Havuz

doc7, belgeleri multimodal model kullanarak görsel içeriği de koruyan AI-ready Markdown çıktısına dönüştüren bir Go aracıdır.

Neden yükseliyor · 56 günde 1137 yıldız, projeye erken dönemde belirgin ilgi olduğunu gösteriyor; ancak son 28 günde commit ve farklı yazar sayısı sıfır, dolayısıyla ilginin sürdürülebilirliği ve bakım temposu henüz kanıtlanmış değil.

DSH Vision Toolkit, DeepSeek Harness içindeki text-only agent'lara görsel görevler için araçlar ve yerel entegrasyon sunan bir plugin'dir.

Neden yükseliyor · 45 günlük yaşına karşılık 883 yıldız ve son 28 günde 49 commit dikkat çekici; README de kolay kurulum ve doğrudan görsel kullanımı vurguluyor. Ancak bu yükselişin nedenini veya gerçek kullanım yaygınlığını verilen verilerle doğrulamak mümkün değil; 49 commit'in yalnızca 2 farklı yazardan gelmesi geliştirme faaliyetinin yoğun ama dar bir ekibe bağlı olduğunu gösteriyor.

Bilingual (中文+EN) ML / LLM / diffusion / agent interview cheat sheets for AI 秋招 — generated by ARIS /interview-cheatsheet, rendered by /rend

🚀🚀 A multimodal System One decision model that gives calibrated answers to typed questions about screens, photos, video and text in one forw

Sleek, mobile-friendly web UI for NVIDIA LocateAnything-3B — open-vocabulary object detection & grounding on your own GPU, via one docker co

NEWTON: Agentic Planning for Physically Grounded Video Generation

基于多模态视觉感知与 LLM Agent 的 macOS 微信自动化框架 | Visual RPA for WeChat

Browser-based Android phone agent using WebADB/WebUSB and OpenAI-compatible vision models

Can my GPU run this? Pick a GPU to get 3 fitting AI models, quantization, VRAM, speed ranges, and run commands. 한국어·English.

A token-efficient, provider-agnostic vision layer for reasoning models via MCP.

UI-UX: a 4B multimodal LLM for mobile UX defect diagnosis, with UXBench — the first vision–language benchmark for UX reasoning (CVPR 2026 Fi

Qwen3.8-Flash-Next 177B MoE (IQ3_S 3.44bpw) on RTX 5090 Laptop 24GB: full-model quality @ 97.7 tok/s · dual-lane 256K full-context unlocked

framework for using llms on object grounding

Gemma 4 31B Abliterated — quality-preserving guardrail removal for Google's most capable open model. Apache 2.0. Runs on Apple Silicon via M

[ECCV 2026] Towards Spatial Supersensing in the Wild

Qwen3.8-27B on RTX 5090 Laptop 24GB — TWIN-TURBO NVFP4 (native llama.cpp) vs Coder390 Q3 LynnStyle (.ninfer + NInfer engine in WSL2/Docker):

An agent skill that reads a photo the way Sherlock Holmes reads a stranger (福尔摩斯): observe first, deduce second, every conclusion cites its

Finds a stolen bike on second-hand marketplaces by ranking photos, not keywords — a CLIP pre-gate in front of a vision-LLM recognition pass.

Native Rust, single-binary MLX inference server for Apple Silicon. OpenAI/Anthropic-compatible LLM serving — text, vision, audio, embeddings

Open-source video compiler that turns videos into structured knowledge for AI agents.

Save UI inspiration from any app; a vision model curates it into a private wiki and turns your taste into a DESIGN.md your coding agent buil

A temporal visual reasoning benchmark over egocentric robot manipulation video — 39 scenarios, 7 dimensions, 34,713 questions

DeepSeek-V4.1-Flash-1017 on one 256 GB M5 Ultra: 1M context, 70.7-104.9 tok/s decode with DSpark drafts, 1,739-1,867 tok/s cold prompts (8k-

A long-horizon CRPG benchmark for frontier agents. Raw 320x200 frames, Traditional Chinese, isometric navigation, twelve books to find.

browser-use, but for the entire desktop — give any AI Agent eyes and hands on Windows. MCP + SDK, VLM-optional, with visual loop, introspect

Apple Silicon 本地视觉语言模型 CLI：用 MLX/Qwen3-VL 描述图片、图像问答与提示词反推。Offline vision-language CLI for image understanding, visual Q&A and reverse prompt