Back to list
Meituan LongCat Releases General 365: A New Rigorous Benchmark for AI Reasoning Evaluation
Industry NewsMeituanAI BenchmarkingReasoning

Meituan LongCat Releases General 365: A New Rigorous Benchmark for AI Reasoning Evaluation

The Meituan LongCat team has officially launched General 365, a new benchmark specifically designed to evaluate the reasoning capabilities of large language models. In an initial assessment involving 26 mainstream AI models, the benchmark revealed a significant performance gap in the industry. Gemini 3 Pro, currently regarded as one of the most capable models, achieved an accuracy rate of only 62.8%. Furthermore, the evaluation found that the vast majority of tested models failed to reach a 60% accuracy threshold, which is considered a basic passing grade. This release by Meituan sets a new standard for measuring cognitive depth in AI, highlighting that complex reasoning remains a formidable challenge for even the most advanced systems currently available.

美团技术团队

Key Takeaways

  • Launch of General 365: Meituan's LongCat team has introduced a new evaluation standard focused on AI reasoning.
  • Industry-Wide Testing: The benchmark was applied to 26 mainstream models to assess the current state of the industry.
  • Gemini 3 Pro Performance: As the top-performing model in the test, Gemini 3 Pro reached an accuracy of 62.8%.
  • The 60% Threshold: Most models evaluated failed to achieve a 60% accuracy score, indicating a widespread struggle with complex reasoning tasks.

In-Depth Analysis

The Launch of General 365 by Meituan LongCat

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 reasoning, a critical area where many large language models (LLMs) still face significant hurdles. By focusing on reasoning rather than simple knowledge retrieval or linguistic fluency, General 365 aims to provide a more granular look at the logical processing capabilities of modern AI. The introduction of this benchmark by a major technology team like Meituan suggests a growing need for specialized tools that can distinguish between surface-level performance and deep cognitive reasoning.

Performance Disparity Among Mainstream Models

In the inaugural testing phase of General 365, the LongCat team evaluated 26 mainstream models. The results provide a sobering look at the current limitations of artificial intelligence. Even the most advanced model currently available, Gemini 3 Pro, only managed to secure an accuracy rate of 62.8%. While this score places it at the top of the current field, it also highlights how much room for improvement remains in the realm of complex reasoning. The fact that the "strongest on earth" model is barely clearing the 60% mark suggests that the tasks within General 365 are designed to push models to their absolute logical limits.

The Reasoning Performance Gap

Perhaps the most striking finding from the Meituan LongCat report is that the vast majority of models failed to reach the 60% accuracy threshold. In many academic and professional settings, 60% is considered the minimum passing grade. The failure of most mainstream models to meet this benchmark indicates a systemic gap in the reasoning capabilities of current AI architectures. This data suggests that while AI has made massive strides in natural language processing and creative generation, the ability to maintain logical consistency and solve complex multi-step reasoning problems remains an elusive goal for the bulk of the industry's current offerings.

Industry Impact

The release of General 365 is likely to have a significant impact on how AI models are developed and marketed. By establishing a benchmark where even the industry leaders struggle, Meituan is shifting the focus toward logical precision and cognitive depth. This may encourage AI labs to move beyond scaling parameters and instead focus on architectural innovations that enhance reasoning. Furthermore, as more companies look to integrate AI into decision-making processes, benchmarks like General 365 will become essential for identifying which models can actually handle complex, real-world logic versus those that merely simulate intelligence through pattern matching.

Frequently Asked Questions

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

Answer: General 365 is specifically designed to evaluate the reasoning capabilities of AI models, serving as a new standard for measuring logical depth and problem-solving accuracy.

Question: Which model performed the best on the General 365 evaluation?

Answer: Gemini 3 Pro achieved the highest accuracy among the 26 models tested, with a score of 62.8%.

Question: How did the majority of mainstream AI models perform?

Answer: Most of the 26 mainstream models tested were unable to reach the 60% accuracy mark, which is often considered the baseline for a passing grade in reasoning tasks.

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.