Skip to content
Live
Trending storyDeveloping

Hugging Face releases LoRA adapter to improve Qwen3.8 programming success rate

1 reports1 outlets5 hours agoUpdate

The story so far

AI overview

Hugging Face has released the Qwen3.8-27B-pi-adapted LoRA adapter, which uses reinforcement learning to fine-tune the Qwen3.8-27B model, increasing Qwen3.8's SWE-bench Pro (hard 51) score in pi programming agents from 39.2% to 64.7%, while reducing inference token usage by about 4.5 times.

AI-written from the reports · Updated 5 hours ago

Coverage timeline

Follow the coverage to see every side of the story.

19 Nov
  1. MarkTechPost
    New LoRA Adapter Lifts Qwen3.8 Coding Success From 39% to 65%

    Hugging Face releases Qwen3.8-27B-pi-adapted LoRA adapter, which fine-tunes the Qwen3.8-27B model via reinforcement learning, boosting its SWE-bench Pro (hard 51) score from 39.2% to 64.7% while reducing inference token usage by about 4.5 times

Heat over time

Now
9
Peak
1019 Nov, 01:00
24h change
–

Hover or tap the chart to see each hour; use the arrow keys on a keyboard.