Benchmark
Enterprise-Bench / L1-L2 Bench
Scores are percentages.
Scores from Harbor Hub Leaderboards.
- Models measured
- 19
- Scores recorded
- 19
- Sources
- 1
- IQ weight
- 0.99
Score distribution
- 77.4%
- Mean
- 78.6%
- Median
- 18.8%
- Std dev
- 53.6%
- 10th percentile
- 96.4%
- 90th percentile
- 32.9%
- Min
This benchmark has a weight of 0.99 for chat models, measured over 19 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | |
|---|---|---|
| 1 | Claude Opus 5main· Harbor Hub Leaderboards | 97.9% |
| 2 | Claude Opus 5.5main· Harbor Hub Leaderboards | 96.4% |
| 3 | Claude Sonnet 5.5main· Harbor Hub Leaderboards | 96.4% |
| 4 | Claude Opus 4.6main· Harbor Hub Leaderboards | 95.7% |
| 5 | Claude Opus 4.8main· Harbor Hub Leaderboards | 94.3% |
| 6 | GLM 5.2main· Harbor Hub Leaderboards | 92.9% |
| 7 | GLM 5.3main· Harbor Hub Leaderboards | 92.9% |
| 8 | GPT-6 Astramain· Harbor Hub Leaderboards | 92.9% |
| 9 | GPT-6.1 Solmain· Harbor Hub Leaderboards | 87.9% |
| 10 | DeepSeek V4 Promain· Harbor Hub Leaderboards | 78.6% |
| 11 | Grok 4.5main· Harbor Hub Leaderboards | 74.3% |
| 12 | Gemini 4 Argonmain· Harbor Hub Leaderboards | 73.6% |
| 13 | Qwen3.8 Maxmain· Harbor Hub Leaderboards | 72.9% |
| 14 | GPT-5.6 Solmain· Harbor Hub Leaderboards | 70% |
| 15 | Mistral Large 4main· Harbor Hub Leaderboards | 62.1% |
| 16 | DeepSeek V4 Pro 0813main· Harbor Hub Leaderboards | 56.4% |
| 17 | Kimi K3main· Harbor Hub Leaderboards | 53.6% |
| 18 | GPT-6 Lunamain· Harbor Hub Leaderboards | 49.3% |
| 19 | Kimi K2.7 Codemain· Harbor Hub Leaderboards | 32.9% |
Tokens the model generated per task, as measured by the source. Comparable within this benchmark only.