LBHustonThe model shows strong ideological guardrails consistent with Chinese training alignment, reducing neutrality on certain geopolitical topics.
Read the full article: Assessment of Qwen3.5-9b in LMStudio
▸ https://lttr.ai/ApXVL
Overall, Qwen3.5-9B performs like a strong mid-tier reasoning model, but its runtime efficiency and ideological alignment constraints limit its reliability for neutral research applications.
Read more 👉 https://lttr.ai/ApVNE
Based on the provided prompt–response dataset, the Qwen3.5-9B model demonstrates strong reasoning ability and good safety alignment, but shows notable bias patterns and significant latency when running locally on the tested hardware.
Read more 👉 https://lttr.ai/ApU6j
Moonstone2487Stürzt bei noch jemandem in #lmstudio das #model ohne weitere Info ab, wenn er eine Datei in den Chat gibt und Vision verwendet? Bei mir jedes mal. An #qwen oder der Systemlast kann es nicht liegen.
Hab schon mit #ollama probiert. Hat zwar 5 Minuten gebraucht, um die Katze zu beschreiben, weil er noch Thinking gemacht hat, aber das Ergebnis war zufriedenstellend und ich hatte keinen Crash.
"The model has crashed without additional information. (Exit code: null)"
TevinFinally, an open-source alternative that's easy to use and performs better than #LMStudio on Mac with support for MLX!
A bit shady that it uploads the benchmark result without noticing or asking tho
Alex Grant#lmstudio model loading panel has an extremely useful and time saving interface that shows you an estimate of the effective memory usage based on your model runtime parameter settings.
Great tool 🙏
wiulinu
DansNull
sayzardNobu-Kobayashi : Generative AI Technology (@nyaa_toraneko)
LM Studio에 직접 웹 검색 기능을 붙이는 오픈소스 프로젝트(mrkrsl/web-search-mcp)를 소개합니다. 이를 통해 로컬 LLM의 지식 컷오프를 신경 쓰지 않고 최신 웹 정보를 검색해 활용할 수 있어, 로컬 LLM 개발·운용에서 매우 편리하고 실무적으로 유용합니다.
ich hab da mal was aufgeschrieben:
https://text.tchncs.de/wiulinu/goose-und-lm-studio
