惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

博客园_首页
IT之家
IT之家
博客园 - Franky
Stack Overflow Blog
Stack Overflow Blog
宝玉的分享
宝玉的分享
Recent Announcements
Recent Announcements
Engineering at Meta
Engineering at Meta
S
SegmentFault 最新的问题
V
Visual Studio Blog
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
Last Week in AI
Last Week in AI
H
Help Net Security
V
V2EX
H
Hackread – Cybersecurity News, Data Breaches, AI and More
量子位
博客园 - 叶小钗
J
Java Code Geeks
博客园 - 【当耐特】
月光博客
月光博客
爱范儿
爱范儿
人人都是产品经理
人人都是产品经理
酷 壳 – CoolShell
酷 壳 – CoolShell
小众软件
小众软件

Hacker News: Front Page

SPICE simulation → oscilloscope → verification with Claude Code — Lucas Gerads Introducing Claude Opus 4.7 Qwen Studio The Future of Everything is Lies, I Guess: Where Do We Go From Here? GitHub - SeanFDZ/macmind: Single-layer transformer in HyperTalk for the classic Macintosh Show HN: Agent-cache – Multi-tier LLM/tool/session caching for Valkey and Redis Ancient DNA reveals pervasive directional selection across West Eurasia [pdf] AI cybersecurity is not proof of work Moving a large-scale metrics pipeline from StatsD to OpenTelemetry / Prometheus GitHub - Nightmare-Eclipse/RedSun: The Red Sun vulnerability repository GitHub - SethPyle376/hiraeth: Local AWS emulator focused on fast integration testing, with SQS support, SQLite-backed state, and a debug-friendly web UI. A Better Ludum Dare; Or, How to Ruin a Legacy GitHub - macOS26/Agent: Any AI, replaces Claude Code, Cursor, OpenClaw. Over 18 LLM providers (Claude, OpenAI, Gemini, Ollama, Zai, HF, Qwen) wired into a native Mac app that writes code, builds Xcode projects, bumps versions, manages git, automates Safari, use AppleScript, JS or Accessibility, extend Agent! w/ MCP Servers, run tasks from your iPhone via Messages. YouTube now lets you turn off Shorts I Made a Terminal Pager Burgers | マクドナルド公式 Commands — HackerNews CLI documentation ChatGPT for Excel PiCore - Raspberry Pi Port of Tiny Core Linux Live Nation illegally monopolized ticketing market, jury finds Google Broke Its Promise to Me. Now ICE Has My Data. Founding Engineer at Adaptional | Y Combinator CRISPR takes important step toward silencing Down syndrome’s extra chromosome GitHub - saffron-health/libretto: The AI toolkit for building reliable browser automations US v. Heppner (S.D.N.Y. 2026) no attorney-client privilege for AI chats [pdf] Unexpected €54k billing spike in 13 hours: Firebase browser key without API restrictions used for Gemini requests Fragments: April 14 Cal.com Goes Closed Source: Why AI Security Is Forcing Our Decision | Cal.com - Scheduling Software for Online Bookings Laravel raised money and now injects ads directly into your agent Codex Hacked a Samsung TV
Qwen3.6-35B-A3B on my laptop drew me a better pelican tha...
Simon Willison · 2026-04-17 · via Hacker News: Front Page

16th April 2026

For anyone who has been (inadvisably) taking my pelican riding a bicycle benchmark seriously as a robust way to test models, here are pelicans from this morning’s two big model releases—Qwen3.6-35B-A3B from Alibaba and Claude Opus 4.7 from Anthropic.

Here’s the Qwen 3.6 pelican, generated using this 20.9GB Qwen3.6-35B-A3B-UD-Q4_K_S.gguf quantized model by Unsloth, running on my MacBook Pro M5 via LM Studio (and the llm-lmstudio plugin)—transcript here:

The bicycle frame is the correct shape. There are clouds in the sky. The pelican has a dorky looking pouch. A caption on the ground reads Pelican on a Bicycle!

And here’s one I got from Anthropic’s brand new Claude Opus 4.7 (transcript):

The bicycle frame is entirely the wrong shape. No clouds, a yellow sun. The pelican is looking behind itself, and has a less pronounced pouch than I would like.

I’m giving this one to Qwen 3.6. Opus managed to mess up the bicycle frame!

I tried Opus a second time passing thinking_level: max. It didn’t do much better (transcript):

The bicycle frame is entirely the wrong shape but in a different way. Lines are more bold. Pelican looks a bit more like a pelican.

I don’t think Qwen are cheating

A lot of people are convinced that the labs train for my stupid benchmark. I don’t think they do, but honestly this result did give me a little glint of suspicion. So I’m burning one of my secret backup tests—here’s what I got from Qwen3.6-35B-A3B and Opus 4.7 for “Generate an SVG of a flamingo riding a unicycle”:

I’m giving this one to Qwen too, partly for the excellent <!-- Sunglasses on flamingo! --> SVG comment.

What can we learn from this?

The pelican benchmark has always been meant as a joke—it’s mainly a statement on how obtuse and absurd the task of comparing these models is.

The weird thing about that joke is that, for the most part, there has been a direct correlation between the quality of the pelicans produced and the general usefulness of the models. Those first pelicans from October 2024 were junk. The more recent entries have generally been much, much better—to the point that Gemini 3.1 Pro produces illustrations you could actually use somewhere, provided you had a pressing need to illustrate a pelican riding a bicycle.

Today, even that loose connection to utility has been broken. I have enormous respect for Qwen, but I very much doubt that a 21GB quantized version of their latest model is more powerful or useful than Anthropic’s latest proprietary release.

If the thing you need is an SVG illustration of a pelican riding a bicycle though, right now Qwen3.6-35B-A3B running on a laptop is a better bet than Opus 4.7!