Hackernews posts about Qwen3:30B MoE

RSS feed for Qwen3:30B MoE

01
newShow HN: Nibia Fabric – Turn everyday computers into a shared AI compute clustergithub.com
▲ 1abelop 6d 💬 3
02
newLaunch HN: Magnitude (YC S25) – Self-optimizing inference engine for agentsgithub.com
▲ 194anerli 4d 💬 99
03
newShow HN: Token compression CLI to save Codex/Astra costs
▲ 12yolandac 4d 💬 5
04
newRunning Qwen3:30B MoE on an RTX 3070 laptop with Ollamablog.kekepower.com
▲ 18kekePower 489d 💬 1
05
newAsk HN: HotPin – lossless 120B MoE inference on 24GB RAM (CPU, 50 loc)
▲ 14LozzKappa 71d discuss
06
newShow HN: Moe-Direct – MoE Models far larger than your RAM, on a consumer desktopgithub.com
▲ 1tmxkzm1925-max 36d discuss
07
newShow HN: Flint – A 30B model fine-tuned for less repetitionspringboards.ai
▲ 6thmsmxwll 172d 💬 2
08
newShow HN: Open-Source Browser Use LLM (30B, A3B)huggingface.co
▲ 2gregpr07 289d discuss
09
newA 30B Qwen model walks into a Raspberry Pi and runs in real timebyteshape.com
▲ 363dataminer 271d 💬 135
10
newCline and LM Studio: the local coding stack with Qwen3 Coder 30Bcline.bot
▲ 80Terretta 400d 💬 20
11
newShow HN: First Claude Code client for Ollama local modelsgithub.com
▲ 44SerafimKorablev 255d 💬 27
12
newCline and LM Studio: the local coding stack with Qwen3 Coder 30Bcline.bot
▲ 3ljosifov 402d discuss
13
newQwen 3.5 9B, 4B models beating 30B, 80B modelshuggingface.co
▲ 2satvikpendem 216d discuss
14
newShow HN: Shoehorn, a library to quantize an LLM to fit your Mac's VRAMgithub.com
▲ 6rhgraysonii 52d discuss
15
newShow HN: Levi – run AlphaEvolve on your Claude Code/Codex for dirt cheapttanv.github.io
▲ 5ttanv 119d discuss
16
newQwen 3 1.7B running locally on mobile at 30 tokens/stwitter.com
▲ 4chmjkb 507d 💬 3
17
newShow HN: FC-Eval – CLI to Benchmark Local or Cloud LLMs on Function Callinggithub.com
▲ 3gauravvij137 202d discuss
18
newShow HN: Off Grid – Run AI text, image gen, vision offline on your phonegithub.com
▲ 124ali_chherawalla 232d 💬 66
19
newShow HN: Dicta.to – Local voice dictation for Mac with on-device AIdicta.to
▲ 2alamparelli 223d discuss
20
newShow HN: MyDeviceAI – Local AI with web search for your phoneapps.apple.com
▲ 3navedmerchant 499d discuss
21
newI gave Qwen 3.8 27B a reverse-engineering job and it finished in 30 minuteswww.xda-developers.com
▲ 369raybb 43d 💬 153
22
newShow HN: Learn from 30 historical figures, open source, nonprofit, self-hostedgithub.com
▲ 49micstradev 118d 💬 24
23
newShow HN: Off Grid: On-device AI-web browsing, tools vision,image,voice–3x faster
▲ 12ali_chherawalla 222d 💬 5
24
newShow HN: How clanker are you? A reverse Turing testhowclankerareyou.com
▲ 11niklio 93d 💬 11
25
newShow HN: I am running 3 coding agents non-stop over the last 3 days. Here is how
▲ 10sermakarevich 113d 💬 3
26
newI've tested some local LLMs on prosumer hardware, here are some findings
▲ 7felineflock 43d discuss
27
newShow HN: OS Megakernel that match M5 Max Tok/w at 2x the Throughput on RTX 3090github.com
▲ 6GreenGames 180d 💬 1
28
newRun 35B LLMs on Dual Pascal GPUs with QLoRA
▲ 4rickesh_tn 363d discuss
29
newShow HN: Sentient OS – On-device intelligence layer for your entire digital lifesentient-os.ai
▲ 3TechExpert2910 155d 💬 2
30
newShow HN: Minna – On-device semantic search across your own documents (macOS)github.com
▲ 3ActuallyTaylor 47d discuss
31
newShow HN: Diktafon – Voice memos on cassette tapes, transcribed on-devicegithub.com
▲ 1jaromiru 75d 💬 1
32
newOff Grid: On-device AI-web browsing, tools, vision, image gen, voice – 3x faster
▲ 1ali_chherawalla 222d 💬 1
33
newThe AI Conundrum: We are living in highly subsidized, interesting times
▲ 10pyeri 105d 💬 1