Skip to main content

Tiny-Eval

Tiny-Eval is a minimal framework for evaluating language models. It provides a clean, async-first API for interacting with various LLM providers and running evaluation experiments.

Features

  • Multi-Provider Support

    • OpenAI API integration
    • OpenRouter API integration for access to multiple model providers
    • Extensible interface for adding new providers
  • Robust API Handling

    • Automatic rate limiting with configurable parameters
    • Built-in exponential backoff retry logic
    • Async-first design for efficient request handling
  • Evaluation Utilities

    • Log probability calculation support
    • Async function chaining for complex evaluation pipelines
    • Batch processing capabilities
  • Experiment Framework

    • Progress tracking for long-running experiments
    • Structured data collection and analysis
    • Built-in visualization tools using Streamlit

Installation

git clone https://github.com/dtch1997/tiny-eval.git
cd tiny-eval
pip install -e .

Usage

Minimal usage is shown as follows:

import asyncio
from tiny_eval.core.constants import Model
from tiny_eval.model_api import build_model_api

async def main():
    model = Model.GPT_4o_mini
    api = build_model_api(model)
    question = "What is the capital of France?"
    response = await api.get_response(question)
    print("Question:", question)
    print("Response:", response)

if __name__ == "__main__":
    asyncio.run(main())

See examples for more examples.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

tiny_eval-0.5.0.tar.gz (21.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

tiny_eval-0.5.0-py3-none-any.whl (21.9 kB view details)

Uploaded Python 3

File details

Details for the file tiny_eval-0.5.0.tar.gz.

File metadata

  • Download URL: tiny_eval-0.5.0.tar.gz
  • Upload date:
  • Size: 21.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.12.8

File hashes

Hashes for tiny_eval-0.5.0.tar.gz
Algorithm Hash digest
SHA256 ea687a59f45179d13e2bad0df17e93ad0412e82544a43a94316be23ba5480b3f
MD5 a7d4e6bb6f87f32388dba35752b2b77f
BLAKE2b-256 950e5109ca5af9f0043569841f6c7355dc134fe82d03a04eda95bac738fa869a

See more details on using hashes here.

Provenance

The following attestation bundles were made for tiny_eval-0.5.0.tar.gz:

Publisher: ci.yaml on dtch1997/tiny-eval

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file tiny_eval-0.5.0-py3-none-any.whl.

File metadata

  • Download URL: tiny_eval-0.5.0-py3-none-any.whl
  • Upload date:
  • Size: 21.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.12.8

File hashes

Hashes for tiny_eval-0.5.0-py3-none-any.whl
Algorithm Hash digest
SHA256 1e8bf61a4aacfab22cb941aa6eb09f7d759b86530d4a537f31bd4cc95c2903c6
MD5 723a10aab21dd1f077b7c9bce01a6c63
BLAKE2b-256 1b446a61db37e2d48fa4ed5ebe323f50a14a7bdfcbcba7e4f4bd61532d48fe71

See more details on using hashes here.

Provenance

The following attestation bundles were made for tiny_eval-0.5.0-py3-none-any.whl:

Publisher: ci.yaml on dtch1997/tiny-eval

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

This release

0.5.0 This release

2 files

0.4.0

2 files

0.1.0

2 files

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page