State of the Art of Coding Models, According to Hacker News Commenters
Details
- External ID
- 47990708
- Source
- HN
- Company
- —
- Product
- State of the Art of Coding Models, According to Hacker News Commenters
- Website domain
- hnup.date
- Launched
- May 2, 2026
- Cohort
- —
- Upvotes
- 168
- Upvotes percentile
- 0.9507269789983845
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
Hello HN,I was away from my computer for two weeks, and after coming back and reading the latest discussions on HN about coding assistants (models, harnesses), I felt very out of the loop. My normal process would have been to keep reading and figure out the latest and greatest from people's comments, but I wanted to try and automate this process.Basically the goal is to get a quick overview over which coding models are popular on HN. A next iteration could also scan for harnesses that people use, or info on self-hosting or hardware setups.I wrote a short intro on the page about the pipeline that collects and analyzes the data, but feel free to ask for more details or check the Google Sheet for more info.https://hnup.date/hn-sota
Enrichment
- Theme
- Hacker News clients, datasets, and tools
- Vertical
- Horizontal
- Function
- Analytics & BI
- Audience
- Developer
- AI stance
- Not AI
- Project type
- Hobby / open-source project
- Normalized one-liner
- analysis of coding models discussion
- Manually corrected
- False
Could you build this?
Yes This is a content synthesis / data analysis project aggregating Hacker News comments via the Firebase/Algolia API and running summarization prompts with an LLM.
Discussion
20 comments analyzed.
Competitors mentioned: Claude, Claude Code, Codex, GPT-5.5, Kimi
Concerns raised: Gemini may have model bias influencing ratings, Claude fails with shell commands on Windows (PowerShell/cmd permissions issues), Claude Code uses significantly more tokens than alternatives, LLMs struggle with complex, poorly-documented integrations and hallucinate frequently, Unpredictable and systematic performance degradation at different times/demand levels
Feature requests: Show the prompt used for model evaluations, Normalize model versions (e.g., 'opus-all') to show trendlines over time, Improve long-running agent capabilities, Better healing mechanisms for common LLM errors (thinking loops, syntax errors), Easy Python library swap support for open-source model cloud providers
Competitors
Other products that read as similar to this one — 103 launches clear the similarity bar, closest 8 shown.
Attention rank: #14 of 104 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 183 days after the earliest competitor.
- Was tired of drowning in HN comments, so I built an AI Chief of Staff · hn · 2026-01-26 · 5 upvotes · similarity 0.49
- I built an interactive HN Simulator · hn · 2025-11-24 · 538 upvotes · similarity 0.49
- HN.watch · hn · 2026-09-28 · 212 upvotes · similarity 0.49
- I built a tool that helps predict HN front page success · hn · 2026-05-03 · 26 upvotes · similarity 0.48
- Hacker Atlas · hn · 2026-09-25 · 86 upvotes · similarity 0.47
- Hackobar · hn · 2026-05-25 · 5 upvotes · similarity 0.46
- 20 years of Hacker News discussions, clustered and visualized · hn · 2026-03-22 · 7 upvotes · similarity 0.45
- How much of Hacker News is about AI? · hn · 2026-08-26 · 74 upvotes · similarity 0.43
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a analytics & bi tool for Legal yet.