Skip to main content

Deterministic cost-bounded execution for multi-step LLM workflows

Project description

Execution Constraint Engine (ECE)

Runtime decision layer for enforcing cost constraints in multi-step LLM workflows.

ECE evaluates projected cost before each step and returns a BLOCK decision when execution would exceed a defined budget.


Install

pip install veloryn

Minimal usage

from veloryn import execute

def run_task():
    return {
        "output": "model response",
        "usage": {
            "input_tokens": 120,
            "output_tokens": 300,
        },
    }

execute(
    task_id="demo",
    run_task=run_task,
    input_data="Analyze input",
    model="gpt-4o",
    max_cost=0.05,
)

run_task must return usage (input/output tokens).
ECE does not infer cost automatically.


Full documentation and examples:
https://github.com/veloryn-intel/execution-constraint-engine

Project details


Download files

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

Source Distribution

veloryn-0.1.1.tar.gz (7.9 kB view details)

Uploaded Source

Built Distribution

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

veloryn-0.1.1-py3-none-any.whl (9.5 kB view details)

Uploaded Python 3

File details

Details for the file veloryn-0.1.1.tar.gz.

File metadata

  • Download URL: veloryn-0.1.1.tar.gz
  • Upload date:
  • Size: 7.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.3

File hashes

Hashes for veloryn-0.1.1.tar.gz
Algorithm Hash digest
SHA256 3456c1fdfefa5020bc6b5ec3fb231cb0435ff571d18736ab19d90d859192c263
MD5 8645acb7618bf240571bfe06f29faaf3
BLAKE2b-256 34f28377492a1e6a75006071947221cacf02c88015a1c05031f81fe136a756a5

See more details on using hashes here.

File details

Details for the file veloryn-0.1.1-py3-none-any.whl.

File metadata

  • Download URL: veloryn-0.1.1-py3-none-any.whl
  • Upload date:
  • Size: 9.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.3

File hashes

Hashes for veloryn-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 ac9a1724bfbc4bddb72210d65cbbaed3bd8310eaabc999758361e6173986bc04
MD5 4f5a0fc80921c8d6a7ade1549eefcb65
BLAKE2b-256 ed6c55aeb3ea12ae1000aacf78d8bc3de270a0bbf8f390b3aa4c79bded2c5e72

See more details on using hashes here.

Supported by

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