Hacker News: xyc

New comment by xyc in "Open models by OpenAI"

xyc — Wed, 06 Aug 2025 00:09:10 +0000

I'm running it with ROG Flow Z13 128GB Strix Halo and getting 50 tok/s for 20B model and 12 tok/s for 120B model. I'd say it's pretty usable.

New comment by xyc in "MCP in LM Studio"

xyc — Fri, 27 Jun 2025 17:29:49 +0000

For TypeScript you can refer to https://github.com/modelcontextprotocol/typescript-sdk/blob/...

There isn't much documentation available right now but you can ask coding agent eg. Claude Code to generate an example.

New comment by xyc in "MCP in LM Studio"

xyc — Wed, 25 Jun 2025 23:19:52 +0000

Great to see more local AI tools supporting MCP! Recently I've also added MCP support to recurse.chat. When running locally (LLaMA.cpp and Ollama) it still needs to catch up in terms of tool calling capabilities (for example tool call accuracy / parallel tool calls) compared to the well known providers but it's starting to get pretty usable.

New comment by xyc in "LM Studio is now an MCP Host"

xyc — Wed, 25 Jun 2025 23:13:54 +0000

It's a protocol that doesn't dictate how you are calling the tool. You can use in-memory transport without needing to spin up a server. Your tool can just be a function, but with the flexibility of serving to other clients.

New comment by xyc in "Ask HN: What Does Your Self-Hosted LLM Stack Look Like in 2025?"

xyc — Thu, 05 Jun 2025 17:19:21 +0000

recurse.chat + M2 max Mac

New comment by xyc in "Cursor 1.0"

xyc — Wed, 04 Jun 2025 23:16:43 +0000

I recently discovered toolhive which is pretty handy too https://github.com/stacklok/toolhive

New comment by xyc in "Vision Now Available in Llama.cpp"

xyc — Sat, 17 May 2025 00:39:10 +0000

If you are on a Mac, give https://recurse.chat/ a try. As simple as download the model and start chatting. Just added the new multimodal support in LLaMA.cpp.

New comment by xyc in "Show HN: Clippy – 90s UI for local LLMs"

xyc — Tue, 06 May 2025 19:58:37 +0000

Actually this is a good way to find product ideas. I placed a query in Grok to find posts about what people want, similar to this. Then it performs multiple searches on X including embedding search, and suggested people want stuff like tamagotchi, ICQ etc. back.

New comment by xyc in "Show HN: Clippy – 90s UI for local LLMs"

xyc — Tue, 06 May 2025 17:47:24 +0000

It seems that this is possibly not necessary, since LLaMA.cpp already integrates Jinja with CPP implementation (through minja)

New comment by xyc in "How I run LLMs locally"

xyc — Mon, 30 Dec 2024 23:47:45 +0000

Check out https://recurse.chat/

New comment by xyc in "Limbo: A complete rewrite of SQLite in Rust"

xyc — Tue, 10 Dec 2024 18:16:06 +0000

The fact that there's no alternative implementation of SQLite also seems to play a part in preventing standardization of WebSQL.

https://www.w3.org/TR/webdatabase/

"The specification reached an impasse: all interested implementors have used the same SQL backend (Sqlite), but we need multiple independent implementations to proceed along a standardisation path."

New comment by xyc in "Llama.cpp guide – Running LLMs locally on any hardware, from scratch"

xyc — Sat, 30 Nov 2024 05:35:01 +0000

If anyone on macOS wants to use llama.cpp with ease, check out https://recurse.chat/. Supports importing ChatGPT history & continue chats offline using llama.cpp. Built this so I can use local AI as a daily driver.

New comment by xyc in "Llama.cpp guide – Running LLMs locally on any hardware, from scratch"

xyc — Sat, 30 Nov 2024 05:30:35 +0000

You can get a release binary from https://github.com/ggerganov/llama.cpp/releases too.

New comment by xyc in "Model Context Protocol"

xyc — Wed, 27 Nov 2024 08:15:00 +0000

Made tool use work! check out demo here: https://x.com/chxy/status/1861684254297727299

New comment by xyc in "Model Context Protocol"

xyc — Tue, 26 Nov 2024 09:13:21 +0000

sharing the messy code here just for funsies: https://gist.github.com/xyc/274394031b41ac7e8d7d3aa7f4f7bed9

New comment by xyc in "Model Context Protocol"

xyc — Tue, 26 Nov 2024 07:05:53 +0000

Just tried out the puppeteer server example if anyone is interested in seeing a demo: https://x.com/chxy/status/1861302909402861905. (Todo: add tool use - prompt would be like "go to this website and screenshot")

I appreciate the design which left the implementation of servers to the community which doesn't lock you into any particular implementation, as the protocol seems to be aiming to primarily solve the RPC layer.

One major value add of MCP I think is a capability extension to a vast amount of AI apps.

New comment by xyc in "Model Context Protocol"

xyc — Mon, 25 Nov 2024 22:38:16 +0000

^ asked the question in the discussion: https://github.com/modelcontextprotocol/specification/discus...

New comment by xyc in "Model Context Protocol"

xyc — Mon, 25 Nov 2024 22:27:37 +0000

Thanks for the pointers! Will do. I've fired up https://github.com/modelcontextprotocol/inspector and the code looks helpful too.

I'm looking at integrating MCP with desktop app. The spec (https://spec.modelcontextprotocol.io/specification/basic/tra...) mentions "Clients SHOULD support stdio whenever possible.". The server examples seem to be mostly stdio as well. In the context of a sandboxed desktop app, it's often not practical to launch a server as subprocess because:

- sandbox restrictions of executing binaries

- needing to bundle binary leads to a larger installation size

Would it be reasonable to relax this restriction and provide both SSE/stdio for the default server examples?

New comment by xyc in "Model Context Protocol"

xyc — Mon, 25 Nov 2024 20:25:44 +0000

Superb work and super promising! I had wished for a protocol like this.

Is there a recommended resource for building MCP client? From what I've seen it just mentions Claude desktop & co are clients. SDK readme seems to cover it a bit but some examples could be great.

New comment by xyc in "Show HN: Embed an SQLite database in your PostgreSQL table"

xyc — Tue, 19 Nov 2024 17:13:32 +0000

With Claude you barely had to learn the language this days as you just need to prompt, but SQLite column is an interesting idea.