Filter
Exclude
Time range
-
Near
Replying to @Grummz
Why would I use this over LLMstudio
3
1
9
17,121
Replying to @0xSero
great talk and you convinced me to install LLMstudio, btw
2
671
Replying to @AtrakcyjnyAdam
Warto sb obczaić zamiennik llmstudio, który ma wsparcie MTP, DFlash i TurboQuant(lub jego szybszy fork) dla tego Qwena Chyba że llmstudio już to ma, nwm
1
6
393
Replying to @egebese
İlk defa geçen haftasonu denedim. Sabaha kadar uyuyamamışım. Bu hafta tekrar uğraşınca yazarım modelleri ve detayları. Mac mini m4 24gb var. Bir sürü model denedim hepsini de yazdım ama buraya sığmaz şimdi. Qwen3coder iyiydi baya onu hatırlıyorum. Huggingface modellerinden çok verim alamadım Ollamada. Bu haftasonu llmstudio ile deneyeceğim.
3
1,023
Mar 31
Replying to @fipa
llmstudio to umi a doporucil bych ho pred ollama
1
2
533
Replying to @Ced_haurus
D'où l'intérêt de stocker sur son pc un ou plusieurs modèles d'IA via llmstudio par exemple. Bon ce n'est pas Byzance, mais c'est déjà pas mal.
2
8
1,560
Llmstudio 2-200 gb vram ;) firefox 2/3 gb ram. Good os not use ram !!! User software use ram !! Gaming use ram. Run local ai :-). Every KB saved by the OS is a KB I can use for software, this also applies to the CPU, every unused cycle is an available cycle
2
148
Ni org2 ngerender video AI pake NVIDIA brp GB deh? 16GB? Gw pake GTX1050 4GB aja install LLMStudio masih lemot

aespa - Karina
Community note
This AI-generated video uses footage of aespa Karina from the K-Link Festival Oct 5 2024 The creation and circulation of this content constitute sexual harassment and violate South Korea’s laws on digital sexual crimes, including regulations on AI/deepfake sexual exploitation. youtu.be/VwEoybH0rPA?si
2
1,696
11 Nov 2025
Ever felt that traditional AI simply doesn’t speak your business language? Today’s enterprises need AI that’s powerful yet controlled, intelligent yet contextual. Off-the-shelf models can’t always speak your language. That’s where Cogentiq LLM Studio redefines the game: 🔸 It gets your data AI-ready in just a few clicks. No heavy tech skills needed. 🔸 Models are tuned and tested as per your business context 🔸 Once deployed, they’re governed, automated, and scaled seamlessly on the cloud. Cogentiq LLMs cuts deployment cycles from months to weeks to deliver measurable ROI Customize smarter decisions, seamless automation, and real results at scale with Cogentiq LLM Studio : fractal.ai/products/cogentiq… #Cogentiq #LLMStudio
4
88
12 Oct 2025
@ivanfioravanti Do you have some performance comparation with the same model between llmstudio, ollama and Mlx-lm/mlx-vlm? Thanks!!
1
3
597
well, I downloaded llmstudio, installed it, and did a test run. needless to say for now, I will continue to use my paid version of chatgpt, unless of course, you have a recommendation that peforms even better.
5
3
17
587
Pues si se puede y no es tan difícil, de hecho no necesitas que alguien la haga ya están herramientas como LLMStudio o similares. Eso solo es la interfaz, lo que es difícil es el modelo detrás pero usando APIs es bastante viable de hecho.
2
4
3,956
11 Mar 2025
Replying to @Tono_Ken3
やべぇ、僕もLLMstudioにDeepseek落としてローカル運用してたんですが、やっぱそっちだとヤバいですよね…いいなぁ…僕は初代ipod touchからのiOS信者をiphone8でやめてから、根っからのWindows Android信者になったんですが、しんかまんさんのポストでMacコスパいいなぁって思ってます…いいなぁ
1
4
798
AI-Infra-Guard:AIインフラセキュリティ評価ツール ✅ 主な機能 AIインフラストラクチャのセキュリティリスクを検出・評価 アプリケーション詳細: 0.0.0.0:8081 (ai-infra-guard, ステータスコード200, リスク数0) 0.0.0.0:8080 (Page not found at /, ステータスコード404, リスク数0) 0.0.0.0:8000 (MLflow, ステータスコード200, mlflow:2.11.0, リスク数13) ✅ 特徴 効率的なスキャン:28種類のAIフレームワークのフィンガープリント識別、200以上の脆弱性データベースに対応! 使いやすさ:複雑な設定不要、YAML形式のルール定義、柔軟なマッチング構文 軽量:クリーンで効率的なコアコンポーネント、小さなバイナリサイズ、低いリソース使用量、クロスプラットフォーム対応 ✅ WEBUIとコマンドライン コマンドラインでも利用可能! ✅ インストールと使い方 1️⃣ 以下からダウンロード: 2️⃣ 実行! WEBUI: `./ai-infra-guard -ws` ローカルスキャン: `./ai-infra-guard -localscan` 単一ターゲット: `./ai-infra-guard -target [IP/ドメイン]` 複数ターゲット: `./ai-infra-guard -target [IP/ドメイン] -target [IP/ドメイン]` ファイルから読み込み: `./ai-infra-guard -file target.txt` AI分析: `./ai-infra-guard -target [IP/ドメイン] -ai -token [Hunyuan token]` ✅ 詳細な情報 28種類ものAIコンポーネントに対応 anythingllm, langchain, Chuanhugpt, clickhouse, comfyui 関連, dify, fastchat 関連, feast, gradio, jupyter 関連, kubeflow, kubepi, llamafactory, llmstudio, ollama, open-webui, pyload-ng, qanything, ragflow, ray, tensorboard, vllm, xinference, triton-inference-server フィンガープリントマッチングルール WEBフィンガープリント識別コンポーネントに基づいているそうです。 `data/fingerprints`ディレクトリにフィンガープリントルール `data/vuln`ディレクトリに脆弱性マッチングルール AIインフラのセキュリティ対策に、ぜひ活用したいですね! #AI #セキュリティ #Tencent
1
3
134
Myth: Prompt engineering is too complex for the average user. Fact: With tools like LLMStudio and Azure PromptFlow, anyone can harness LLMs! 🚀💡 #PromptEngineering #AI #Innovation Link below.
2
2
31
You cant censor or alter if you host your own AI LLM model! I just added support for Kobold, LLMStudio, and any OpenAI WebUI frontend. We want to add local text to speech too.
2
5
353
Replying to @gaffar_olamide
llmstudio doesnt have advance tool capabilities so you are not able to have agents also it doesnt have crypto integrations
1
1
2
31
Replying to @awnihannun
This is my experience from Llama 8B distilled version on LLMStudio.
2
503
- NovaSky blog post: novasky-ai.github.io/posts/s… - All source code (model and training data) at: github.com/NovaSky-AI/SkyTho… - On @huggingface at huggingface.co/NovaSky-AI/Sk… (note: only some quantized forks can run on Jan/LLMstudio... for now)

2
166