FastFlowLM/FastFlowLM v0.1.8

## ⚡ FastFlowLM v0.1.8 — Faster Decoding & User Agreements Added

on GitHub

FastFlowLM v0.1.8 delivers a performance boost and introduces official usage agreements for better transparency.

✨ What's New

🚀 1. Improved Decoding Performance

• Optimized decoding pipeline.
• Achieves up to 20% faster generation speed depending on model and prompt length.

📄 2. User Agreements Introduced

• FastFlowLM now includes User Agreements to clarify usage conditions and ensure proper compliance.
• Terms are displayed upon installation.

✅ Usage Examples

Interactive CLI:

flm run qwen3:1.7b

Server Mode:

flm serve

Model Management:

flm help
flm version

This release improves user experience with faster inference and sets the foundation for responsible usage through clearly defined terms.