Token optimization and compression for Claude API requests
Project description
Tailspin AI
Token optimization and compression for Claude API requests. Reduces token usage while maintaining context quality through intelligent compression, caching, and context management.
Installation
pip install tailspin-ai
Usage
See main Tailspin repository for documentation and examples.
Features
- Smart Compression: Automatic token optimization for text, code, and logs
- Semantic Caching: Cache-aware compression that preserves prefix stability
- LiteLLM Integration: Route requests to any LLM provider via LiteLLM
- Context Management: Intelligent context windowing and summarization
- Compression Pipeline: Extensible transform pipeline for custom compression logic
Configuration
Configure via environment variables or config.py:
litellm_enabled: bool = False
litellm_endpoint: str = ""
litellm_api_key: str = ""
litellm_provider: str = "hosted_vllm"
Contributing
Contributions welcome. See main repository.
License
MIT
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file tailspin_ai-0.5.29.tar.gz.
File metadata
- Download URL: tailspin_ai-0.5.29.tar.gz
- Upload date:
- Size: 912.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.11.15
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
d5d6ae8437cf211795927bcfb4f176761039e981674c14dd40caebc85378c105
|
|
| MD5 |
6186652ccf56e55d4b5c63058010c19d
|
|
| BLAKE2b-256 |
cec17868a057b28b0a5e0726882c4f0e5345f4336612311d76ba5b7060efbd8b
|
File details
Details for the file tailspin_ai-0.5.29-py3-none-any.whl.
File metadata
- Download URL: tailspin_ai-0.5.29-py3-none-any.whl
- Upload date:
- Size: 1.1 MB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.11.15
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
24b0cf9a9fb45c738348f59815d967d121c03dc846ee7f8c6554f6c0a4af601a
|
|
| MD5 |
6517ac61472dc4704e144652f9fd4bcf
|
|
| BLAKE2b-256 |
fc90d4b682ceae9f3abcdc10d5de7325086b2978e0f758a8f2288a4733c51912
|
