Close Menu
NCIJ Network NCIJ Network
    What's Hot

    America’s ranchers are the last holdout against industrial animal agriculture. Is their future at risk?

    November 18, 2026

    ICJ to Deliver Landmark Verdict in Myanmar Genocide Case

    November 18, 2026

    Video of red lights in the sky predates suspected drone incident at RAF Lakenheath – Full Fact

    November 18, 2026
    Facebook X (Twitter) Instagram
    Trending
    • America’s ranchers are the last holdout against industrial animal agriculture. Is their future at risk?
    • ICJ to Deliver Landmark Verdict in Myanmar Genocide Case
    • Video of red lights in the sky predates suspected drone incident at RAF Lakenheath – Full Fact
    • Deutschland und Österreich fordern europäischen Eigentümer für TikTok
    • New LoRA Adapter Lifts Qwen3.8 Coding Success From 39% to 65%
    • Two stars in the Andromeda galaxy merged before astronomers’ eyes
    • Orlen spreads its wings across Central and Eastern Europe with new gas pacts
    • It’s hard getting into farming. But for many, retiring is harder.
    • About
      • Our Team
      • Editorial Policy
      • Editorial Independence
      • International Support
    • Trust & Standards
      • AI Usage Policy
      • Conflict of Interest Policy
      • Corrections Policy
      • Ethics Policy
      • Fact-Checking Policy
      • Source Protection
    • Get Involved
      • Guide for Sources
      • Support Independent Journalism
    • Legal
      • Cookie Policy
      • Privacy Policy
      • Terms of Use
    Facebook X (Twitter) Instagram
    NCIJ Network NCIJ Network
    Wednesday, November 18
    • Home
    • World
    • Ai
    • Business
    • Politics
    • Health
    • Crypto
    • Science
    • Technology
    • Cybersecurity
    • Defense & Security
    • Economy
    • Energy
    • Europe
    • More
      • Fact Check
      • Investigations
      • Opinion & Analysis
      • Environment
    NCIJ Network NCIJ Network
    Home»Artificial Intelligence

    New LoRA Adapter Lifts Qwen3.8 Coding Success From 39% to 65%

    NCIJ NETWNCIJ NETWORKBy NCIJ NETWNCIJ NETWORKNovember 18, 2026 Artificial Intelligence No Comments4 Mins Read
    Share
    Facebook Twitter LinkedIn Pinterest Email

    Hugging Face’s harness-hub team has released Qwen3.8-27B-pi-adapted, a LoRA adapter for Qwen3.8-27B. It was trained with reinforcement learning to work inside the pi coding agent. The Qwen3.8-27B-pi-adapted LoRA adapter targets one harness instead of general chat. For developers running local coding agents, that is the point. You keep the open base model and swap in a small, harness-specific adapter.

    TL;DR

    • Size: LoRA rank 32 (alpha 64) on Qwen3.8-27B, a 27B dense hybrid with 262,144 native context. Merged BF16 checkpoint is about 54 GB.
    • Runs on: vLLM with LoRA enabled, or llama.cpp with a GGUF adapter on a Q8_0 base. Inference VRAM is not disclosed. Training used 2x NVIDIA B200.
    • Performance: Large pass@1 gains inside pi with thinking off, plus far shorter sessions.
    • Best: 64.7% on SWE-bench Pro (hard 51), up from 39.2%.
    • Worst: 55.8% on Terminal-Bench 2.1. The base model with thinking already reaches 64.7% on SWE-bench Pro.
    • Bottom line: Best: thinking-level accuracy at about 4.5x fewer tokens. Worst: small eval sets, self-reported, pi only.

    What is Qwen3.8-27B-pi-adapted?

    Qwen3.8-27B-pi-adapted is a PEFT LoRA adapter that turns Qwen3.8-27B into a model tuned for pi. pi is an MIT-licensed agent toolkit with a coding agent CLI. The adapter is released under Apache-2.0. It applies to the q, k, v, o, gate, up and down projections. It only works on Qwen3.8-27B.

    The base model is a dense hybrid. It alternates Gated DeltaNet linear-attention blocks with Gated Attention blocks across 64 layers.

