Benchmark
Multi-IF
Instruction following benchmarks measure how closely models follow instructions. Scores are percentages.
All instruction following benchmarksScores from LLM Stats data.
- Models measured
- 20
- Scores recorded
- 22
- Sources
- 1
- IQ weight
- 1.22
Score distribution
- 74%
- Mean
- 75.1%
- Median
- 6.6%
- Std dev
- 66.1%
- 10th percentile
- 80.1%
- 90th percentile
- 57.2%
- Min
This benchmark has a weight of 1.22 for chat models, measured over 20 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Ling 3.0 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 87.7% |
| 2 | Qwen 3 235B A22B Thinking 2507· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 80.6% |
| 3 | LFM2.5 2.6B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 80.1% |
| 4 | o3 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 79.5% |
| 5 | Qwen3 VL 235Bthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 79.1% |
| 6 | Qwen3 VL 32B Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 78% |
| 7 | Qwen3 Next 80B A3Bthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 77.8% |
| 8 | Qwen3 235B A22B Instruct 2507· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 77.5% |
| 9 | Qwen3 VL 235B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 76.3% |
| 10 | Qwen3 Next 80B A3B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 75.8% |
| 11 | Qwen: Qwen3 VL 8B Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 75.1% |
| 12 | Qwen3 VL 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 75.1% |
| 13 | Qwen3 14B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 74.8% |
| 14 | qwen3-vl-30bthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 73% |
| 15 | Qwen3 30B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 72.2% |
| 16 | Qwen3 VL 32B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 72% |
| 17 | GPT 4.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 70.8% |
| 18 | GPT-4.5 Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 70.8% |
| 19 | GPT-4.1 mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 67% |
| 20 | qwen3-vl-30b· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 66.1% |
| 21 | GPT 4o· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 60.9% |
| 22 | GPT-4.1 nano· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 57.2% |
Data credits: Data by LLM Stats