Claude Instant
Claude Instant.
10
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Anthropic Aug 2023
Claude Instant
Closed weights
114.2
#194 of 256
Epoch Capabilities Index
- Situational Awareness Dataset (SAD) 43.9% #12 /17
- MedQA 55.9% #55 /71
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 | 86.3% | #9 of 75 | Claim in Releasing Claude Instant 1.2, via Epoch AI | |
| MMLU | 73.4% | #20 of 85 | Claim in Model Card and Evaluations for Claude Models, via Epoch AI | |
| TriviaQA | 78.9% | #18 of 38 | Claim in Model Card and Evaluations for Claude Models, via Epoch AI | |
| EQ-Bench v2 | 69.0 | #55 of 85 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GSM8K | 86.7% | #11 of 70 | Claim in Releasing Claude Instant 1.2, via Epoch AI |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HumanEval | 57.3% | #31 of 46 | Official leaderboard | |
| HumanEval+ | 50.6% | #31 of 46 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MedQA | 55.9% | #55 of 71 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Situational Awareness Dataset (SAD) | 43.9% | #12 of 17 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HELM Lite | 39.9 | #50 of 71 | Official leaderboard |
Anthropic's models over time
Anthropic'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)