Models · Compare
Llama 3.1 8B Instruct vs Claude 3.5 Sonnet
Rows highlighted in warm gray are where the models differ. Numbers carry their as-of date and primary source.
Specs
| Field | A: Llama 3.1 8B Instruct | B: Claude 3.5 Sonnet |
|---|---|---|
| Released | 2024-07-23 | 2024-06-20 |
| Developer | Meta | Anthropic |
| Openness | Source-available | Proprietary |
| License | Llama 3.1 Community License | Proprietary |
| OSI-approved | no | no |
| Data released | no | no |
| Training code | no | no |
| Architecture | dense | unknown |
| Total params | 8.0B | — |
| Active params | — | — |
| Experts | — | — |
| Context window | 131K | 200K |
| Attention | gqa | unknown |
| Position enc. | rope-llama3 | unknown |
| Pretraining tokens | — | — |
| Post-training | sft, dpo | rlhf, constitutional |
| Training hardware | H100 | — |
| $/M input | — | — |
| $/M output | — | — |
| Output tok/sec | — | — |
Benchmarks
Missing scores render as not reported; never inferred. Bold highlights the leader per benchmark.
General reasoning
| GPQA-Diamond | 30.4 2024-07-23 | — |
Code
Context · A
Extended the 3 series to 128K context via continued pretraining and tool-use post-training. The 405B sibling launched the same day was the first open-weights model at GPT-4-class scale.
Context · B
The first Claude release to beat its own larger sibling (Claude 3 Opus) on most benchmarks. Established Artifacts, driving a wave of code-and-canvas product copies.