SAGE (Service Agent) - Logical Compliance: leaderboard
Metric: Logical Compliance Score (0-100): 0.4 classification accuracy + 0.4 path correctness + 0.2 action correctness against the graph, averaged over the six scenarios, over SAGE customer-service dialogues: standard operating procedures formalized as dialogue graphs, an adversarial simulated user with varied intents and personas, turns 1, 5, 10, 15 and the final turn scored; three unnamed judge LLMs (majority vote and mean) plus a deterministic rule engine produce the reference; closed models through their APIs, open models served with vLLM; 0-100 scale; higher is better. Source: arxiv.org. Saturation forecast: Around July 2027. 27 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.5 | 72.1 |
| 2 | DeepSeek V3.2 | 72.08 |
| 3 | Gemini 3 Pro (Preview) | 71.27 |
| 4 | Claude Sonnet 4.5 | 70.1 |
| 5 | DeepSeek R1 | 70 |
| 6 | Kimi K2.5 | 69.16 |
| 7 | GLM-4.7 | 68.68 |
| 8 | GPT-4.1 | 68.31 |
| 9 | Gemini 3 Flash (Preview) | 68.16 |
| 10 | DeepSeek V3 | 67.9 |
| 11 | MiniMax-M2.1 | 66.59 |
| 12 | Qwen 2.5 32B Instruct | 66.49 |
| 13 | Llama 3.3 70B Instruct | 66.06 |
| 14 | Qwen 3 32B | 65.97 |
| 15 | Seed 1.8 | 65.78 |
Interactive version: theaggregate.ai/benchmark?slug=sage-service-agent-logical-compliance · How It Works · Data refreshed daily, snapshot 2026-10-07.