chatglm2-6b
chatglm2-6b.
Released 24 Jun 2023
12
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Unknown lab Jun 2023
chatglm2-6b
98.8
#249 of 271
Epoch Capabilities Index
- L-Eval 38.5% #4 /5
- LAMBADA 54.3% #26 /26
- AlpacaEval 47.1% #18 /19
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ARC-Challenge | 61% | #32 of 75 | Epoch AI | |
| BoolQ | 79% | #40 of 75 | Epoch AI | |
| L-Eval | 38.5% | #4 of 5 | Official leaderboard | |
| MMLU | 47.9% | #102 of 127 | Epoch AI | |
| HellaSwag | 57% | #64 of 74 | Epoch AI | |
| BIG-Bench Hard | 33.7% | #41 of 45 | Epoch AI | |
| AlpacaEval | 47.1% | #18 of 19 | Official leaderboard | |
| PIQA | 69.6% | #57 of 58 | Epoch AI | |
| LMArena (Elo) | 1,025 | #328 of 333 | Official leaderboard | |
| LAMBADA | 54.3% | #26 of 26 | Epoch AI |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GSM8K | 32.4% | #55 of 82 | Epoch AI |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SuperCLUE-Safety | 79.4 | #4 of 7 | Official leaderboard |
The lab's models over time
's models in color, against every other model on the Epoch Capabilities Index, by release date.