# AA-HLE → ARC-AGI-1

Calibration · https://benchgap.net/calibration/aa-hle/current/to/arcagi1/current · data retrieved 2026-11-18

ARC-AGI-1 is estimated from AA-HLE with a linear curve fitted on 36 models measured on both: y = 0.7403·x + 0.5746, R² = 0.52, cross-validated error 8.1 pp. It is used for 2 estimates.

| Estimated model | AA-HLE | ARC-AGI-1 | Source |
|---|---|---|---|
| [Phi-4 Multimodal Instruct](https://benchgap.net/model/phi-4-multimodal-instruct.md) | 5.0% | 61.2% | estimated ± 8.1 pp, low confidence |
| [Llama 3.1 8B](https://benchgap.net/model/llama-3-1-8b.md) | 4.3% | 60.6% | estimated ± 8.1 pp, low confidence |

## Fitted on

| Model | AA-HLE | ARC-AGI-1 |
|---|---|---|
| [Claude Fable 5](https://benchgap.net/model/claude-fable.md) | 55.5% | 98.5% |
| [Claude Fable 5.1](https://benchgap.net/model/claude-fable-5-1.md) | 59.1% | 97.5% |
| [Claude Haiku 5.5](https://benchgap.net/model/claude-haiku-5-5.md) | 44.4% | 89.0% |
| [Claude Opus 4.5 Thinking](https://benchgap.net/model/claude-opus-4-5-thinking.md) | 30.1% | 80.0% |
| [Claude Opus 4.6 (Adaptive)](https://benchgap.net/model/claude-opus-4-6-thinking.md) | 39.9% | 93.0% |
| [Claude Opus 4.8](https://benchgap.net/model/claude-opus-4-8.md) | 48.7% | 92.5% |
| [Claude Opus 5](https://benchgap.net/model/claude-opus-5.md) | 54.9% | 97.5% |
| [Claude Opus 5.5](https://benchgap.net/model/claude-opus-5-5.md) | 61.4% | 97.5% |
| [Claude Sonnet 4.6](https://benchgap.net/model/claude-sonnet-4-6.md) | 13.3% | 86.0% |
| [DeepSeek V4 Flash 0731](https://benchgap.net/model/deepseek-v4-flash-0731.md) | 38.6% | 89.0% |
| [DeepSeek V4 Pro](https://benchgap.net/model/deepseek-v4-pro.md) | 41.0% | 90.0% |
| [Gemini 3.5 Flash](https://benchgap.net/model/gemini-3-5-flash.md) | 42.7% | 92.5% |
| [Gemini 3.6 Flash](https://benchgap.net/model/gemini-3-6-flash.md) | 40.8% | 91.2% |
| [Gemini 3.7 Flash](https://benchgap.net/model/gemini-3-7-flash.md) | 47.9% | 95.5% |
| [Gemini 3.8 Flash](https://benchgap.net/model/gemini-3-8-flash.md) | 47.8% | 98.5% |
| [Gemini 3 Pro](https://benchgap.net/model/gemini-3-pro.md) | 39.7% | 75.0% |
| [Gemini 4 Argon](https://benchgap.net/model/gemini-4-argon.md) | 57.1% | 98.5% |
| [GPT-5.1](https://benchgap.net/model/gpt-5-1.md) | 28.5% | 72.8% |
| [GPT-5.2](https://benchgap.net/model/gpt-5-2.md) | 37.7% | 86.2% |
| [GPT-5.4](https://benchgap.net/model/gpt-5-4.md) | 43.7% | 93.7% |
| [GPT-5.4 mini](https://benchgap.net/model/gpt-5-4-mini.md) | 28.1% | 63.7% |
| [GPT-5.4 nano](https://benchgap.net/model/gpt-5-4-nano.md) | 28.3% | 51.5% |
| [GPT-5.5](https://benchgap.net/model/gpt-5-5.md) | 45.8% | 95.0% |
| [GPT-5.6 Luna](https://benchgap.net/model/gpt-5-6-luna.md) | 39.5% | 88.0% |
| [GPT-5.6 Sol](https://benchgap.net/model/gpt-5-6-sol.md) | 49.5% | 96.5% |
| [GPT-5.6 Terra](https://benchgap.net/model/gpt-5-6-terra.md) | 42.9% | 96.5% |
| [GPT-6.1 Sol](https://benchgap.net/model/gpt-6-1-sol.md) | 52.9% | 96.5% |
| [GPT-6 Astra](https://benchgap.net/model/gpt-6-astra.md) | 54.7% | 97.5% |
| [GPT-6 Luna](https://benchgap.net/model/gpt-6-luna.md) | 38.5% | 86.7% |
| [GPT-6 Sol](https://benchgap.net/model/gpt-6-sol.md) | 47.9% | 95.5% |
| [Grok 4.5](https://benchgap.net/model/grok-4-5.md) | 42.7% | 85.7% |
| [Grok 4.6](https://benchgap.net/model/grok-4-6.md) | 42.9% | 87.0% |
| [Grok 4.7](https://benchgap.net/model/grok-4-7.md) | 43.1% | 89.5% |
| [Inkling 2 Small](https://benchgap.net/model/inkling-2-small.md) | 44.1% | 96.5% |
| [Inkling-Small](https://benchgap.net/model/inkling-small.md) | 33.3% | 84.0% |
| [Kimi K3](https://benchgap.net/model/kimi-k3.md) | 46.9% | 94.5% |
