IFBench
AI benchmark results and model performance
Type: percentageGeneral
Recorded results
28
Average score
66.25%
Score range
21% - 81.20%
Leading model
81.20% - Grok 4.20
Scores over time
Individual benchmark scores plotted by date.
Model results
| Organisation | Models | Reported | Top score | Info | Self reported | Source |
|---|---|---|---|---|---|---|
| Grok 4.20 | 17 Feb 2026 | 81.20% | Artificial Analysis structured model metrics | No | Source | |
| Grok 4.3 | 30 Apr 2026 | 81% | - | No | Source | |
| Beam | 05 Oct 2026 | 79.70% | Reflection-reported launch score. Source row: IFBench. Accessed 2026-10-05T22:51:12.456Z. Harness, reasoning effort, sampling and trial count are not specified per row. Not independently reproduced. | Yes | Source | |
| Qwen 3.8 27B | 14 Aug 2026 | 79.50% | - | Yes | Source | |
| Qwen 3.7 Max | 21 May 2026 | 79.10% | - | Yes | Source | |
| Muse Glimmer 30B | 10 Aug 2026 | 77% | High reasoning | Yes | Source | |
| Qwen 3.5 27B | 24 Feb 2026 | 76.50% | - | Yes | Source | |
| Qwen 3.5 397B A17B | 16 Feb 2026 | 76.50% | - | Yes | Source | |
| Qwen 3.5 Flash | 23 Feb 2026 | 76.50% | inferred family alias from qwen3.5-27b (score=0.4147; benches=81) | Yes | Source | |
| Qwen 3.5 122B A10B | 24 Feb 2026 | 76.10% | - | Yes | Source | |
| Qwen 3.6 Plus | 01 Apr 2026 | 74.20% | - | Yes | Source | |
| GLM 5 Turbo | 15 Mar 2026 | 73.20% | Artificial Analysis structured model metrics | No | Source | |
| Nemotron 3 Super | 11 Mar 2026 | 72.56% | - | Yes | Source | |
| Nemotron 3.5 Lightning 30B A3B | 11 Aug 2026 | 71.88% | Loose; NVIDIA release evaluation suite | Yes | Source | |
| Solar Pro 3 (2026-03-23) | 23 Mar 2026 | 71.20% | Artificial Analysis structured model metrics | No | Source | |
| Mercury 2 | 24 Feb 2026 | 71% | - | Yes | Source | |
| Qwen 3.5 35B A3B | 24 Feb 2026 | 70.20% | - | Yes | Source | |
| MiniMax M2.1 | 23 Dec 2025 | 70% | Instruction following benchmark | Yes | Source | |
| K EXAONE | 31 Dec 2025 | 67.30% | inferred modality/version alias from k-exaone-236b-a23b | Yes | Source | |
| Qwen 3.5 9B | 02 Mar 2026 | 64.50% | - | Yes | Source | |
| Qwen 3.5 4B | 02 Mar 2026 | 59.20% | - | Yes | Source | |
| Trinity Large Thinking | 01 Apr 2026 | 52.30% | Hugging Face model card benchmark table (arcee-ai/Trinity-Large-Thinking) | Yes | Source | |
| Mistral Small 1.0 | 26 Feb 2024 | 48% | inferred family alias from mistral-small-latest (score=0.3650; benches=9) | Yes | Source | |
| Mistral Small 2.0 | 17 Sept 2024 | 48% | inferred family alias from mistral-small-latest (score=0.3650; benches=9) | Yes | Source | |
| Mistral Small 4 | 16 Mar 2026 | 48% | - | Yes | Source |
