RideWay - Success Rate: leaderboard
The success-rate view of RideWay (2026), a Chinese-language ridehailing tool-agent benchmark. Success rate is the share of trials that pass the three-judge LLM success panel over the pooled 58-task suite with three attempts per model, in percent, higher is better.
Metric: Success rate (%; share of trials passing the three-judge LLM success gate over the pooled 58-task suite, 3 attempts per model). Source: arxiv.org. Saturation forecast: Around February 2027. 24 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | MiMo-V2.5-Pro (Non-reasoning) | 87.4 | #274 (MiMo-V2.5-Pro) |
| 2 | Claude Sonnet 5 | 82.8 | #108 |
| 3 | Gemini 3.5 Flash (Minimal) | 82.2 | #98 (Gemini 3.5 Flash) |
| 4 | Qwen 3.7 Plus (Non-reasoning) | 82.2 | #203 (Qwen 3.7 Plus) |
| 5 | Claude Opus 4.6 | 81.6 | #100 |
| 6 | DeepSeek V4 Pro (Non-reasoning) | 81 | #181 (DeepSeek V4 Pro) |
| 7 | GLM-5.2 (Non-reasoning) | 78.7 | #137 (GLM-5.2) |
| 8 | Qwen 3.6 27B (Non-reasoning) | 77 | #339 (Qwen 3.6 27B) |
| 9 | Gemma 4 31B | 75.9 | #393 |
| 10 | GPT-5.5 (Non-reasoning) | 75.3 | #48 (GPT-5.5) |
| 11 | Qwen 3.7 Max (Non-reasoning) | 74.1 | #142 (Qwen 3.7 Max) |
| 12 | Qwen 3.6 Plus (Non-reasoning) | 71.8 | #217 (Qwen 3.6 Plus) |
| 13 | Kimi K2.6 (Non-reasoning) | 61.5 | #173 (Kimi K2.6) |
| 14 | Gemma 4 26B | 60.3 | #614 |
| 15 | Qwen 3.5 9B (Non-reasoning) | 59.2 | #803 (Qwen 3.5 9B) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=rideway-success-rate · How It Works · Data refreshed daily, snapshot 2026-11-12.