MPT-30B
MPT-30B.
11
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
MosaicML Jun 2023 MPT-30B
Open weights
85.5
#232 of 256
Epoch Capabilities Index
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 | |
|---|---|---|---|---|
| PIQA | 81.9% | #24 of 58 | Claim in Llama 2: Open Foundation and Fine-Tuned Chat Models, via Epoch AI | |
| BoolQ | 79% | #34 of 61 | Claim in Llama 2: Open Foundation and Fine-Tuned Chat Models, via Epoch AI | |
| WinoGrande | 71% | #45 of 76 | Claim in Llama 2: Open Foundation and Fine-Tuned Chat Models, via Epoch AI | |
| OpenBookQA | 52% | #25 of 41 | Claim in Llama 2: Open Foundation and Fine-Tuned Chat Models, via Epoch AI | |
| ARC-Challenge | 50.6% | #46 of 75 | Claim in Llama 2: Open Foundation and Fine-Tuned Chat Models, via Epoch AI | |
| TriviaQA | 73.6% | #25 of 38 | Claim in Llama 2: Open Foundation and Fine-Tuned Chat Models, via Epoch AI | |
| MMLU | 47.9% | #59 of 85 | Claim in Qwen Technical Report, via Epoch AI | |
| BIG-Bench Hard | 38% | #34 of 45 | Claim in Qwen Technical Report, via Epoch AI | |
| LMArena (Elo) | 1,151 | #307 of 335 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GSM8K | 34.4% | #7 of 24 | Stanford HELM, via Epoch AI |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HELM Classic | 71.5 | #7 of 17 | Official leaderboard |
MosaicML's models over time
MosaicML's models in color, against every other model on the Epoch Capabilities Index, by release date.
Source: Epoch Capabilities Index, Epoch AI (CC BY 4.0)