Hands-on ETL tasks: Nemotron 3.5 Bolt vs Qwen Flash Next each win one round
CulturalKing5623 · reddit · 2026-10-17
A data engineer benchmarked Nemotron 3.5 Bolt against Qwen Flash Next (UD-IQ4XS, llama.cpp) on real work tasks.
Method: plan tasks in Gemini, write a markdown spec-sheet with acceptance criteria, paste it into a fresh pi.dev session, then have Gemini judge both outputs against the spec.
Task 1 (run.sh refactor): Nemotron won — it anchored log paths to SCRIPTDIR (Qwen's $(pwd)/logs breaks when called from /tmp), checked for tee/aws before redirecting output, and dropped exec so shell EXIT traps actually fire. Qwen preserved exec, silently bypassing traps.
Task 2 (orchestrator.py): a script generating daily/monthly time windows and invoking run.sh sequentially. Gemini judged Qwen Flash Next the better implementation.
Takeaway: the two models split wins on everyday engineering tasks, and the post demonstrates a reusable spec-sheet-plus-arbiter comparison workflow.
More from coding & agent
- Dev ditches Gemini 4 Argon for Grok 4.5, which eats 96.6% of his Cursor tokens — gaganghotra_ · 2026-10-17
- Apple releases RSI-Inference Bench: best agent solves just 26.7% of vLLM/SGLang tasks — SumitGup · 2026-10-17
- Supabase ships MIT-licensed Middleware: write fetch logic once, run on any runtime — dshukertjr · 2026-10-17
- Cognition launches Devin Oncall as Opus 5.5 and GPT-6 Astra both ace SREGym-Lite at 100% — tianyin_xu · 2026-10-17
- msgblast: Open-Source Mac App Broadcasts One Prompt to All Your AI Agents — msg · 2026-10-17
- Vibe coder uses Deep Research to reverse-engineer why QuickTime screen recording looks so good — gaganghotra_ · 2026-10-17