0.0.33

What's Changed

• fix-orjson by @michaelfeil in #201
• Add EngineArray Multi-Model [1/3] by @michaelfeil in #200
• Openapi tests by @michaelfeil in #199
• refactor BatchHandler into ModelWorker by @michaelfeil in #202
• Add fp32 as runtime dtype by @michaelfeil in #211

Full Changelog: 0.0.32...0.0.33

0.0.32

What's Changed

You can now run a model with a alias. This will help you communicating with the API.

infinity_emb --served-model-name "your_nickname"

You can now use preload models. This acts as a "run download and load into ram" test. Upon execution, all files are cached, which will speedup consecutive loads. For additonal speedups, use --no-model-warmup to skip model warmup after loading.

infinity_emb --preload-only --model--name-or-path BAAI/bge-large-en-v1.5 

PR's

• feat: add served_model_name argument for the infinity_server by @bufferoverflow in #180
• FIX: import crossencoder without torch installed and git push of creds by @michaelfeil in #181
• update default model_name to be unified name across routes by @michaelfeil in #179
• python39 type hints by @michaelfeil in #182
• pydantic cli / args validation by @michaelfeil in #183
• update defered moving to cpu & type hints improvement by @michaelfeil in #187
• Update README.md - add Contributors by @michaelfeil in #189
• update infinity offline solution by @michaelfeil in #195
• update offline-mode: deployment docs v2 by @michaelfeil in #196

New Contributors

• @bufferoverflow made their first contribution in #180 Thanks!

Full Changelog: 0.0.31...0.0.32

0.0.31

What's Changed

• Create ISSUE_TEMPLATE by @michaelfeil in #168
• bump sentence transformers to v.2.6.0 by @michaelfeil in #169
• Embedding quant by @michaelfeil in #170
• refactor ENUM..TypeHint into a function by @michaelfeil in #172
• refactored more imports by @michaelfeil in #171
• redirect to /docs and optional imports by @michaelfeil in #175
• update typing by @michaelfeil in #176
• update lock by @michaelfeil in #177

Full Changelog: 0.0.30...0.0.31

0.0.30

What's Changed

• remove fastembed by @michaelfeil in #141
• Sentence transformers bump to 2.5.0 by @michaelfeil in #142
• Revert "Sentence transformers bump to 2.5.0" by @michaelfeil in #143
• Update README.md by @michaelfeil in #145
• update poetry lock - sentence-transformers 2.5.0 by @michaelfeil in #144
• Support for Inferentia2 (draft) by @michaelfeil in #118
• Add bettertransformer to cli by @michaelfeil in #152
• Fp8 support by @michaelfeil in #153
• Some docstring and typing fixes by @lckr in #156
• add async tokenization to reranker in torch by @michaelfeil in #154
• Update README.md by @sherwin684 in #167

New Contributors

• @lckr made their first contribution in #156
• @sherwin684 made their first contribution in #167

Full Changelog: 0.0.29...0.0.30

0.0.29

What's Changed

• OpenAI models compatability and update docs and by @michaelfeil in #140

This will be the last release with fastembed - fastembed and optimum provide similar capabilities. Please use optimum going forward.

Full Changelog: 0.0.28...0.0.29

0.0.28

What's Changed

• add macos ci by @michaelfeil in #133
• Quantization: int8 by @michaelfeil in #134
• add docs via mkdocs by @michaelfeil in #137

Full Changelog: 0.0.27...0.0.28

0.0.27

What's Changed

• BREAKING: EngineArgs by @michaelfeil in #124
  • new stable interface. batch-size=32 as default. using michaelfeil/bge-small as default model. You can overwrite the pooling method now.
• multiple os ci and python 3.12 support by @michaelfeil in #131
• add michaelfeil/bge-small as default model by @michaelfeil in #135

Full Changelog: 0.0.26...0.0.27

0.0.26

What's Changed

• hf_transfer is automatically used by @michaelfeil in #112
• add revision to onnx by @michaelfeil in #109
• bump sentence-transformers to 2.4.0 by @michaelfeil in #113
• Adds benchmarking by @michaelfeil in #110
• ONNX/Optimum now works on windows by @michaelfeil in #117

Full Changelog: 0.0.25...0.0.26

0.0.25

What's Changed

• add engine args similar to vllm by @michaelfeil in #102
• ct2 bump by @michaelfeil in #103
• Deps free by @michaelfeil in #104
• fix: cli start by @michaelfeil in #105

Full Changelog: 0.0.24...0.0.25

0.0.24

What's Changed

• Update README.md contribution guidelines by @michaelfeil in #91
• Update tensorrt, onnxruntime, cuda base by @michaelfeil in #93
• Update dependencies by @NirantK in #96
• Torch dynamic shapes by @michaelfeil in #97
• update poetry version + cache in ci by @michaelfeil in #99
• pydantic upgrade by @michaelfeil in #100
• pydantic-v1-backwards-fixes by @michaelfeil in #101

New Contributors

• @NirantK made their first contribution in #96

Full Changelog: 0.0.23...0.0.24