    How does the harness adapter training work?

    The method comes from Hugging Face’s harness adapters post from November 6. The idea is 1 LoRA adapter per model and harness. Training runs inside the unmodified harness on verifiable tasks.

    For this checkpoint, the model card lists these details:

    • Algorithm: AsyncGRPO from TRL, 8 rollouts per task at temperature 0.8.
    • Tasks: 750 scale800 tasks interleaved with 300 TermGrade tasks.
    • Reward: verifier correctness plus a tool-efficiency bonus with weight 0.1.
    • Budget: up to 50 agent steps and 1,800 seconds per rollout, thinking off.
    • Checkpoint: step 160. The run stopped at step 191 of 1,000 planned.

    Every tool call passes through safeagents, a rule-based proxy. It blocks calls such as fetching benchmark answers from the Hub. Blocked rollouts score 0, so RL pushes the model away from reward hacking.

    Benchmarks: how much does the LoRA adapter help?

    All numbers below are self-reported. They use the pi harness, 1 attempt per task.

    Benchmark (pi, thinking off) Base Qwen3.8-27B pi-adapted Change
    SWE-bench Pro (hard 51) 39.2% 64.7% +25.5 ± 14.2 pts (p = 0.0023)
    Terminal-Bench 2.1 (86 tasks) 39.5% 55.8% +16.3 ± 9.1 pts (p = 0.0013)

    The adapter solved 15 new SWE-bench Pro tasks and lost 2. On Terminal-Bench 2.1, it gained 16 and lost 2.

    Efficiency is the larger story. On tasks both models solved, median turns fell from 59 to 23 on SWE-bench Pro. On Terminal-Bench 2.1, they fell from 18 to 8. The adapter also generates about 4.5x fewer tokens than the base model with thinking.

    adapter Coding lifts LoRA Qwen3.8 success
    NCIJ NETWNCIJ NETWORK
    • Website

    Keep Reading

    Anthropic Adds Code Mode to Claude Code: Claude Now Chains Tool Calls in JavaScript

    SID AI Releases SID-2 Agentic Search Model With 0.720 NDCG Swarm Mode

    Clearing Up the Confusion on AI Data Retention: Two Standard Designs Explained

    Perplexity Computer Adds Multiagent: Up to 16 AI Agents That Cross-Check Work

    OpenAI Releases GPT-6.2 Astra and Sol With a 20% Cheaper Astra API

    Astra for Law leads in legal intelligence

    Add A Comment
    Leave A Reply Cancel Reply

    Editors Picks

    America’s ranchers are the last holdout against industrial animal agriculture. Is their future at risk?

    November 18, 2026

    ICJ to Deliver Landmark Verdict in Myanmar Genocide Case

    November 18, 2026

    Video of red lights in the sky predates suspected drone incident at RAF Lakenheath – Full Fact

    November 18, 2026

    Deutschland und Österreich fordern europäischen Eigentümer für TikTok

    November 18, 2026
    Latest Posts

    After one week wearing the Pixel Watch 5, here’s what I want most

    August 21, 2026

    Targeted by an Islamophobic Provocateur, a City’s Muslims Are on Edge

    August 21, 2026

    Back on track: can Burnham fix the north’s woeful rail services? | Rail industry

    August 21, 2026

    Subscribe to News

    Get the latest sports news from NewsSite about world, sports and politics.

    NCIJ Network is an independent digital news platform delivering trusted investigative journalism, European and global news, in-depth analysis, and fact-based reporting with accuracy, transparency, and integrity.

    Facebook X (Twitter) Instagram Pinterest YouTube

    America’s ranchers are the last holdout against industrial animal agriculture. Is their future at risk?

    November 18, 2026

    ICJ to Deliver Landmark Verdict in Myanmar Genocide Case

    November 18, 2026

    Video of red lights in the sky predates suspected drone incident at RAF Lakenheath – Full Fact

    November 18, 2026

    Subscribe to Updates

    Get the latest creative news from FooBar about art, design and business.

    Type above and press Enter to search. Press Esc to cancel.