MongoDB - Build AI That Scales
Mini toolsARC Challenge
AI2 Reasoning Challenge (Challenge set)
Grade-school science multiple-choice, hard subset. Saturated by frontier models but still in many evaluation harnesses.
Knowledge
Text
Accuracy
Max 100.0%
Released Mar 2018
Saturated
Possibly contaminated
10
Results
9
Models scored
554.0%
Top: Gemma 2
91.3%
Median
Best results
Top primary scores; one row per model.
Frontier over time
Each dot is one model result; the line traces the running best score.
All results
Showing all configurations including non-primary alternates. · Show only primary
| # | Model | Score | Conditions | Eval date | Source | Flags |
|---|---|---|---|---|---|---|
| 1 | Gemma 2 | 554.0% | — | 25 Feb 2025 | Self-reported | Primary |
| 2 | Claude Opus 3 | 96.4% | 25-shot | 22 Oct 2024 | Self-reported | Primary |
| 3 | Nemotron 3 Super | 96.1% | 25-shot | 03 Apr 2026 | Self-reported | Primary |
| 4 | Nova Pro | 94.8% | 0-shot | 03 Dec 2024 | Self-reported | Primary |
| 5 | Nova Lite | 92.4% | 0-shot | 03 Dec 2024 | Self-reported | Primary |
| 6 | Claude 2 | 91.0% | 5-shot · standard | 11 Jul 2023 | Self-reported | |
| 7 | Nova Micro | 90.2% | 0-shot | 03 Dec 2024 | Self-reported | Primary |
| 8 | Claude Haiku 3 | 89.2% | 25-shot · standard | 04 Mar 2024 | Self-reported | |
| 9 | GPT 3.5 | 85.2% | 25-shot · standard | 14 Mar 2023 | Self-reported | |
| 10 | Llama 3.2 | 78.6% | 0-shot | 22 Oct 2024 | Self-reported | Primary |
| 11 | Mixtral 8x7B | 59.7% | — | 01 Dec 2023 | Self-reported | Primary |
| 12 | Mixtral 8x7B | 59.7% | — | 08 Jan 2024 | Self-reported | Primary |
| 13 | Mistral 7B | 55.6% | — | 01 Sep 2023 | Self-reported | Primary |
