Indicator Retrieval on BacktestBench synthetic (test)
87.15AccuracyGemini 3 Pro
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Gemini 3 ProAccessibility=Closed-Source2026.05 | 87.15 | 96.08 | 96.07 | 96.08 | |
| Kimi K2 ThinkingAccessibility=Open-Source2026.05 | 79.82 | 95.09 | 96.86 | 95.96 | |
| Seed 1.8Accessibility=Closed-Source, Version=1.82026.05 | 76.93 | 94.27 | 96.24 | 95.25 | |
| Qwen3 235BAccessibility=Open-Source, Parameters=235B2026.05 | 76.41 | 93.62 | 94.75 | 94.18 | |
| Kimi Linear 48BAccessibility=Open-Source, Parameters=48B2026.05 | 74.37 | 79.38 | 82.05 | 80.69 | |
| GLM 4.7Accessibility=Open-Source2026.05 | 74.35 | 93.48 | 94.53 | 94 | |
| Qwen3 Next 80BAccessibility=Open-Source, Parameters=80B2026.05 | 73.8 | 92.57 | 94.54 | 93.54 | |
| Qwen3 MaxAccessibility=Closed-Source2026.05 | 72.42 | 91.75 | 95.82 | 93.74 | |
| Qwen3 30BAccessibility=Open-Source, Parameters=30B2026.05 | 69.73 | 91.78 | 93.25 | 92.51 | |
| Deepseek V3.2Accessibility=Open-Source2026.05 | 68.98 | 91.01 | 94.88 | 92.9 | |
| Seed OSS 36BAccessibility=Open-Source, Parameters=36B2026.05 | 68.07 | 90.49 | 91.89 | 91.18 | |
| GPT OSS 120BAccessibility=Open-Source, Parameters=120B2026.05 | 67.65 | 90.46 | 93.35 | 91.88 | |
| GPT OSS 20BAccessibility=Open-Source, Parameters=20B2026.05 | 67.54 | 90.86 | 92.76 | 91.8 | |
| Qwen3 32BAccessibility=Open-Source, Parameters=32B2026.05 | 67.41 | 90.44 | 93.99 | 92.18 | |
| Qwen3 Coder PlusAccessibility=Closed-Source2026.05 | 65.09 | 89.94 | 94.24 | 92.04 | |
| Qwen3 14BAccessibility=Open-Source, Parameters=14B2026.05 | 60.92 | 87.56 | 90.5 | 89.01 | |
| Qwen3 8BAccessibility=Open-Source, Parameters=8B2026.05 | 60.61 | 87.33 | 90.66 | 88.96 | |
| Ministral 3 14BAccessibility=Open-Source, Parameters=14B2026.05 | 59.62 | 86.77 | 90.75 | 88.71 | |
| Mimo V2 FlashAccessibility=Open-Source2026.05 | 58.58 | 86.67 | 93.38 | 89.9 | |
| Qwen3 4BAccessibility=Open-Source, Parameters=4B2026.05 | 58.34 | 87.16 | 89.84 | 88.48 | |
| MiniMax M2.1Accessibility=Open-Source2026.05 | 57.09 | 86.65 | 90.76 | 88.66 | |
| GLM 4.7 FlashAccessibility=Open-Source2026.05 | 56.18 | 88.03 | 90.8 | 89.39 | |
| Ministral 3 8BAccessibility=Open-Source, Parameters=8B2026.05 | 47.99 | 82.2 | 87.69 | 84.86 |