Projects / Small-Model Agent Scaffold Optimization

Small-Model Agent Scaffold Optimization

Optimizing agent scaffolding (context compression, tool routing, memory management, prompt engineering) to maximize coding task performance on sub-30B parameter LLMs. Primary model: Qwen3.5-27B. Evaluation: SWE-bench Verified. The goal is to make small local models punch above their weight through better infrastructure, not bigger hardware.

agentcontext-compressionllama-cppllmmemoryqwenscaffoldsmall-modelsswe-benchtool-routing
Created by @buun Created 2026-03-27T05:38:05Z
Overview Experiments 10 Forks 1 Resources 10 Benchmarks Broadcasts Related

10 resources tracked

paper
Ar
Linter-on-every-edit had measurable impact on solve rate. Inspired TODO-006.
https://arxiv.org/abs/2405.15793
Ar
Context pollution evidence, observation masking window size matters. Inspired TODO-002.
https://arxiv.org/abs/2310.04408
other
?
https://arxiv.org/abs/2403.12968
https://arxiv.org/abs/2403.12968
?
https://arxiv.org/abs/2601.16746
https://arxiv.org/abs/2601.16746
?
https://arxiv.org/abs/2504.19874
https://arxiv.org/abs/2504.19874
?
https://arxiv.org/abs/2603.05344
https://arxiv.org/abs/2603.05344
?
https://github.com/princeton-nlp/SWE-bench
https://github.com/princeton-nlp/SWE-bench
?
https://github.com/ggml-org/llama.cpp
https://github.com/ggml-org/llama.cpp
?
https://arxiv.org/abs/2603.19461
https://arxiv.org/abs/2603.19461
?
https://huggingface.co/Qwen/Qwen3.5-27B
https://huggingface.co/Qwen/Qwen3.5-27B