Back to list
Meituan LongCat Team Launches General 365: A New Benchmark Revealing AI Reasoning Limitations
Industry NewsMeituanAI BenchmarkingReasoning

Meituan LongCat Team Launches General 365: A New Benchmark Revealing AI Reasoning Limitations

The Meituan LongCat team has officially released General 365, a new evaluation benchmark specifically designed to measure the reasoning capabilities of large language models. In an extensive test involving 26 mainstream models, the benchmark has highlighted a significant performance gap in the current AI landscape. According to the results, Gemini 3 Pro emerged as the top performer but only managed an accuracy rate of 62.8%. Strikingly, the vast majority of the tested models failed to reach the 60% threshold, which is typically considered a passing grade. This development suggests that while AI has made strides in general tasks, complex reasoning remains a formidable challenge for even the most advanced systems currently available on the market.

美团技术团队

Key Takeaways

  • Meituan's LongCat team has introduced General 365, a rigorous new benchmark for evaluating AI reasoning.
  • A comprehensive test of 26 mainstream models shows that most current AI systems struggle with complex reasoning tasks.
  • Gemini 3 Pro recorded the highest accuracy at 62.8%, setting the current ceiling for the benchmark.
  • The majority of tested models failed to achieve a 60% accuracy score, indicating a widespread "reasoning gap" in the industry.

In-Depth Analysis

The Debut of General 365

The Meituan LongCat team has officially entered the AI evaluation space with the release of General 365. This benchmark is positioned as a new yardstick for measuring the logical and reasoning depth of large language models (LLMs). As the AI industry moves beyond basic text generation and information retrieval, the ability to perform multi-step reasoning and maintain logical consistency has become the new frontier. General 365 aims to provide a standardized metric to quantify these high-level cognitive abilities, offering a clearer picture of how models perform when faced with complex problem-solving scenarios.

Benchmarking the Giants: Gemini 3 Pro and Others

To establish the baseline for General 365, the LongCat team conducted empirical tests on 26 of the most prominent mainstream models currently available. The results serve as a reality check for the state of artificial intelligence. Gemini 3 Pro, which is widely regarded as one of the most capable models in the world, achieved an accuracy rate of 62.8%. While this score placed it at the top of the leaderboard, the figure itself suggests that even the industry's leading models have significant room for improvement. The fact that the highest score sits just above 60% underscores the difficulty of the General 365 evaluation criteria.

The 60% Threshold: A Critical Performance Gap

One of the most significant findings from the Meituan LongCat report is the failure of most models to reach the 60% accuracy mark. In many academic and professional contexts, 60% is viewed as the minimum threshold for competency or a "passing grade." The discovery that the majority of mainstream models could not reach this level on General 365 highlights a critical deficiency in current AI development. It suggests that while models are becoming increasingly proficient at mimicking human language, their underlying reasoning engines are not yet robust enough to handle the complexities presented by this new benchmark. This gap between linguistic fluency and logical reasoning is a primary hurdle that the next generation of AI models will need to overcome.

Industry Impact

The release of General 365 by Meituan's LongCat team is expected to have a notable impact on how AI models are developed and marketed. By providing a benchmark where even top-tier models like Gemini 3 Pro struggle, Meituan is pushing the industry toward a more rigorous standard of accountability. This will likely encourage AI researchers to pivot their focus from increasing parameter counts to improving the qualitative aspects of machine reasoning. Furthermore, General 365 provides a transparent framework for enterprises to evaluate which models are truly capable of handling sophisticated logic-based tasks, potentially influencing future investments and adoption strategies across the tech sector.

Frequently Asked Questions

Question: What is the primary purpose of the General 365 benchmark?

General 365 was developed by the Meituan LongCat team to specifically evaluate and set a new standard for the reasoning capabilities of large language models, moving beyond general performance metrics.

Question: How did the top-performing model fare on General 365?

Gemini 3 Pro was the highest-scoring model among the 26 tested, achieving an accuracy rate of 62.8%. However, this was the only model to significantly exceed the 60% mark.

Question: What does the failure of most models to reach 60% accuracy signify?

It indicates that complex reasoning remains a major weakness for the majority of mainstream AI models. The results suggest that current AI technology still faces substantial challenges in performing logical tasks that meet a basic threshold of competency as defined by the General 365 benchmark.

Related News

How Google Duplex Evolved to Call Businesses on Your Behalf in Gemini
Industry News

How Google Duplex Evolved to Call Businesses on Your Behalf in Gemini

A Google Research Blog entry titled “How Google Duplex evolved to also call businesses for you in Gemini” describes its subject as the evolution of Google Duplex to support calls to businesses on a user’s behalf in Gemini. The entry is dated November 20, 2026, and the supplied article text consists only of the label “Generative AI.” The headline identifies the capability and connects it to Duplex, but the available text does not explain how the system evolved, how users request a call, or where the feature is available. The headline establishes the topic, but the supplied material does not provide a technical account or rollout details.

How AI and Lean Are Helping Automate Formal Verification to Fight Software Slop
Industry News

How AI and Lean Are Helping Automate Formal Verification to Fight Software Slop

In Latent Space, Richard MacManus examines formal verification as an emerging way to fight what the headline calls “slop.” The central claim is that proving software correct with mathematics has traditionally required specialist work, and that AI and Lean are now helping automate that process. The distinction is important: the reported change concerns how the work of proving correctness is done, not what counts as a proof. The supplied excerpt presents automation as assistance, without saying that it replaces specialists or completes proofs independently. It also does not specify which parts of the process AI and Lean help automate, how much work they save, or which software has been verified this way.

AI and Human Claim Solutions to the 167-Year-Old Riemann Hypothesis
Industry News

AI and Human Claim Solutions to the 167-Year-Old Riemann Hypothesis

TechCrunch AI reports that AI claimed a solution to the Riemann hypothesis, a mathematical problem the headline describes as 167 years old. A human also put forward a competing solution. The report, written by Tim Fernholz and published on November 20, 2026, describes mathematicians as astonished by the AI-driven result. The supplied excerpt does not explain either proposed solution, identify the AI system or the human claimant, or say whether either claim has been verified. The distinction between a claimed solution and a confirmed one is central to reading the report: the excerpt establishes that two claims are being reported, but it does not establish that either has resolved the hypothesis.