# Web Framework Benchmarks

**URL:** <https://forums.fsharp.org/t/web-framework-benchmarks/1347>\
**Category:** General\
**Tags:** web\
**Created:** [June 23, 2020, 1:56pm UTC](https://forums.fsharp.org/t/web-framework-benchmarks/1347 "2020-06-23T13:56:02Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![panesofglass](https://yyz2.discourse-cdn.com/flex030/user_avatar/forums.fsharp.org/panesofglass/32/755_2.png) [@panesofglass](https://forums.fsharp.org/u/panesofglass)\
**Post date:** [June 23, 2020, 1:56pm UTC](https://forums.fsharp.org/t/web-framework-benchmarks/1347/1 "2020-06-23T13:56:02Z")

</div>

In case you are curious, a [new benchmark](https://github.com/the-benchmarker/web-frameworks) appeared, and I was able to enter several F# frameworks for comparison. The benchmark is really simple at present, but I think their plans are to increase the complexity over time to show a more real-world use case than something like the [TechEmpower benchmarks](https://www.techempower.com/benchmarks/) which are typically optimized for performance. The F# frameworks are holding their own pretty well, especially when compared with vanilla [ASP.NET](http://ASP.NET) Core.

 ![image](https://canada1.discourse-cdn.com/flex030/uploads/fsharp/original/1X/7e89a5a9058f36830d116559aabfbe0644ba6d4a.png)

---

<div class="post-metadata">

**Author:** ![panesofglass](https://yyz2.discourse-cdn.com/flex030/user_avatar/forums.fsharp.org/panesofglass/32/755_2.png) [@panesofglass](https://forums.fsharp.org/u/panesofglass)\
**Post date:** [June 23, 2020, 1:57pm UTC](https://forums.fsharp.org/t/web-framework-benchmarks/1347/2 "2020-06-23T13:57:15Z")

</div>

Also, I cannot seem to stop from chanting, “Go! Go! Go! Go!” silently to myself while reading through the list, particularly in this section.

---

<div class="post-metadata">

**Author:** ![chethusk](https://yyz2.discourse-cdn.com/flex030/user_avatar/forums.fsharp.org/chethusk/32/11_2.png) [@chethusk](https://forums.fsharp.org/u/chethusk)\
**Post date:** [June 23, 2020, 2:03pm UTC](https://forums.fsharp.org/t/web-framework-benchmarks/1347/3 "2020-06-23T14:03:12Z")

</div>

I’m very interested in seeing what giraffe and saturn do when the EndpointRouting-based-handlers work merges in! One thing that’s unclear to me in these benchmarks is the gap between giraffe and saturn, as saturn’s own testing shows that it’s a very very small delta between them for common use cases:

```auto
25h| Method | Categories | Mean | Error | StdDev | Ratio | RatioSD |
25h|------------ |----------- |---------:|--------:|--------:|------:|--------:|
25h| SaturnGet | GET | 258.9 us | 5.04 us | 8.13 us | 0.99 | 0.03 |
25h| GiraffeGet | GET | 262.8 us | 5.21 us | 7.31 us | 1.00 | 0.00 |
25h| | | | | | | |
25h| SaturnGetf | GETF | 290.3 us | 5.76 us | 6.86 us | 1.07 | 0.04 |
25h| GiraffeGetf | GETF | 273.6 us | 5.30 us | 4.96 us | 1.00 | 0.00 |
25h| | | | | | | |
25h| SaturnPost | POST | 270.1 us | 4.86 us | 5.60 us | 0.97 | 0.04 |
25h| GiraffePost | POST | 280.3 us | 5.39 us | 7.38 us | 1.00 | 0.00 |

```