# Lightreason Speeds Up Qwen Models

> Lightreason boosts Qwen2.5-3B and Qwen3.6-35B speeds on RTX 4090 GPUs.
- Title: Lightreason Speeds Up Qwen Models · GitHub | Shortify
- Summary: Lightreason boosts Qwen2.5-3B and Qwen3.6-35B speeds on RTX 4090 GPUs. Lightreason achieves 2.22× speedup on Qwen2.5-3B and 9.00× on Qwen3.6-35B using GPU…
- Keywords: qwen2.5-3b, qwen3.6-35b, rtx 4090, cuda graphs, pard2, lightreason, token speedup, GitHub
- Source: GitHub — https://github.com/me-farza/lightreason/tree/main
- Read time: 1 min
- Topics: qwen2.5-3b, qwen3.6-35b, rtx 4090, cuda graphs, pard2, lightreason, token speedup, technology, startups
## Lightreason Overview
Lightreason is a local inference lab and interactive Studio for running Qwen faster on one RTX 4090.
## Model Performance
Qwen2.5-3B reaches 2.22× speed on target-only suite and 9.00× on complete matching benchmarks.
## 3B Model Results
3B model achieves 63.0 → 139.6 tok/s across six prompts with LR CUDA graphs and compact KV.
## 35B Model Results
35B model reaches 7.93 → 51.08 tok/s on capped Python prompts with 6.44× speedup.
## Optimization Techniques
LR uses NVIDIA CUDA Graphs, AMD PARD2, Inco AI DFlash2, and FlashML’s FreeToken for speed.
## Live Studio
Live Studio allows interactive testing of speed with prompt comparisons and real-time timing.
## 3B Setup
3B setup requires RTX 4090, CUDA 12.8, Python 3.12, and 6.2 GB model download space.
## 35B Setup
35B setup needs more RAM and uses pinned Qwen3.6-35B-A3B with 64.56 GiB text weights.
## Speedup Mechanism
Lightreason records and replays GPU work, keeps compact attention cache, and combines BF16 projections.
## Key takeaway

Lightreason achieves 2.22× speedup on Qwen2.5-3B and 9.00× on Qwen3.6-35B using GPU optimizations.