Hacker News: jboss10

New comment by jboss10 in "Qwen 3.6 27B is the sweet spot for local development"

jboss10 — Mon, 29 Jun 2026 20:25:19 +0000

Qwen 3.6 35B runs on 32GB with a 1080. That GPU is from 2017.

New comment by jboss10 in "Qwen 3.6 27B is the sweet spot for local development"

jboss10 — Mon, 29 Jun 2026 20:23:16 +0000

I have 8GB VRAM but 32GB RAM. Qwen 3.6 35B runs nicely.

You should look at gemma-4-26B-A4B. 16+8=24gb and Q4 is about 16GB. Not much context left, but might run.

New comment by jboss10 in "Qwen 3.6 27B is the sweet spot for local development"

jboss10 — Mon, 29 Jun 2026 20:20:46 +0000

I have 8GB VRAM, but 32GB sys ram. I can run qwen 3.6 35B at 30 tok/s. I also use pi, and it's smart enough to extend itself(multishot and maybe a few tries)

For you, you could try gemma-4-26B-A4B

New comment by jboss10 in "Qwen 3.6 27B is the sweet spot for local development"

jboss10 — Mon, 29 Jun 2026 20:13:19 +0000

I don't understand the talk about how expensive the hardware is. These models can run on very old or old and low end. I've been running Qwen3.6-35B Q4 on an old 1080 GPU(8GB vram) with 32GB sys RAM. I have a i7-12700.

It does about 30 tok/s which is enough for me. It's about half what the online models do, but it's enough.

I've heard their 9B models are also good, but they aren't much faster if you have the ram and a nice cpu.

These qwen3.6 models are the first ones I find can do much. GPT OSS was good, and Gemma4 is better. Gemma knows more facts, but qwen3.6 is smarter.

New comment by jboss10 in "Qwen 3.6 27B is the sweet spot for local development"

jboss10 — Mon, 29 Jun 2026 20:08:30 +0000

They can be ran on 32GB with 8GB VRAM. I don't think these will be on 16GB for a while. (35B MoE)

New comment by jboss10 in "Qwen 3.6 27B is the sweet spot for local development"

jboss10 — Mon, 29 Jun 2026 20:05:31 +0000

For the 35B model, ofloading to RAM doesn't slow it down much. If you have a nice CPU and a weak GPU, it will be fast enough to use.

New comment by jboss10 in "Qwen 3.6 27B is the sweet spot for local development"

jboss10 — Mon, 29 Jun 2026 19:55:07 +0000

https://unsloth.ai/docs/models/qwen3.6 And https://huggingface.co/collections/unsloth/qwen36

New comment by jboss10 in "Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?"

jboss10 — Mon, 15 Jun 2026 21:03:13 +0000

Have you tried qwen3.6 or pi?

New comment by jboss10 in "Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?"

jboss10 — Mon, 15 Jun 2026 20:50:19 +0000

llama.cpp It's faster and more open source. Ollama has some mixed history. I use llama-swap to emulate the Ollama experience.

Construction Manager Vibe Codes Paperwork Tool, Accidentally Breaks It

jboss10 — Wed, 08 Apr 2026 20:25:08 +0000

Article URL: https://www.businessinsider.com/construction-manager-vibe-coded-paperwork-tool-2026-4

Comments URL: https://news.ycombinator.com/item?id=47695808

Points: 1

# Comments: 0

Beyond Python: Why LLMs Need More Stable, Open Source Code

jboss10 — Mon, 12 Jan 2026 12:28:38 +0000

Article URL: https://thenewstack.io/beyond-python-why-llms-need-more-stable-open-source-code/

Comments URL: https://news.ycombinator.com/item?id=46587533

Points: 1

# Comments: 0

My father first came to the UK, people looked after him. Would that happen now?

jboss10 — Sat, 27 Dec 2025 12:58:09 +0000

Article URL: https://www.theguardian.com/commentisfree/2025/dec/24/father-uk-people-looked-after-him-refugees

Comments URL: https://news.ycombinator.com/item?id=46401537

Points: 5

# Comments: 0