Perf improvement: json, json, tls-handshakes-kestrel

| Scenario | Environment | Date | Old RPS | New RPS | Change | Deviation | StDev | Dependencies |
| -------- | ----------- | ---- | ------- | ------- | ------ | --------- | ----- | ------------ |
| json | Json Minimal APIs 4- Trends gold-win | 07/28/2026 19:00:33 | 1,366,242 | 1,401,806 | 2.60 % (35,564) :thumbsup: | 2.90 σ | 12,283.66 | Changes NameVersionDiff |
| json | Json Platform Inline 9- Trends cobalt-cloud-lin-azl3-dual | 07/30/2026 09:21:10 | 264,890 | 286,385 | 8.11 % (21,495) :thumbsup: | 2.47 σ | 8,714.55 | Changes NameVersionDiff Microsoft.AspNetCore.App `11.0.0-preview.7.26366.102` -> `11.0.0-rc.1.26379.102` https://github.com/dotnet/dotnet/compare/d2b524965e50619e8db62ed35943981830e1008d...813f634ceb016018f5acc9bd3c2b16e17dff4686 Microsoft.NETCore.App `11.0.0-preview.7.26366.102` -> `11.0.0-rc.1.26379.102` https://github.com/dotnet/dotnet/compare/d2b524965e50619e8db62ed35943981830e1008d...813f634ceb016018f5acc9bd3c2b16e17dff4686 Microsoft.AspNetCore.Antiforgery.dll `11.0.0-rc.1.26379.102` (__New__ ) n/a Microsoft.AspNetCore.Antiforgery.dll `11.0.0-preview.7.26366.102` (__Removed__) n/a |
| tls-handshakes-kestrel | Kestrel Linux: TLS Handshakes 9- Trends cobalt-cloud-lin-azl3-dual | 07/29/2026 09:45:21 | 1,708 | 1,844 | 0.00 % (136) :thumbsup: | 4.33 σ | 31.43 | Changes NameVersionDiff Microsoft.AspNetCore.App `11.0.0-preview.7.26366.102` -> `11.0.0-rc.1.26378.118` https://github.com/dotnet/dotnet/compare/d2b524965e50619e8db62ed35943981830e1008d...43cdb1f62dde1df80d5f9c3532c8cfacbbc40f53 Microsoft.NETCore.App `11.0.0-preview.7.26366.102` -> `11.0.0-rc.1.26378.118` https://github.com/dotnet/dotnet/compare/d2b524965e50619e8db62ed35943981830e1008d...43cdb1f62dde1df80d5f9c3532c8cfacbbc40f53 Microsoft.AspNetCore.Antiforgery.dll `11.0.0-rc.1.26378.118` (__New__ ) n/a Microsoft.AspNetCore.Antiforgery.dll `11.0.0-preview.7.26366.102` (__Removed__) n/a |

[PowerBI Dashboard](https://aka.ms/aspnet/benchmarks)

@sebastienros

Start with the benchmark results in this issue and compare the JSON and tls-handshakes-kestrel scenarios in the linked PowerBI Dashboard. The payload reports performance changes but does not identify a file, entry point, test, or specific code change, so a concrete target and completion criterion are still needed.

Written by the indexing model from the issue text.