Test your LLM-powered apps with a TypeScript-native, Vitest-based eval runner. No API key required.
active 2024-12-07 → 2026-06-17 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 558 days from 2024-12-07 to 2026-06-17. Pushes: 382 total, peak 41 in a day. Pull requests: 139 total, peak 13 in a day. Issues: 186 total, peak 11 in a day. Comments: 389 total, peak 27 in a day. Stars: 896 total, peak 62 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| mattpocock | 665 | 305 | 75 | 153 |
| github-actions[bot] | 101 | 72 | 29 | 0 |
| vercel[bot] | 63 | 0 | 0 | 63 |
| changeset-bot[bot] | 44 | 0 | 0 | 44 |
| DogPawHat | 38 | 0 | 3 | 29 |
| cantemizyurek | 38 | 0 | 7 | 18 |
| ShiboSoftwareDev | 20 | 0 | 0 | 7 |
| pkg-pr-new[bot] | 13 | 0 | 0 | 13 |
| claude[bot] | 13 | 4 | 0 | 8 |
| dschlabach | 10 | 0 | 1 | 5 |
| dcramer | 5 | 0 | 0 | 3 |
| psinha40898 | 5 | 0 | 2 | 3 |
| christianklotz | 5 | 0 | 0 | 2 |
| jacobparis | 5 | 0 | 3 | 1 |
| dkokotov | 5 | 0 | 0 | 4 |
| focux | 4 | 0 | 0 | 2 |
| martincik | 3 | 0 | 2 | 1 |
| JamesHenry | 3 | 0 | 1 | 1 |
| jreidgreer | 3 | 0 | 0 | 2 |
| Swifilaboroka | 3 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#397changeset-bot[bot]2026-05-31 09:09Support reasoning in OpenAI models
- Issue comment#386030814985182026-04-27 00:57Upgrade to tsgo
- Issue#223mattpocock2026-03-27 10:47Eval duration not being tracked properly
- Issue#386mattpocock2026-03-27 10:47Upgrade to tsgo
- Issue#221mattpocock2026-03-27 09:56Adapter: Implement JSON adapter
- Issue#337mattpocock2026-03-27 09:56Document how to evaluate MCP servers
- Issue#307mattpocock2026-03-27 09:56Add 'copy' button to each page of the docs
- Issue#223mattpocock2026-03-27 09:52Eval duration not being tracked properly
- Issue#372mattpocock2026-03-27 09:31Vercel's default gateway model doesn't work with built-in LLM scorers
- Issue#374mattpocock2026-03-27 09:18`wrapAISDKModel` for `EmbeddingModelV2`
- Issue#361mattpocock2026-03-26 20:37wrapAISDKModel: Unsupported content type: reasoning (AI SDK v5 reasoning parts)
- Issue comment#361mattpocock2026-03-26 20:37wrapAISDKModel: Unsupported content type: reasoning (AI SDK v5 reasoning parts)
- Issue comment#362mattpocock2026-03-26 20:33TypeError: Cannot read properties of undefined (reading 'pool') when rendering errors with vitest 4.x
- Issue comment#358mattpocock2026-03-26 20:28CLI crashes when test data content exceeds terminal width
- Issue comment#385vercel[bot]2026-03-12 23:52feat: allow scorer functions to return arrays of scores (#304)
- Issue comment#385changeset-bot[bot]2026-03-12 23:52feat: allow scorer functions to return arrays of scores (#304)
- Issue#383mattpocock2026-02-20 12:02Fix failing CI on v1
- Pull request#382mattpocock2026-02-20 11:58
- Issue comment#379mattpocock2026-02-20 10:49Bump AI SDK dependencies to v6 and migrate core types
- Issue#380mattpocock2026-02-20 10:05Migrate scorers, test fixtures, and examples to v6 APIs
- Issue#379mattpocock2026-02-20 10:05Bump AI SDK dependencies to v6 and migrate core types
- Issue comment#376cantemizyurek2026-02-05 20:03Support multiple file path filters in CLI
- Issue#375akn-trackunit2026-02-04 07:56Documentation: "Standalone UI (works without server)" claim is misleading
- Issue#374Fox322026-01-29 10:58`wrapAISDKModel` for `EmbeddingModelV2`
- Issue comment#335mmmeff2025-12-23 08:30Command to start the UI without running evals
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 896 stars here means stars gained during the window, not the repo's star count.