# Discovering best models for a task

**URL:** <https://discuss.huggingface.co/t/discovering-best-models-for-a-task/137222>\
**Category:** Models\
**Created:** [January 24, 2025, 8:58pm UTC](https://discuss.huggingface.co/t/discovering-best-models-for-a-task/137222 "2025-01-24T20:58:06Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![aulakh1](https://avatars.discourse-cdn.com/v4/letter/a/f4b2a3/32.png) [@aulakh1](https://discuss.huggingface.co/u/aulakh1)\
**Post date:** [January 24, 2025, 8:58pm UTC](https://discuss.huggingface.co/t/discovering-best-models-for-a-task/137222/1 "2025-01-24T20:58:06Z")

</div>

Is there any specialized search engine to search the latest models from Github or Huggingface for a given task? Google results are always stale.

---

<div class="post-metadata">

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\
**Post date:** [January 25, 2025, 2:52am UTC](https://discuss.huggingface.co/t/discovering-best-models-for-a-task/137222/2 "2025-01-25T02:52:53Z")

</div>

I think it would be really convenient if it existed, but I don’t think it does. The old-fashioned style of relying on word of mouth is still the mainstream. If I had to say, leaderboards are close to what you’re looking for.

> **[Spaces - Hugging Face](https://huggingface.co/spaces?search=leader)**
>
> Discover amazing ML apps made by the community

---

<div class="post-metadata">

**Author:** ![khaysanti](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/khaysanti/32/39842_2.png) [@khaysanti](https://discuss.huggingface.co/u/khaysanti)\
**Post date:** [January 26, 2025, 5:34pm UTC](https://discuss.huggingface.co/t/discovering-best-models-for-a-task/137222/3 "2025-01-26T17:34:01Z")

</div>

Leaderboards are the best way to stay up-to-date with the best models for a given benchmark, but you’ll still want to test them out for your specific use case. Huggingface has a good collection of popular benchmark leaderboards, but Spaces are also helpful for finding niche benchmarks. Their search capabilities in Spaces isn’t the best though so you may need to get creative with your key word strings. [The Big Benchmarks Collection - a open-llm-leaderboard Collection](https://huggingface.co/collections/open-llm-leaderboard/the-big-benchmarks-collection-64faca6335a7fc7d4ffe974a)