Saved toolsSubmit a tool
FOXAI SEARCH

Search FoxAI

AI Models 2119

  • sam3↗

    bodhicitta

  • MiniCPM-V-4.7-35B-A3B-AWQ↗

    openbmb

    > **This repository hosts the AWQ (W4A16, compressed-tensors, llm-compressor) quantized version of MiniCPM-V 4.7 35B-A3B.** For the original BF16 weights and the full model card, please refer to openbmb/MiniCPM-V-4.7-35B-A3B. **MiniCPM-V 4.7** is the latest model in the MiniCPM-V series, available in a **35B-A3B** variant, with **35B-A3B being the first MoE model in the MiniCPM-V series**. It builds on the family's single-image, multi-image, and video understanding capabilities, combining stronger foundation capabilities with efficient visual encoding and on-device inference. Notable features of MiniCPM-V 4.7 include: - 📱 **The First Open-Source MoE Model with Lossless, Smooth On-Device Inference.** With our **Lookahead Expert Scheduling (LES)** technique introduced alongside MiniCPM-V 4.7, combined with fine-grained layer-wise quantization and DSpark speculative sampling integration, MiniCPM-V 4.7 35B-A3B becomes the **first open-source MoE model to run smoothly on-device without compromising performance**, achieving a decoding speed of **13+ tokens/s with just 3 GB of memory on iPhone 17 Pro Max**. - 🔥 **A New Pareto Frontier in Performance and Efficiency.** As model capabilities continue to advance, inference efficiency has become an important measure of a model's value. On resource-constrained devices, delivering the strongest possible capabilities within a limited resource budget calls for higher capability density. The **35B-A3B** variant of MiniCPM-V 4.7 advances the performance–efficiency Pareto frontier for MLLMs. In Instruct mode, 35B-A3B **achieves a higher overall score in less inference time than Qwen3.6 35B-A3B**. - 🚀 **Efficient Architecture.** Based on the intra-ViT early compression technique from LLaVA-UHD v4, MiniCPM-V 4.7 compresses visual tokens early within the vision encoder, reducing computation in subsequent encoder layers. It also supports **mixed 4x/16x visual token compression**, allowing developers to balance visual detail and inference cost across tasks. - 💻 **Broad Platform and Chip Support.** MiniCPM-V 4.7 supports **iOS, Android, and HarmonyOS**, together with open-source on-device deployment support. It also provides **day-0 support for chip platforms from Rockchip, Houmo, Intel, AMD, and Arm**, extending deployment options across mobile and local computing devices. - 🛠️ **Developer Friendly.** MiniCPM-V 4.7 supports in

  • SAME-S↗

    Stability AI

  • krea-realtime-video↗

    krea

    Krea Realtime 14B is distilled from the Wan 2.1 14B text-to-video model using Self-Forcing, a technique for converting regular video diffusion models into autoregressive models. It achieves a text-to-video inference speed of **11fps** using 4 inference steps on a single NVIDIA B200 GPU. For more details on our training methodology and sampling innovations, refer to our technical blog post. - Our model is over **10x larger than existing realtime video models** - We introduce **novel techniques for mitigating error accumulation,** including **KV Cache Recomputation** and **KV Cache Attention Bias** - We develop **memory optimizations specific to autoregressive video diffusion models** that facilitate training large autoregressive models - **Our model enables realtime interactive capabilities**: Users can modify prompts mid-generation, restyle videos on-the-fly, and see first frames within 1 second Krea realtime allows users to stream real videos, webcam inputs, or canvas primitives into the model, unlocking controllable video synthesis and editing

  • VisionPsy-Nano-460M-Flash-GGUFs↗

    qvac

    **VisionPsy-Nano-460M-Flash-GGUFs** provides GGUF weights of VisionPsy-Nano-460M-Flash for fast, fully on-device multimodal inference with our patched **llama.cpp** fork (not stock upstream). Flash is the **efficiency-optimized** sibling of VisionPsy-Nano-460M: same ~460M nanoVLM architecture, but it processes each image with **far fewer visual tokens (as few as 64 at 512x512)**, so it responds an order of magnitude faster on real phones while giving up only a small amount of accuracy. Unquantized **FP32** and **BF16 GGUFs** are included alongside a family of quantization formats, from near-lossless **8-bit** through a high-quality **5-bit** option down to ultra-compact **3-bit**. All sub-8-bit variants (except the legacy `Q4_0`) use **importance-matrix (imatrix) calibration**. If VisionPsy-Nano-460M-GGUF is our best-quality ~0.5B GGUF, **Flash-GGUF is the one you ship when time-to-first-token, battery, and RAM are the binding constraints.** > Each GGUF is a **vision-language** model: alongside the text weights (`*.gguf`) you also need the > multimodal projector (`mmproj-*.gguf`, the SigLIP2 vision encoder) to run image inputs. See > Usage.

