Anthropic Wins Second Best AI Model Market: June 2026 | Lines.com

Anthropic claimed the second spot on the Chatbot Arena LLM Leaderboard as of June 30, 2026, resolving the Polymarket Which company has second best AI model end of June? market in its favor. The outcome confirmed what most traders had priced in for weeks: Anthropic’s model lineup ran so deep that the company occupied both the top and second positions on the benchmark that governed resolution. No other lab came close by the close date.

The market opened at an 88% implied probability for Anthropic and never seriously threatened to flip. Traders pushed the contract to 100% at resolution, with $700,980 in total volume reflecting strong but not outsized conviction. The market correctly favored Anthropic from the start. The only real question during its lifespan was whether OpenAI’s GPT-5.5 would pull ahead of a second Anthropic model for the runner-up position, and that question was answered decisively by mid-June.

How Anthropic Secured Both Top AI Model Slots in June 2026

Anthropic’s Claude Fable 5, released June 9, 2026, captured the top position on the Chatbot Arena LLM Leaderboard. That left Claude Opus 4.8, released in May 2026, holding the second position on the same composite leaderboard. The Polymarket market resolved using Chatbot Arena rankings as of June 30, 2026 at 12:00 PM ET, and Anthropic’s two-model grip on the top of that board triggered simultaneous resolution in both the best and second best markets. A single lab sweeping both positions had not been a common expectation when this market opened.

The Chatbot Arena composite incorporates multiple benchmark dimensions including agentic coding performance on SWE-Bench, the Artificial Analysis Intelligence Index, and long-horizon reasoning evaluations. Claude Fable 5 posted an 80.3% score on SWE-Bench Pro to lead that dimension, while Claude Opus 4.8 scored 69.2% on the same benchmark, keeping it ahead of GPT-5.5 and Gemini 3.1 Pro. The Opus 4.8 result on the composite Artificial Analysis Intelligence Index reached 61.4, edging out OpenAI’s best offering on the aggregated score. The market’s final convergence to 100% reflected a clean leaderboard picture, not a contested finish.

The contract price jumped roughly 7% around June 15, six days after Claude Fable 5 launched. That single-session move signaled that traders recalibrated the second-place question once Fable 5 locked down the top spot. With Anthropic clearly at number one, Opus 4.8 only needed to hold off OpenAI and Google for the silver position, which the benchmark data confirmed it did.

Sponsored Partner

How the Market Performed on Anthropic

The market opened pricing Anthropic at an 88% implied probability, reflecting genuine but incomplete certainty. At that point, traders acknowledged OpenAI’s GPT-5.5 and Google’s Gemini 3.1 Pro as credible alternatives for the second slot. The market moved to 100% at close, a shift of 12 percentage points driven primarily by the Fable 5 release and subsequent leaderboard updates. Under the Lines market accuracy framework, an 88% opening probability that resolved YES represents a correctly favored outcome with strong directional accuracy.

Total volume reached $700,980 across the market’s lifespan, with $4,828,253 in liquidity providing tight price discovery. The 24-hour volume at resolution was just $2,203, consistent with a market that had already converged. The high liquidity relative to volume suggests institutional-scale capital was present to anchor the price, even as late-stage trading activity thinned out.

• Resolution Outcome: Anthropic
• Article-Time Probability: 88%
• Final Price at Close: 100%
• Total Volume: $700,980
• Market Assessment: Correctly priced, strong directional conviction from open

What Anthropic’s Sweep Means for the AI Competitive Landscape

A single company holding both top leaderboard positions at the same time is a meaningful signal in a field where OpenAI, Google, xAI, and Meta all compete aggressively on six-week release cycles. Anthropic’s June 2026 sweep happened because Claude Fable 5 raised the ceiling fast enough that Opus 4.8, a model already capable of beating the prior generation’s best, got pushed into second place within the same house. That kind of intra-company competition at the frontier is a new dynamic and it complicates how prediction markets will structure AI ranking questions going forward.

The binary structure of this market worked reasonably well, but it obscured a more nuanced story. Asking which company holds second place does not capture whether the second-place model is close to or far from the first. In June 2026, the gap between first and second was Anthropic versus Anthropic, a non-gap that tells traders more about one lab’s output velocity than about the competitive spread across the industry. Future markets on AI rankings may need to incorporate margin-of-performance data to price the spread more accurately.

• Anthropic’s July 2026 companion market on best AI model end of July sits at a 96% implied probability, suggesting the market expects the sweep to continue at least one more month.
• Claude Sonnet 5, released June 30, 2026, brings Opus-class coding capability at a lower price tier, widening Anthropic’s deployment footprint and reinforcing its leaderboard durability.
• OpenAI’s GPT-5.5 and Google’s Gemini 3.1 Pro remain the most credible challengers, but neither closed the composite score gap by June 30 according to the Artificial Analysis Intelligence Index.
• The resolution rule using Chatbot Arena with an alphabetical tiebreaker for tied scores adds a structural quirk that traders should factor into future similar markets if top models converge in score.

LINES RESOLUTION VERDICT

ANTHROPIC: CORRECTLY PRICED

Traders read the leaderboard momentum accurately from day one, and Anthropic’s dual-model dominance on Chatbot Arena confirmed the contract’s direction well before June 30 arrived.

What the market showed: The market opened at 88% for Anthropic and closed at 100%. The 12-point move came from a clean benchmark confirmation, not a surprise. Traders held the right conviction and sized it appropriately given the alternatives still in play at open.