I built a guide to the “Best LLMs for Coding” using 11 benchmark boards
DataLearnerAI · reddit · 2026-08-30
A new guide aggregates 11 benchmark boards to rank the best LLMs for coding, covering 98 model series. The methodology splits evidence into repository engineering, agentic coding, live coding, and function generation, converting ranks to percentiles rather than averaging raw metrics. It includes data from SWE-bench, LiveCodeBench, and others, and the author is seeking feedback on benchmark selection and adding local deployment metrics.
More from Models
- Claude's Safety Downgrade Backfires: Wipes User's 700GB Home Directory — 机器之心 · 2026-08-30
- Qwen 350K Context Tested on M5 Max: Performance and Quality — Artistic_Okra7288 · 2026-08-30
- Gemini 3.7 Flash and GPT 5.6 Luna ranked best for automation tasks — burkov · 2026-08-30
- GLM-5.3 vs. Flash: 17x Price Difference and Usage Strategy — togethercompute · 2026-08-30
- Grok $300/Month Subscriber Reports Hitting Only 10% of Weekly Usage Cap — AaronBergman18 · 2026-08-30
- Benchmarking models by hand takes forever, but I care about the data and model welfare — cephaloform · 2026-08-30