Benchmark
Kagi LLM Benchmark
Scores are percentages.
- Models measured
- 110
- Scores recorded
- 123
- Sources
- 1
- IQ weight
- 1.63
Score distribution
- 54.3%
- Mean
- 55.2%
- Median
- 17.8%
- Std dev
- 31.5%
- 10th percentile
- 75%
- 90th percentile
- 6.3%
- Min
This benchmark has a weight of 1.63 for chat models, measured over 110 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | |
|---|---|---|
| 101 | Devstral Small 2505· Kagi LLM Benchmark | 37.7% |
| 102 | Llama 4 Scout· Kagi LLM Benchmark | 36.9% |
| 103 | Qwen2.5 VL 72B Instruct· Kagi LLM Benchmark | 36% |
| 104 | Grok 4 Fast· Kagi LLM Benchmark | 35.4% |
| 105 | Gemma 3 27B IT· Kagi LLM Benchmark | 35.1% |
| 106 | Meta Llama 3 70b· Kagi LLM Benchmark | 35.1% |
| 107 | Claude 3 Haiku· Kagi LLM Benchmark | 34.2% |
| 108 | GPT-4.1 nano· Kagi LLM Benchmark | 33.3% |
| 109 | Codestral 25.01· Kagi LLM Benchmark | 32.5% |
| 110 | Virtuoso Large· Kagi LLM Benchmark | 32.4% |
| 111 | Gemma 3n E4B IT· Kagi LLM Benchmark | 31.5% |
| 112 | Hermes 4 405Bno-thinking· Kagi LLM Benchmark | 30.1% |
| 113 | Command A· Kagi LLM Benchmark | 28.8% |
| 114 | GPT-4o mini· Kagi LLM Benchmark | 28.8% |
| 115 | Jamba Large 1.7· Kagi LLM Benchmark | 26.1% |
| 116 | Gemma 3 4B IT· Kagi LLM Benchmark | 25.2% |
| 117 | Mercury· Kagi LLM Benchmark | 21.6% |
| 118 | mercury-coder· Kagi LLM Benchmark | 20.7% |
| 119 | Spotlight· Kagi LLM Benchmark | 18.9% |
| 120 | Phi 4 Multimodal Instruct· Kagi LLM Benchmark | 17.1% |
| 121 | Magistral Medium· Kagi LLM Benchmark | 16.2% |
| 122 | jamba-mini-1.7· Kagi LLM Benchmark | 11.7% |
| 123 | Magistral Small 2509· Kagi LLM Benchmark | 6.3% |
Showing 101–123 of 123 results.
Tokens per task, input and output together, as measured by the source. Comparable within this benchmark only.