Raon-1 mini is a 30B-A4B Mixture-of-Experts (MoE) language model, designed and trained from scratch by KRAFTON AI. It demonstrates strong capabilities in Korean conversation and agentic tasks.
Could this beat Tiel Coder?
- 10 posts
- 7 comments
Joined 3 months ago
Cake day: July 29th, 2026
You are not logged in. If you use a Fediverse account that is able to follow users, you can follow this user.
Maybe a better alternative to Tiel Coder for people with 32GB RAM and 8GB VRAM
https://piefed.social/comment/13343280
PieFed has the concept of silencing another instance. A silenced instance is not defederated from but their posts do not show in the Popular or All feeds and their communities are not shown during onboarding. Their communities can still be found in the communities list (if you unblock the instance first) and joined in the normal way. Once joined, posts in there show up in the Subscribed feed of the joiner as usual.
Since I added silencing functionality I feel like there’s almost no need for (more) defederation. I can offer protection to newbies who don’t yet know which places to avoid and they can explore the dark corners when they’re ready. It’s a low-drama way to put a bit of distance between groups that rub each other the wrong way.
Yeah default blocks or silencing or something like that
PieFed allows admins different levels of blocking instances, they can do defederation but there are levels softer than that
oh yea feddit blocks shitjustworks https://feddit.online/instance/sh.itjust.works/posts
maybe @xSikes@feddit.online could ask their admins to convert the defederation to a “silenced” or whatever that feature is on PieFed
32GB RAM, RTX 2080 8GB VRAM, I use Tiel Coder 35b-a3b (it’s a fine tune of Qwen 3.6) with Zoo Code
I see the last activity was 2 months ago.
that means no one from your instance is subscribed, if you subscribe then you’ll get new posts
you can see that there are new posts here https://sh.itjust.works/c/localllama
Actually much improved benchmark scores considering this is supposed to be a “minor” update
I don’t know why they don’t just call this v4.2, why don’t LLM labs know how to do version numbers?
I just heard about this finetune of Qwen 3.5 9b. It claims to be better than Ornith 1.5 9b and Ling 3.0 Tiny
--moe-cache-mib doesn’t seem optimized for tight VRAM situations or if your PCIe link gets saturated by it, but it improves performance on some setups.
NVIDIA Nemotron 3.5 Bolt 74B-A9B 24GB is a vision-language model (VLM) designed to run everyday agentic tasks locally, using a native 2-bit quantization to fit in 24GB of memory and run alongside other applications. It optimizes intelligence and speed per GB. It aims to achieve the optimal balance between speed and intelligence, being faster than most popular dense models while smarter than most popular MoE models with a similar memory footprint.
This might actually work in 8GB VRAM? 74B parameters at 24GB means about 3GB for the 9b active parameters.
Muse Spark Open is an 816B total/42B active parameter open weights model with a context window of 1M tokens. Built for coding and agentic work, Muse Spark Open is natively multimodal.
Don’t bother with their download restrictions, just check the quant providers https://huggingface.co/models?other=base_model%3Aquantized%3Ameta-models%2FMuse-Spark-Open-816B-A42B
This might just barely fit in 16GB RAM with their QAD-Q4_0 being only 13.6 GB. Should be good for small VRAM too, like 4GB
Some changes I picked out, you may need to change your configs
- Removed deprecated --mmap/–mlock/–direct-io args in favor of --load-mode
- Made mmproj and draft devices default to the global --device selection
- Enabled --reasoning-preserve by default
I just use llama.cpp with Zoo Code, the VSCode extension
join us !localllama@sh.itjust.works