Hacker News: xjconlyme

New comment by xjconlyme in "Ask HN: What Are You Working On? (April 2026)"

xjconlyme — Mon, 13 Apr 2026 10:43:05 +0000

https://kulikuli.app/translate I'm building a web version of my app, which has better quality than Google/DeepL and also translates the result in reversed for users to double check.

Show HN: Eimi – Turn any information into daily cards

xjconlyme — Wed, 31 Dec 2025 15:56:35 +0000

Hi HN,

I’m the developer behind Eimi (https://eimi.tech/).

The Problem: I found myself constantly distracted. I wanted to learn new words and keep up with specific news, but every time I opened an app, I ended up doomscrolling for 20 minutes. The signal-to-noise ratio was too low.

The Solution: I built Eimi to "calm down" my consumption. It’s a card-based app that delivers content depends on your personal needs. Instead of an infinite feed, you get curated highly customizable cards.

Use Cases:

Language Learning: Instead of flashcards without context, it generates cards with usage examples for words I'm trying to learn.

News/Market Watch: I hook up my interests (e.g., tech news, specific stocks), and it generates a summary card. No clickbait, just the gist.

It's currently in Public Beta. I’d love to hear your thoughts on the UX and what other "card types" you’d find useful.

Thanks!

Comments URL: https://news.ycombinator.com/item?id=46445248

Points: 2

# Comments: 0

Show HN: YPerf – Discover AI apps and their models based on token usage

xjconlyme — Tue, 06 May 2025 11:18:37 +0000

I've created a leaderboard using public data from OpenRouter. It tracks new AI apps based on their token usage and also highlights the top 3 models these apps are adopting and detailed model stats.

The goal is to provide a quick overview of what's gaining traction and which new models are being integrated into real-world applications.

You can check it out here: https://yperf.com

Would love to hear your thoughts or any suggestions!

Comments URL: https://news.ycombinator.com/item?id=43903825

Points: 1

# Comments: 0

New comment by xjconlyme in "Show HN: YPerf – Compare LLM models with performance/cost/uptime metrics"

xjconlyme — Fri, 03 Jan 2025 02:04:05 +0000

It's based on the usage data of the model with multiple providers from OpenRouter.

Show HN: YPerf – Compare LLM models with performance/cost/uptime metrics

xjconlyme — Thu, 02 Jan 2025 13:00:53 +0000

I've iterated YPerf since last submission with features to make LLM selection more data-driven:

- Added benchmark rankings from lmarena.ai to quantify model intelligence - Built-in cost estimator to forecast average usage expenses - Real-time uptime monitoring across providers - Side-by-side provider comparisons for identical models

The goal is to help teams make more informed decisions when choosing LLMs by providing transparency around performance, reliability and costs.

Comments URL: https://news.ycombinator.com/item?id=42574040

Points: 2

# Comments: 2

New comment by xjconlyme in "Show HN: YPerf – Monitor LLM Inference API Performance"

xjconlyme — Mon, 23 Dec 2024 11:11:15 +0000

[Updated] I've added the filtering and the multiple provider comparison!

---

Great suggestion!

I currently pick the fastest TPS among providers, and you can see a detail performance list by clicking the icon at the last column. For example, here's the detailed OR page of Llama 3.3: https://openrouter.ai/meta-llama/llama-3.3-70b-instruct

I'll add the filtering soon.

Show HN: YPerf – Monitor LLM Inference API Performance

xjconlyme — Mon, 23 Dec 2024 06:32:46 +0000

Our team operates several real-time AI applications, where both latency(TTFT) and throughput(TPS) are critical to most of our users. Unfortunately, nearly all of the major LLM APIs lack consistent stability.

To address this, I developed YPerf—a simple webpage designed to monitor the performance of inference APIs. I hope it helps you select better models and discover new trending ones as well.

The data is sourced from OpenRouter, an excellent provider that aggregates LLM API services.

Comments URL: https://news.ycombinator.com/item?id=42492300

Points: 2

# Comments: 2

Show HN: WTranslate – A translator shows reverse translations for double check

xjconlyme — Mon, 02 Dec 2024 08:30:38 +0000

I built WTranslate, a simple translation tool that helps users verify translation accuracy by showing both the forward and reverse translations simultaneously.

