2 vs 0 benchmarks won
Anthropic Claude 3.7 Sonnet | Meta Code Llama | |
|---|---|---|
| Overview | ||
| Company | Anthropic | Meta |
| Release date | Feb 24 2025 | Aug 24 2023 |
| Model type | — | — |
| Open source | No | Yes |
| Specifications | ||
Parameters | — | — |
Context window | — | — |
| Benchmarks | ||
Science reasoning GPQA Diamond | 68% | — |
Software engineering SWE-Bench Verified | 62.3% | — |
Multimodal understanding MMMU | — | — |
| Timeline | ||
| Release gap | Code Llama shipped 550 days before Claude 3.7 Sonnet | |
Claude 3.7 Sonnet leads Code Llama on 2 of the tracked benchmarks (GPQA Diamond, SWE-Bench Verified, MMMU). Code Llama shipped 550 days before Claude 3.7 Sonnet, so benchmark comparisons should account for the intervening progress.
Code Llama is an open-source / open-weight model; Claude 3.7 Sonnet is proprietary.
Direct benchmark comparisons are unavailable — at least one of these models has not published scores on GPQA Diamond, SWE-Bench Verified, or MMMU.