Benchmark
Arena ELO - Industry Life and Physical and Social Science
People compare model answers and choose which they prefer. Higher scores are better.
All human preference benchmarksScores from LMArena data.
- Models measured
- 364
- Scores recorded
- 421
- Sources
- 1
- IQ weight
- 1.24
Score distribution
- 1,316.4
- Mean
- 1,364.4
- Median
- 158.9
- Std dev
- 1,071.5
- 10th percentile
- 1,475.1
- 90th percentile
- 800.9
- Min
This benchmark has a weight of 1.24 for chat models, measured over 364 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 |
|---|---|---|
| 401 | PaLM 2· LMArena data | 1,011.1 |
| 402 | Qwen-14B-Chat· LMArena data | 1,010.2 |
| 403 | OLMo 7B Instruct· LMArena data | 1,007.9 |
| 404 | Vicuna 7B· LMArena data | 1,000.9 |
| 405 | Mistral 7B Instruct· LMArena data | 998.5 |
| 406 | Gemma 1.1 2B IT· LMArena data | 996 |
| 407 | Koala 13B· LMArena data | 986.3 |
| 408 | Qwen1.5-4B-Chat· LMArena data | 976.3 |
| 409 | Gemma 2B IT· LMArena data | 949.7 |
| 410 | ChatGLM3 6B· LMArena data | 931.8 |
| 411 | RWKV-4 Raven 14B· LMArena data | 927.8 |
| 412 | ChatGLM2 6B· LMArena data | 915.8 |
| 413 | MPT 7B Chat· LMArena data | 913.2 |
| 414 | GPT4All 13B Snoozy· LMArena data | 901.9 |
| 415 | OpenAssistant Pythia 12B· LMArena data | 896.6 |
| 416 | Alpaca 13B· LMArena data | 882.8 |
| 417 | FastChat-T5 3B· LMArena data | 865.5 |
| 418 | ChatGLM 6B· LMArena data | 849.4 |
| 419 | LLaMA 13B· LMArena data | 806.2 |
| 420 | StableLM Tuned Alpha 7B· LMArena data | 803.3 |
| 421 | Dolly 2.0 12B· LMArena data | 800.9 |
Showing 401–421 of 421 results.
Data credits: LMArena, CC BY 4.0 (licence)