* Finally to FastAPI `0.70.0`, just upgrading Starlette to the latest version `0.16.0` with additional bug fixes.

This way, in case there was a breaking change for your code in one of the releases, you can still benefit from the previous upgrades. ✨

### Breaking Changes - Upgrade

* ⬆️ Upgrade Starlette to 0.16.0. PR [#4016](https://github.com/tiangolo/fastapi/pull/4016) by [@tiangolo](https://github.com/tiangolo).

015E          ; mapped                 ; 015F          # 1.1  LATIN CAPITAL LETTER S WITH CEDILLA
015F          ; valid                                  # 1.1  LATIN SMALL LETTER S WITH CEDILLA
0160          ; mapped                 ; 0161          # 1.1  LATIN CAPITAL LETTER S WITH CARON
0161          ; valid                                  # 1.1  LATIN SMALL LETTER S WITH CARON

5))) var x158 uint64 var x159 uint64 x158, x159 = bits.Add64(x134, x149, uint64(fiatScalarUint1(x157))) var x160 uint64 _, x160 = bits.Mul64(x150, 0xd2b51da312547e1b) var x162 uint64 var x163 uint64 x163, x162 = bits.Mul64(x160, 0x1000000000000000) var x164 uint64 var x165 uint64 x165, x164 = bits.Mul64(x160, 0x14def9dea2f79cd6) var x166 uint64 var x167 uint64 x167, x166 = bits.Mul64(x160, 0x5812631a5cf5d3ed) var x168 uint64 var x169 uint64 x168, x169 = bits.Add64(x167, x164, uint64(0x0)) x170 :=...

Tseng, seaotterman, Seo Sanghyeon, Sergei Lebedev, Ted Chang, terrytangyuan, Tim
H, tkunic, Tod, vihanjain, Yan Facai (颜发才), Yin Li, Yong Tang, Yukun Chen,
Yusuke Yamada

# Release 1.6.0

## Breaking Changes

*   Prebuilt binaries are now built against CUDA 9.0 and cuDNN 7.
*   Prebuilt binaries will use AVX instructions. This may break TF on older
    CPUs.