Compare frameworks
Pydantic AI vs Vercel AI SDK vs DSPy
Every figure below carries the source that reported it and the moment it was observed. Values that failed validation are withheld rather than shown, and a dash means no source we track carries that field for that framework.
| Field | Pydantic AI | Vercel AI SDK | DSPy |
|---|---|---|---|
| Stars | 20,133Synced from sourceCurrent3h ago | 26,922Synced from sourceCurrent3h ago | 38,243 (highest of the compared values)Synced from sourceCurrent3h ago |
| downloads | 3,076,771Synced from sourceStale1mo ago· past SLA | 17,883,737 (highest of the compared values)Synced from sourceCurrent3h ago | — |
| Last commit | 2026-09-24Synced from sourceCurrent3h ago | 2026-09-24Synced from sourceCurrent3h ago | 2026-09-23Synced from sourceCurrent3h ago |
| Language | pythonSeeded, unreviewedwritten 1mo ago | typescriptSeeded, unreviewedwritten 1mo ago | pythonSeeded, unreviewedwritten 1mo ago |
| Licence | MITSynced from sourceCurrent3h ago | — | MITSynced from sourceCurrent3h ago |
| Archived | noSynced from sourceCurrent3h ago | noSynced from sourceCurrent3h ago | noSynced from sourceCurrent3h ago |
| Category | structured-outputSeeded, unreviewedwritten 1mo ago | orchestrationSeeded, unreviewedwritten 1mo ago | optimisationSeeded, unreviewedwritten 1mo ago |
| Contributors | 589Synced from sourceCurrent3h ago | 715Synced from sourceCurrent3h ago | 459Synced from sourceCurrent3h ago |
| Forks | 2768Synced from sourceCurrent3h ago | 5180Synced from sourceCurrent3h ago | 3353Synced from sourceCurrent3h ago |
| Maintainer | PydanticSeeded, unreviewedwritten 1mo ago | VercelSeeded, unreviewedwritten 1mo ago | Stanford NLPSeeded, unreviewedwritten 1mo ago |
| npm downloads, monthly | — | 87791075Synced from sourceCurrent3h ago | — |
| npm downloads, weekly | — | 17883737Synced from sourceCurrent3h ago | — |
| npm package | — | aiSynced from sourceCurrent3h ago | — |
| npm window ends | — | 2026-09-21Synced from sourceCurrent3h ago | — |
| Open issues | 965Synced from sourceCurrent3h ago | 1462Synced from sourceCurrent3h ago | 726Synced from sourceCurrent3h ago |
| PyPI downloads, monthly | 13478631Synced from sourceStale1mo ago· past SLA | — | — |
| PyPI downloads, weekly | 3076771Synced from sourceStale1mo ago· past SLA | — | — |
| PyPI package | pydantic-aiSynced from sourceCurrent3h ago | — | dspy-aiSynced from sourceCurrent3h ago |
| PyPI released | 2026-09-24T03:23:01.147978ZSynced from sourceCurrent3h ago | — | 2026-08-21T23:06:20.945191ZSynced from sourceCurrent3h ago |
| Requires Python | >=3.10Synced from sourceCurrent3h ago | — | >=3.9Synced from sourceCurrent3h ago |
| PyPI version | 2.49.0Synced from sourceCurrent3h ago | — | 3.3.1Synced from sourceCurrent3h ago |
| Repository | https://github.com/pydantic/pydantic-aiSynced from sourceCurrent3h ago | https://github.com/vercel/aiSynced from sourceCurrent3h ago | https://github.com/stanfordnlp/dspySynced from sourceCurrent3h ago |
Sources: GitHub REST API, npm registry downloads, PyPI registry metadata, pypistats.org downloads. Most recent observation 3h ago. Where a row is marked, ▲ is the highest of the values shown and ▼ the lowest — arithmetic on the figures above, not a ranking or a recommendation. Rows where neither extreme is meaningful are left unmarked rather than given a direction they do not have.
Change the comparison
Remove one, or open the directory to pick a different set. Up to 3 at a time.