Tag

google/gemma-4-12B-it

Instruction-tuned, unified Gemma 4 12B multimodal model that accepts text, image and audio inputs and generates text outputs locally. Encoder-free design reduces multimodal latency and fits on consumer devices while offering long-context support and native thinking/system-prompt features.

Gemma 4 12B Unified

A 12B unified, encoder-free multimodal model that directly ingests text, images and audio and returns text; supports very long contexts (up to 256K tokens), native function-calling/thinking modes, and small-model deployment for local or on-device use.

Step 3.7 Flash

Processes images and text to produce structured, reasoning-rich text outputs for high-throughput agentic workflows. Sparse MoE design (198B total, ~11B active per token), 256k context window and selectable reasoning levels—optimized for single-pass parsing, verification, and multi-step automation.

Keye-VL-2.0-30B-A3B

Performs hour-scale video understanding and fine-grained temporal localization while exposing agent-style multimodal tool/code/search abilities. Built on a sparse-attention long-context architecture (DSA) and a specialized inference stack—best used in GPU-backed research or production evaluation.

LocateAnything-3B

Performs fast, high-quality vision–language grounding: given an image plus a natural-language prompt it returns bounding boxes or points for referred objects. Uses Parallel Box Decoding for parallel coordinate prediction (higher throughput) and targets research/non-commercial use.

MeasL-Bench

Benchmark for evaluating vision–language models on measurement-grounded inputs vs. RGB, emphasizing low-light, HDR, and visibility-sensitive evidence recovery. Contains 2,183 paired test examples with local image assets for controlled RAW↔RGB comparisons.

PaddleOCR-VL-1.6

Performs image-to-text document parsing and OCR for complex elements (tables, formulas, charts, seals), with multilingual support (en/zh). It uses region-aware data optimization and progressive post-training to improve weak-region supervision and is plug-and-play compatible with PaddleOCR-VL-1.5.

Cosmos3-Super-Text2Image

Generates high-fidelity images from text prompts using NVIDIA's 64B Cosmos3-Super multimodal foundation model. Integrates with Hugging Face Diffusers and vLLM‑Omni, is released under OpenMDW1.1 for commercial use, and is optimized for Physical AI workflows (robotics, AV, simulation).

Why Far Looks Up: Probing Spatial Representation in Vision-Language Models

Analyzes spatial representations in vision–language models and reveals a consistent vertical-position ↔ distance entanglement; introduces SpatialTunnel, a synthetic benchmark that exposes this perspective-driven shortcut, and provides code and a project page.

EarlyTom: Early Token Compression Completes Fast Video Understanding

Performs training-free early-stage visual token compression inside the vision encoder to cut time-to-first-token (TTFT) and FLOPs for Video-LLMs. Introduces a decoupled spatial token selection strategy and reports up to 2.65× TTFT reduction and 61% FLOPs savings on LLaVA-OneVision-7B (NVIDIA A100) while preserving full-token accuracy — aimed at latency-sensitive video understanding.

GGT-100K: Generative Ground Truth for Generalizable Real-World Image Restoration

Synthesizes high-quality targets for real-world image restoration by using multimodal foundation models (MFMs) to convert real low-quality photos into HQ references. Provides GGT-100K (103,707 LQ–HQ training pairs + 500 test pairs) with multi-stage quality control and demonstrates consistent generalization gains for a range of restoration models, especially for finetuning generative restorers.

unsloth/gemma-4-12b-it-GGUF

A GGUF-quantized, locally runnable build of Gemma 4 12B Unified (image-text-to-text) packaged by unsloth; preserves multimodal (image/audio) input support under an Apache-2.0 license and is compatible with common GGUF runtimes and Unsloth Studio.