SAGE (Service Agent) - Ecommerce Refund: leaderboard
SAGE overall assessment score in the ecommerce refund customer-service scenario: ecommerce refunds (deep decision tree on product status and credit level), combining procedure compliance and conversational quality on a 0-100 scale, higher is better. It is unrelated to the SAGE robustness benchmark.
Metric: Overall Assessment Score (0-100), 0.8 x logical compliance + 0.2 conversational quality, in the Ecommerce Refund scenario (ecommerce refunds (deep decision tree on product status and credit level)), 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 May 2028. 27 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.5 | 72.73 | #160 |
| 2 | Qwen 2.5 32B Instruct | 72.07 | #1247 |
| 3 | DeepSeek V3.2 | 71.51 | #419 |
| 4 | Claude Sonnet 4.5 | 71.21 | #261 |
| 5 | Gemini 3 Pro (Preview) | 70.99 | #119 |
| 6 | GPT-4.1 | 70.32 | #505 |
| 7 | Seed 1.8 | 70.25 | #252 |
| 8 | DeepSeek R1 | 70.16 | #482 |
| 9 | Kimi K2.5 | 69.69 | #266 |
| 10 | DeepSeek V3 | 69.62 | #729 |
| 11 | Gemini 2.5 Pro | 69.46 | #276 |
| 12 | Qwen 2.5 7B Instruct | 69.05 | #2723 |
| 13 | Qwen 2.5 3B Instruct | 68.92 | #5392 |
| 14 | Llama 3.3 70B Instruct | 68.6 | #1385 |
| 15 | MiniMax-M2.1 | 68.33 | #575 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=sage-service-agent-ecommerce-refund · How It Works · Data refreshed daily, snapshot 2026-11-21.