AI benchmark results and model performance
Individual benchmark scores plotted by date.
| Organisation | Model | Reported | Top Score | Info | Self Reported | Source |
|---|---|---|---|---|---|---|
| Gemini 2.0 Flash | 05 Feb 2025 | 22.20% | Whole | Yes | Source | |
| Claude 3.5 Haiku | 04 Nov 2024 | 28% | - | No | Source | |
| Gemini 2.5 Flash Lite Preview (2025-06-17) | 17 Jun 2025 | N/A | No Thinking | Yes | Source | |
| GPT 4.1 Nano | 14 Apr 2025 | 8.90% | - | No | Source | |
| Gemini 2.5 Flash Lite Preview (2025-09-25) | 25 Sept 2025 | 26.70% | inferred alias from gemini-2.5-flash-lite | Yes | Source | |
| GPT 4o (2024-08-06) | 06 Aug 2024 | 30.70% | - | Yes | Source | |
| GPT 4o Audio Preview (2025-06-03) | 03 Jun 2025 | 30.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Audio Preview (2024-10-01) | 01 Oct 2024 | 30.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Audio Preview (2024-12-17) | 17 Dec 2024 | 30.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Realtime Preview (2025-06-03) | 03 Jun 2025 | 30.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Realtime Preview (2024-10-01) | 01 Oct 2024 | 30.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Search Preview | 11 Mar 2025 | 30.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Transcribe | 20 Mar 2025 | 30.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Transcribe Diarize | 15 Oct 2025 | 30.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4.1 Mini | 14 Apr 2025 | 32.40% | - | No | Source | |
| o1 mini | 12 Sept 2024 | 32.90% | - | No | Source | |
| GPT OSS 20b | 05 Aug 2025 | 34.20% | High Reasoning Effort | Yes | Source | |
| GPT OSS 120b | 05 Aug 2025 | 44.40% | High Reasoning Effort | Yes | Source | |
| GPT 4.5 Preview | 27 Feb 2025 | 44.90% | - | No | Source | |
| Magistral Medium 1.0 | 10 Jun 2025 | 47.10% | - | Yes | Source | |
| Magistral Medium 1.1 | 24 Jul 2025 | 47.10% | inferred version-family alias from magistral-medium | Yes | Source | |
| Magistral Medium 1.2 | 17 Sept 2025 | 47.10% | inferred version-family alias from magistral-medium | Yes | Source | |
| GPT 5 Nano | 07 Aug 2025 | 48.40% | High Reasoning Effort, Diff Method | Yes | Source | |
| DeepSeek OCR | 20 Oct 2025 | 49.60% | inferred family alias from deepseek-v3 (score=0.3000; benches=20) | Yes | Source | |
| DeepSeek V2 (2024-06-28) | 28 Jun 2024 | 49.60% | inferred family alias from deepseek-v3 (score=0.4159; benches=20) | Yes | Source |
