Aphrodite-engine

What's Changed
* Set CPU Affinity: Electric Boogaloo V2 by KaraKaraWitch in https://github.com/PygmalionAI/aphrodite-engine/pull/187
* chore: backlog 1 by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/191
* feat: support GPTQ 2, 3, and 8bit quants by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/181
* feat: FP8 KV Cache (ENG-4) by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/185
* feat: tokenizer endpoint for OpenAI API by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/195
* feat: rejection sampler by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/197
* feat: better mixtral parallelism by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/193
* fix: triton compile error by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/200
* feat: reduce sampler overhead by making it less blocking by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/198
* fix: test units by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/201
* merge branch 'dev' into 'main' by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/203
* feat: bump cuda to 12.1 by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/205
* bump version to 0.4.6 by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/204

New Contributors
* KaraKaraWitch made their first contribution in https://github.com/PygmalionAI/aphrodite-engine/pull/187

**Full Changelog**: https://github.com/PygmalionAI/aphrodite-engine/compare/v0.4.5...v0.4.6

What's Changed
> [!IMPORTANT]
> Version 0.4.4 was skipped.

Quite a few changes this time around, most notably:

* Implement [DeciLM](https://huggingface.co/Deci/DeciLM-7B) by AlpinDale in #158
* Support prompt logprobs by AlpinDale in 162
* Support safetensors for Mixtral along with expert parallelism for better multi-gpu by AlpinDale in 167
* Implement CUDA graphs for better multi-GPU and optimizing smaller models by AlpinDale in 172
* Fix peak memory profiling to allow higher gmu values by AlpinDale in 166
* Restore compatibility with Python 3.8 and 3.9 by g4rg in 170
* Lazily import model classes to avoid import overhead by AlpinDale in 165
* Add RoPE scaling support for Mixtral models by g4rg in 174
* Make OpenAI API keys optional by AlpinDale in 176

**Full Changelog**: https://github.com/PygmalionAI/aphrodite-engine/compare/v0.4.4...v0.4.5

This is a big release! We've had many new and exciting changes.

What's New
* Mixtral 8x7B support by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/155
* add ROCm support for MI200-300 GPUs by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/95
* implement fused Add RMSNorm kernels by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/125
* add SqueezeLLM support by AlpinDAle in https://github.com/PygmalionAI/aphrodite-engine/pull/140
* add chat templates for the OpenAI endpoint by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/138
* speed up compilation times by 2 to 3x by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/130
* support Phi 1.5 models by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/121

NOTE: You'll need to run `pip install megablocks` if you're using the wheels.

New Contributors
* miku448 made their first contribution in https://github.com/PygmalionAI/aphrodite-engine/pull/108

**Full Changelog**: https://github.com/PygmalionAI/aphrodite-engine/compare/v0.4.2...v0.4.3

What's Changed
* fix: correct auto ntk scaling_factor for 4k ctx case by sandwichdoge in https://github.com/PygmalionAI/aphrodite-engine/pull/101
* fix: cpu memory limit detection for containers by g4rg in https://github.com/PygmalionAI/aphrodite-engine/pull/103
* feat: yi support by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/104
* fix: docker port by Krisseck in https://github.com/PygmalionAI/aphrodite-engine/pull/105
* feat: min_p by StefanGliga in https://github.com/PygmalionAI/aphrodite-engine/pull/106
* chore: api keys for OAI server by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/107

New Contributors
* sandwichdoge made their first contribution in https://github.com/PygmalionAI/aphrodite-engine/pull/101
* Krisseck made their first contribution in https://github.com/PygmalionAI/aphrodite-engine/pull/105

**Full Changelog**: https://github.com/PygmalionAI/aphrodite-engine/compare/v0.4.1...v0.4.2

What's Changed
* fix: AWQ NotImplemented errors by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/102


**Full Changelog**: https://github.com/PygmalionAI/aphrodite-engine/compare/v0.4...v0.4.1

What's Changed
* Make entrypoint executable by city-unit in https://github.com/PygmalionAI/aphrodite-engine/pull/83
* Correct Conda Env Creation in Dockerfile by city-unit in https://github.com/PygmalionAI/aphrodite-engine/pull/82
* feat: prompt logprobs and batched samplers by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/77
* feat: add mistral support for GPTQ by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/86
* feat: finish up tests and workflows by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/87
* feat: flattened 1D tensor -> 2D tensor by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/85
* chore: reformats by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/90
* fix: pylint complaints by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/91
* fix: remove unnecessary lines by g4rg in https://github.com/PygmalionAI/aphrodite-engine/pull/81
* fix: sync CPU delay in sampler by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/93
* New Mirostatv2 implementation by 50h100a in https://github.com/PygmalionAI/aphrodite-engine/pull/96
* feat: spaces between special tokens by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/94
* chore: clean up endpoints by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/98
* feat: add exllamav2 for GPTQ by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/99
* fix: force v2 for ctxlen larger than 8192 by AlpinDale in https://github.com/PygmalionAI/aphrodite-engine/pull/100

New Contributors
* city-unit made their first contribution in https://github.com/PygmalionAI/aphrodite-engine/pull/83

**Full Changelog**: https://github.com/PygmalionAI/aphrodite-engine/compare/v0.3.7...v0.4