Firethering1d ago

MiniCPM Desk Pet turns the MiniCPM AI model into a desktop companion that lives alongside your workflow. Install the app, follow the setup wizard, and within a few minutes you can chat with a local AI pet directly from a floating desktop bubble.

Conversations run on your machine using the local model. The pet can stay visible while you work, react to activity from tools like Cursor, Claude Code, and Codex.

https://firethering.com/minicpm-desk-pet-open-source-local-ai-desktop-pet/
#minicpm #coding #opensource #ai #claudecode #developers

MiniCPM Desk Pet: Open Source AI Desktop Companion That Runs Locally - Firethering

MiniCPM Desk Pet turns the MiniCPM model into a desktop companion that lives alongside your workflow. Install the app, follow the setup wizard, and within a few minutes you can chat with a local AI pet directly from a floating desktop bubble. The app checks your environment, downloads the model, warms it up, and simplify the complexity of the setup Once everything is ready, conversations run on your machine using the local model. The pet can stay visible while you work, react to activity from tools like Cursor, Claude Code, and Codex, and even take on different personalities through character adapters. It's part local AI assistant, part desktop pet.

Firethering
Firethering2d ago

MiniCPM5-1B Shows Why the Small-Model Race Isn’t Over
https://firethering.com/minicpm5-1b-small-model-reasoning/

#ai #llm #minicpm #genai #opensource #openweights #technews #tech

MiniCPM5-1B Shows Why the Small-Model Race Isn't Over - Firethering

A 1B model scoring 40.42 on AIME 2025 should not be possible. AIME is the American Invitational Mathematics Examination, the kind of test that filters out most humans who attempt it. Qwen3-0.6B scores 16.25 on the same benchmark. LFM2.5-1.2B, a larger model, scores 31.88. MiniCPM5-1B, at roughly one billion parameters, beats both. OpenBMB just dropped MiniCPM5-1B, the first model in their MiniCPM5 series, and it's built specifically for the scenarios like on-device deployment, resource-constrained environments, local inference on consumer hardware. The AIME score is surprising. The telecom agent benchmark is even more surprising. And then there's the desktop pet. We'll get to that.

Firethering
H@R0👨🏻‍💻3d ago

最近的小模型可以這麼小？ 想知道它能做到甚麼，與qwen3.5 9b比較有多大差異？

https://github.com/openbmb/minicpm

#minicpm

GitHub - OpenBMB/MiniCPM: MiniCPM5-1B: A SOTA 1B on-device LLM, small yet powerful.

MiniCPM5-1B: A SOTA 1B on-device LLM, small yet powerful. - OpenBMB/MiniCPM

GitHub
Show threadSome(oz)Mar 10, 2025

Me resigno a pagar OpenAI, o otro LLM, porque no tengo un servidor lo suficiente potente donde correr MiniCPM-V con Ollama.

Este modelo es "ligero" pero se traga 4GB de memoria de todos modos, y corre mucho mejor si tiene acceso a un GPU. En este contexto son más económicas las APIs de OpenAI o Anthropic.

Creo que para extraer datos de una imagen pequeña, será gratis o cerca de 0 de todos modos, pero perfiero no depender de servicios externos. 🤔

#ollama #minicpm

Alexander GerberFeb 8, 2025

Zur Einordnung:
"[...] Kombination aus kleineren Modellen und lokaler Datenverarbeitung die Kosten senkt und den Schutz der #Privatsphäre verbessert.

Das neueste Modell von #Modelbest, #MiniCPM 3.0, hat
NUR
vier #Milliarden #Parameter, erreicht aber bei verschiedenen Benchmarks die Leistung des alten #chatgpt35."

https://t3n.de/news/was-kommt-nach-deepseek-diese-vier-ki-startups-aus-china-solltest-du-dir-merken-1671497/?utm_source=firefox-newtab-de-de

Nach Deepseek: Diese KI-Startups aus China solltest du kennen

In China gibt es einige KI-Firmen, die mit OpenAI, Anthropic und Co. konkurrieren könnten – und das nicht nur in ihrer Heimat.

t3n Magazin
王永帥🍥Jan 15, 2025

#開源分享 面壁開源了其最新模型：MiniCPM-o 2.6，首個支持在 iPad等端側設備進行多模態即時流式交互的多模態模型，視覺、語音和多模態流式能力說是達到了GPT-4o-202405級別

1、總參數量 8B

2、支持可配置聲音的中英雙語語音對話，同時具備情感、語速、風格控制、端到端聲音複製、角色扮演等進階能力

3、能接受連續影片和音訊串流，進行即時語音互動。在StreamingBench上，超過了GPT-4o-202408和Claude 3.5 Sonnet

4、增強了OCR、可信行為、多語言支持和影片理解等視覺能力

專案地址： github.com/OpenBMB/MiniCPM-o

#面壁小鋼炮 #MiniCPMo #MiniCPM

Dr. Fortyseven 🥃 █▓▒░Sep 11, 2024

Good ol' Squeaky Pete.

#llm #minicpm

Banana Pi Open Source HardwareJul 19, 2024
#BananaPi BPI-F3 Now this is the model of #ChatGLMTV2-6b, the model size is 3.6G, the performance is 0.7t/s in Chinese, 1.3T/s in English.such as #minicpm, #qwen2, #phi3, #llama3, #qwen1.5, #chatglm, etc.
https://forum.banana-pi.org/t/bpi-f3-large-model-demo/18541
#ai #llm
BPI-F3 large model Demo

BPI-F3 large model Demo has been put into github source code after V1.0.8, you can experience the following installation:\ sudo apt-get update sudo apt-get install inferllm-chatglm; sudo apt-get install inferllm-chatglm; Now this is the model of ChatGLMTV2-6b, the model size is 3.6G, the performance is 0.7t/s in Chinese, 1.3T/s in English; SpacemiT Key Stone™ K1 implements local operation of many mainstream miniaturized large models such as minicpm, qwen2, phi3, llama3, qwen1.5, chatglm, e...

banana pi single board computer open source project official forum BPI team