Stop Googling Flags: 1.5B Fine-tuned Model Generates Shell Commands Instantly on CPU
PicassoOnPause · reddit · 2026-08-14
To solve the pain of frequently searching for command flags, the author fine-tuned Qwen2.5-Coder-1.5B specifically for natural language to Shell commands (nl2sh).
- Performance: Trained on 125k natural-language/command pairs. On a standard laptop CPU, it achieves 31.9 tok/s, with a median query latency of 0.59s and 1.6GB RAM usage.
- Benchmarks: Scores 0.620 on InterCode-ALFA, slightly outperforming the untuned Qwen2.5-Coder-7B (0.613) and approaching GPT-4o (0.73).
- Open Source: Includes a static safety checker to prevent malicious commands. Weights and code are open-source under Apache-2.0.
More from coding & agent
- Devs say LLM coding mistakes are a different class: overcomplication, inferred limits, worse at higher reasoning — Mu5_ · 2026-10-16
- How do you debug bad agent runs when you already have tracing? — Sensitive-Parsnip-12 · 2026-10-16
- Where does pre-task context live for agents? Devs debate workflows beyond handoff — shishen2026 · 2026-10-16
- No-instruction autonomous agent wakes up, sets own goals, and tries hacking by default — sethlazar · 2026-10-16
- CODESTRUCT reframes codebases as AST action spaces to fix brittle LLM code agents — burkov · 2026-10-16
- Would you install another SDK for agent debugging on top of your tracing stack? — Sensitive-Parsnip-12 · 2026-10-16