**Anthropic holds a commanding 94.5% implied probability on Polymarket for the best overall LiveBench model at end of August 2026.** Traders view its recent Claude releases—particularly variants like Fable 5 and Opus 4.x/5 series—as delivering the strongest contamination-resistant performance across LiveBench’s math, coding, reasoning, data analysis, instruction following, and language categories. These models have topped or closely led multiple snapshots of the dynamic benchmark through demonstrated gains in agentic and reasoning tasks, outpacing OpenAI’s o-series and GPT-5 lines as well as offerings from Google, Moonshot, and others. With the month nearly over and no major competing releases reported in the final days, the market reflects high confidence in Anthropic’s current edge. A late August model drop or benchmark refresh could still shift the outcome, though such moves appear unlikely given typical development cycles.
Experimental AI-generated summary referencing Polymarket data. This is not trading advice and plays no role in how this market resolves. · UpdatedAnthropic 94.5%
OpenAI 2.7%
SpaceXAI 1.4%
Meta 1.3%
$23,658 Vol.
$23,658 Vol.
Aug 31, 2026
Anthropic
95%
OpenAI
3%
SpaceXAI
1%
Meta
1%
StepFun
1%
DeepSeek
1%
Z.ai
1%
Nvidia
1%
Baidu
1%
Amazon
1%
Moonshot
<1%
Tencent
<1%
Alibaba
<1%
Microsoft
<1%
<1%
ByteDance
<1%
MiniMax
<1%
Mistral
<1%
Meituan
<1%
Xiaomi
<1%
Thinky
<1%
Anthropic 94.5%
OpenAI 2.7%
SpaceXAI 1.4%
Meta 1.3%
$23,658 Vol.
$23,658 Vol.
Aug 31, 2026
Anthropic
$9,602 Vol.
95%
OpenAI
$1,992 Vol.
3%
SpaceXAI
$649 Vol.
1%
Meta
$633 Vol.
1%
StepFun
$629 Vol.
1%
DeepSeek
$748 Vol.
1%
Z.ai
$654 Vol.
1%
Nvidia
$576 Vol.
1%
Baidu
$625 Vol.
1%
Amazon
$599 Vol.
1%
Moonshot
$544 Vol.
<1%
Tencent
$630 Vol.
<1%
Alibaba
$678 Vol.
<1%
Microsoft
$480 Vol.
<1%
$631 Vol.
<1%
ByteDance
$716 Vol.
<1%
MiniMax
$777 Vol.
<1%
Mistral
$704 Vol.
<1%
Meituan
$597 Vol.
<1%
Xiaomi
$604 Vol.
<1%
Thinky
$589 Vol.
<1%
This market will resolve according to the company which owns the model with the highest Overall score on LiveBench.ai when the leaderboard is checked on August 31, 2026, at 12:00 PM ET.
Results from the “Overall” column of the leaderboard at https://livebench.ai/, with the latest available LiveBench release selected and the category set to “All,” 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.”**Anthropic holds a commanding 94.5% implied probability on Polymarket for the best overall LiveBench model at end of August 2026.** Traders view its recent Claude releases—particularly variants like Fable 5 and Opus 4.x/5 series—as delivering the strongest contamination-resistant performance across LiveBench’s math, coding, reasoning, data analysis, instruction following, and language categories. These models have topped or closely led multiple snapshots of the dynamic benchmark through demonstrated gains in agentic and reasoning tasks, outpacing OpenAI’s o-series and GPT-5 lines as well as offerings from Google, Moonshot, and others. With the month nearly over and no major competing releases reported in the final days, the market reflects high confidence in Anthropic’s current edge. A late August model drop or benchmark refresh could still shift the outcome, though such moves appear unlikely given typical development cycles.
This market will resolve according to the company which owns the model with the highest Overall score on LiveBench.ai when the leaderboard is checked on August 31, 2026, at 12:00 PM ET.
Results from the “Overall” column of the leaderboard at https://livebench.ai/, with the latest available LiveBench release selected and the category set to “All,” 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 “Overall” column of the leaderboard at https://livebench.ai/, with the latest available LiveBench release selected and the category set to “All,” 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.”
Market Opened: Jul 29, 2026, 6:17 PM ET
Volume
$23,658End Date
Aug 31, 2026Market Opened
Jul 29, 2026, 6:17 PM ETResolution Source
https://livebench.ai/Resolver
0x69c47De9D...This market will resolve according to the company which owns the model with the highest Overall score on LiveBench.ai when the leaderboard is checked on August 31, 2026, at 12:00 PM ET.
Results from the “Overall” column of the leaderboard at https://livebench.ai/, with the latest available LiveBench release selected and the category set to “All,” 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.”**Anthropic holds a commanding 94.5% implied probability on Polymarket for the best overall LiveBench model at end of August 2026.** Traders view its recent Claude releases—particularly variants like Fable 5 and Opus 4.x/5 series—as delivering the strongest contamination-resistant performance across LiveBench’s math, coding, reasoning, data analysis, instruction following, and language categories. These models have topped or closely led multiple snapshots of the dynamic benchmark through demonstrated gains in agentic and reasoning tasks, outpacing OpenAI’s o-series and GPT-5 lines as well as offerings from Google, Moonshot, and others. With the month nearly over and no major competing releases reported in the final days, the market reflects high confidence in Anthropic’s current edge. A late August model drop or benchmark refresh could still shift the outcome, though such moves appear unlikely given typical development cycles.
This market will resolve according to the company which owns the model with the highest Overall score on LiveBench.ai when the leaderboard is checked on August 31, 2026, at 12:00 PM ET.
Results from the “Overall” column of the leaderboard at https://livebench.ai/, with the latest available LiveBench release selected and the category set to “All,” 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 “Overall” column of the leaderboard at https://livebench.ai/, with the latest available LiveBench release selected and the category set to “All,” 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
$23,658End Date
Aug 31, 2026Market Opened
Jul 29, 2026, 6:17 PM ETResolution Source
https://livebench.ai/Resolver
0x69c47De9D...**Anthropic holds a commanding 94.5% implied probability on Polymarket for the best overall LiveBench model at end of August 2026.** Traders view its recent Claude releases—particularly variants like Fable 5 and Opus 4.x/5 series—as delivering the strongest contamination-resistant performance across LiveBench’s math, coding, reasoning, data analysis, instruction following, and language categories. These models have topped or closely led multiple snapshots of the dynamic benchmark through demonstrated gains in agentic and reasoning tasks, outpacing OpenAI’s o-series and GPT-5 lines as well as offerings from Google, Moonshot, and others. With the month nearly over and no major competing releases reported in the final days, the market reflects high confidence in Anthropic’s current edge. A late August model drop or benchmark refresh could still shift the outcome, though such moves appear unlikely given typical development cycles.
Experimental AI-generated summary referencing Polymarket data. This is not trading advice and plays no role in how this market resolves. · Updated
Beware of external links.
Beware of external links.
Frequently Asked Questions