HAL USACO: leaderboard
Princeton HAL leaderboard for USACO: 307 competitive programming problems from the USA Computing Olympiad across 4 difficulty tiers. Tests algorithmic problem-solving ability.
Metric: Accuracy (%). Source: hal.cs.princeton.edu. Saturation forecast: Around January 2027. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 (Medium) | 69.06 | #98 (GPT-5) |
| 2 | O4 Mini (High) | 64.82 | #180 (O4 Mini) |
| 3 | O4 Mini (Low) | 53.09 | #180 (O4 Mini) |
| 4 | Claude Opus 4.1 (20250805) | 48.21 | #132 |
| 5 | O3 (Medium) | 46.25 | #124 (O3) |
| 6 | GPT-4.1 (2025-04-14) | 44.95 | #250 |
| 7 | DeepSeek R1 | 38.11 | #251 |
| 8 | Claude 3.7 Sonnet (20250219) | 29.32 | #201 |
| 9 | Gemini 2.0 Flash (001) | 27.04 | #360 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=hal-usaco · How It Works · Data refreshed daily, snapshot 2026-10-20.