aime-api-server-0.9.0

Changes

Version 0.9.0

• Llama 3.1 example endpoint added
• FLUX.1-Dev example endpoint added, worker implementation: https://github.com/aime-labs/flux

aime-api-server-0.8.5

Changes

Version 0.8.5

• Stable Diffusion 3 (sd3) example endpoint added
• Tortois TTS example endpoint added
• Improved and optimized ffmpeg image and audio conversion
• Input parameters attribute 'keep_aspect_ratio' added as option how to rescale images
• Llama3 and Mixtral endpoints: added max_gen_tokens input parameter

aime-api-server-0.8.1

Changes

Version 0.8.1

• New input parameter type 'json', for submitting escaped raw JSON data to endpoint, added
• Llama3 (llama3_chat) instruct chat example endpoint added
• Mixtral 8x7B, 8x22B (mixtral_chat) instruct chat example endpoint added
• Improved authentication mechanism
• improved benchmark tool to benchmark instruct chat models throughput

aime-api-server-0.8.0

AIME API Server - The Scalable Model Inference API Server

With AIME API one deploys deep learning models (Pytorch, Tensorflow) through a job queue as scalable API endpoint capable of serving millions of model inference requests.

Turn a console Python script to a secure and robust web API acting as your interface to the mobile, browser and desktop world.

The central part is the AIME API Server, an efficient asynchronous HTTP/HTTPS web server which can be used stand-alone web server or integrated into Apache, NGINX or similar web servers. It takes the client requests, load balances the requests and distributes them to the API compute workers.

Features

• Fast - asynchronous and multi process API server
• Scalable & Robust- distributed cluster ready architecture
• Secure - type safe interface and input validation
• Aggregates API requests to GPU batch jobs for maximum throughput
• Easy integratable into exisiting Python and Tensorflow projects
• High performance image and audio input/ouput conversion for common web formats
• Pythonic - easily extendable in your favourite programming language

Changes

Version 0.8.0

Initial Release