When you translate text, it shows: 1. The original text 2. The translation to your target language 3. A reverse translation back to the original language

This helps catch potential translation errors or ambiguities quickly.

I built this because I often found myself doing manual back-translations to verify accuracy when using other translation services.

Would love to get the community's feedback on the usefulness of this approach and suggestions for additional features.

Comments URL: https://news.ycombinator.com/item?id=42294263

Points: 1

# Comments: 0

Show HN: A GPT for answering questions about Gemini

xjconlyme — Thu, 07 Dec 2023 05:39:36 +0000

Article URL: https://chat.openai.com/g/g-fkNFBS0HL-everything-about-gemini

Comments URL: https://news.ycombinator.com/item?id=38553232

Points: 2

# Comments: 0

New comment by xjconlyme in "Show HN: Knit – A Better LLM Playground"

xjconlyme — Tue, 08 Aug 2023 12:56:51 +0000

thx! Prompting is still hard, people need better tools.

Show HN: Knit – A Better LLM Playground

xjconlyme — Tue, 08 Aug 2023 11:33:42 +0000

Knit was created to solve pains of other LLM playgrounds.

Some of the highlights:

- Smart prompt builder, create prompt with simple requirement and few shot learning, fast and effortlessly.

- Function call simulation, visualize the function callings and you can also setup a mocked value to return.

- Support OpenAI/Anthropic/Azure models.

- Manage prompts with projects and members.

- And so much more!

I have been developing Knit by myself for over 4 months now, and am looking for ways to improve it. Any feedback is appreciated.

Comments URL: https://news.ycombinator.com/item?id=37047402

Points: 14

# Comments: 4

New comment by xjconlyme in "Show HN: Open-Source AI Playground"

xjconlyme — Tue, 18 Jul 2023 03:32:32 +0000

I'm daily user of ChatHub, author has updated a lot of new features in the past months. I think it's totally fine to share it again on HN.

New comment by xjconlyme in "Show HN: Knit – A prompt managing tool, now supports both OpenAI and Claude"

xjconlyme — Wed, 07 Jun 2023 16:05:46 +0000

Not yet, it’s on the roadmap, Knit will be the home for all of your prompts.

Show HN: Knit – A prompt managing tool, now supports both OpenAI and Claude

xjconlyme — Wed, 07 Jun 2023 13:40:50 +0000

Knit supports Claude API now, you can also design and debug your prompts with the new multi-variable feature.

Comments URL: https://news.ycombinator.com/item?id=36226678

Points: 2

# Comments: 2

Show HN: Knit, a SaaS for teams to build better prompts

xjconlyme — Mon, 29 May 2023 03:59:53 +0000

I've seen people store, edit and share prompts in Notes/Sheets and copy-paste in different IMs, so I build a SaaS for teams to collaborate on prompt development.

Comments URL: https://news.ycombinator.com/item?id=36110506

Points: 2

# Comments: 0

New comment by xjconlyme in "Gordon Moore has died"

xjconlyme — Sat, 25 Mar 2023 12:31:31 +0000

Thank you Mr.Moore, RIP.

New comment by xjconlyme in "Show HN: Use ChatGPT, Bing and Bard in one app"

xjconlyme — Fri, 24 Mar 2023 13:08:31 +0000

Let them chat.

New comment by xjconlyme in "Show HN: Chrome extension to display ChatGPT response besides Google Search"

xjconlyme — Sun, 04 Dec 2022 15:23:18 +0000

It’s interesting to see what’s Google’s next move. laMDA was also impressive.

New comment by xjconlyme in "Mechanical Watch"

xjconlyme — Thu, 05 May 2022 08:12:58 +0000

This one is insanely good.

New comment by xjconlyme in "Show HN: Tellery: an open-source Notion-like tool for analytics"

xjconlyme — Thu, 08 Jul 2021 11:18:02 +0000

This looks interesting. More like a conjunction of notion and metabase, I often use metabase and always want to write and share posts with synced data charts. Tellery looks like a solution for this. Will give it a shot.