://livebench.ai/#/?cats=Mathematics, with the latest available LiveBench release selected and the category set to “Mathematics,” will be used to resolve this market.
Models will be ranked according to the specified score, with higher scores ranked ahead of lower scores. If two or more models have exactly the same score as displayed on the leaderboard, the model with the lower listed "cost per successful task" will be ranked ahead. If a tie still remains, alphabetical order of company names as listed in this market group will be used as a final tiebreaker (e.g., if the two models are tied by exact score and cost per successful task, “Google” would be ranked ahead of “SpaceXAI”). This market will resolve based on the company that occupies first place under this ranking.
The resolution source for this market is the LiveBench leaderboard. 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 it becomes permanently unavailable, this market will resolve to “Other.”
Marché ouvert : Jul 29, 2026, 6:27 PM ET
Volume
$1,785Date de fin
31 août 2026Marché ouvert
Jul 29, 2026, 6:27 PM ETSource de résolution
https://livebench.ai/#/?cats=MathematicsResolver
0x69c47De9D...This market will resolve according to the company which owns the model with the highest Mathematics score on LiveBench.ai when the leaderboard is checked on August 31, 2026, at 12:00 PM ET.
Results from the “Mathematics” column of the leaderboard at https://livebench.ai/#/?cats=Mathematics, with the latest available LiveBench release selected and the category set to “Mathematics,” will be used to resolve this market.
Models will be ranked according to the specified score, with higher scores ranked ahead of lower scores. If two or more models have exactly the same score as displayed on the leaderboard, the model with the lower listed "cost per successful task" will be ranked ahead. If a tie still remains, alphabetical order of company names as listed in this market group will be used as a final tiebreaker (e.g., if the two models are tied by exact score and cost per successful task, “Google” would be ranked ahead of “SpaceXAI”). This market will resolve based on the company that occupies first place under this ranking.
The resolution source for this market is the LiveBench leaderboard. 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 it becomes permanently unavailable, this market will resolve to “Other.”Recent LiveBench mathematics results reflect a tight race among frontier large language models, with market odds clustering near 45-50% for leading contenders as traders weigh incremental gains from reasoning-focused releases. OpenAI's GPT-5 series and o-series variants currently post the strongest scores on the contamination-resistant math tasks, driven by scaled test-time compute and targeted fine-tuning, while Anthropic's Claude 5 Opus and related reasoning modes trail but remain competitive on proof-style and competition problems. Chinese labs like DeepSeek and Moonshot trail further due to narrower gaps in advanced contest math, though rapid iteration keeps them in play. With resolution at month-end, any new model drops, efficiency updates, or benchmark refreshes in August could shift the narrow margins, underscoring how live evaluation rewards timely capability jumps over static leaderboards.
This market will resolve according to the company which owns the model with the highest Mathematics score on LiveBench.ai when the leaderboard is checked on August 31, 2026, at 12:00 PM ET.
Results from the “Mathematics” column of the leaderboard at https://livebench.ai/#/?cats=Mathematics, with the latest available LiveBench release selected and the category set to “Mathematics,” will be used to resolve this market.
Models will be ranked according to the specified score, with higher scores ranked ahead of lower scores. If two or more models have exactly the same score as displayed on the leaderboard, the model with the lower listed "cost per successful task" will be ranked ahead. If a tie still remains, alphabetical order of company names as listed in this market group will be used as a final tiebreaker (e.g., if the two models are tied by exact score and cost per successful task, “Google” would be ranked ahead of “SpaceXAI”). This market will resolve based on the company that occupies first place under this ranking.
The resolution source for this market is the LiveBench leaderboard. 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 it becomes permanently unavailable, this market will resolve to “Other.”
Results from the “Mathematics” column of the leaderboard at https://livebench.ai/#/?cats=Mathematics, with the latest available LiveBench release selected and the category set to “Mathematics,” will be used to resolve this market.
Models will be ranked according to the specified score, with higher scores ranked ahead of lower scores. If two or more models have exactly the same score as displayed on the leaderboard, the model with the lower listed "cost per successful task" will be ranked ahead. If a tie still remains, alphabetical order of company names as listed in this market group will be used as a final tiebreaker (e.g., if the two models are tied by exact score and cost per successful task, “Google” would be ranked ahead of “SpaceXAI”). This market will resolve based on the company that occupies first place under this ranking.
The resolution source for this market is the LiveBench leaderboard. 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 it becomes permanently unavailable, this market will resolve to “Other.”
Volume
$1,785Date de fin
31 août 2026Marché ouvert
Jul 29, 2026, 6:27 PM ETSource de résolution
https://livebench.ai/#/?cats=MathematicsResolver
0x69c47De9D...Recent LiveBench mathematics results reflect a tight race among frontier large language models, with market odds clustering near 45-50% for leading contenders as traders weigh incremental gains from reasoning-focused releases. OpenAI's GPT-5 series and o-series variants currently post the strongest scores on the contamination-resistant math tasks, driven by scaled test-time compute and targeted fine-tuning, while Anthropic's Claude 5 Opus and related reasoning modes trail but remain competitive on proof-style and competition problems. Chinese labs like DeepSeek and Moonshot trail further due to narrower gaps in advanced contest math, though rapid iteration keeps them in play. With resolution at month-end, any new model drops, efficiency updates, or benchmark refreshes in August could shift the narrow margins, underscoring how live evaluation rewards timely capability jumps over static leaderboards.
Résumé expérimental généré par IA à partir des données Polymarket. Ceci n'est pas un conseil de trading et ne joue aucun rôle dans la résolution de ce marché. · Mis à jour
Méfiez-vous des liens externes.
Méfiez-vous des liens externes.
Questions fréquentes