  • ZR1-1.5B↗

    Zyphra

    ZR1-1.5B is a small reasoning model trained extensively on both verified coding and mathematics problems with reinforcement learning. The model outperforms Llama-3.1-70B-Instruct on hard coding tasks and improves upon the base R1-Distill-1.5B model by over 50%, while achieving strong scores on math evaluations and a 37.91% pass@1 accuracy on GPQA-Diamond with just 1.5B parameters. For training we utilized the PRIME Eurus-2-RL dataset which combines the following math and code datasets: - NuminaMath-CoT - APPS, CodeContests, TACO, and Codeforces train set We filtered math data by validating that questions are correctly graded when calling the evaluator with reference ground truth, and we removed all code examples with an empty list of test cases. Our final dataset comprised roughly 400k math + 25k code samples.

  • SmolLM3-3B-checkpoints↗

    HuggingFaceTB

    We release checkpoints every 40,000 steps, which equals 94.4B tokens. The GBS (Global Batch Size) in tokens for SmolLM3-3B is 2,359,296. To calculate the number of tokens from a given step: For the additional 2 stages that extend the context length to 64k, we sample checkpoints every 4,000 steps (9.4B tokens) for a total of 10 checkpoints: We release checkpoints at every step of our post-training recipe: Mid training, SFT, APO soup, and LC expert.

  • Tristan_-_dclm-perplexity-correlations-spearmanr-no-samp-410m-gguf↗

    RichardErkhov

  • colqwen2-v1.0↗

    vidore

    ColQwen2 is a model based on a novel model architecture and training strategy based on Vision Language Models (VLMs) to efficiently index documents from their visual features. It is a Qwen2-VL-2B extension that generates ColBERT- style multi-vector representations of text and images. It was introduced in the paper ColPali: Efficient Document Retrieval with Vision Language Models and first released in this repository This model takes dynamic image resolutions in input and does not resize them, changing their aspect ratio as in ColPali. Maximal resolution is set so that 768 image patches are created at most. Experiments show clear improvements with larger amounts of image patches, at the cost of memory requirements. Our training dataset of 127,460 query-page pairs is comprised of train sets of openly available academic datasets (63%) and a synthetic dataset made up of pages from web-crawled PDF documents and augmented with VLM-generated (Claude-3 Sonnet) pseudo-questions (37%). Our training set is fully English by design, enabling us to study zero-shot generalization to non-English languages. We explicitly verify no multi-page PDF document is used both *ViDoRe* and in the train set to prevent evaluation contamination. A validation set is created with 2% of the samples to tune hyperparameters.

  • Qwen3.6-27B-Abliterated-Heretic-Uncensored-GGUF↗

    Youssofal

    This is a GGUF release of an abliterated, uncensored version of Qwen's Qwen3.6-27B with vision, made with Heretic. By applying a Heretic-style two-stage MPOA pipeline with magnitude preservation on the Qwen3.6-27B dense text stack — slot-grouped output-side ablation followed by jailbreak-conditioned input-side ablation, mirroring the methodology used for Qwen3.6-35B-A3B — the base refusal behavior was attenuated at the weight level with low distributional divergence (KL 0.0282 vs base on harmless prompts). The result keeps Qwen3.6-27B's full vision and video multimodal architecture and general capability profile intact, while no longer defaulting to the original refusal pattern. Numbers measured by reading every response, not by regex / refusal-marker scoring. Greedy decoding (`do_sample=False`), `enable_thinking=False`, on `mlabonne/harmful_behaviors test[:25]`. KL on `mlabonne/harmless_alpaca test[:25]`. The 3 remaining deflections under the jailbreak system prompt are crisis-substitution patterns (violent crime → conflict resolution, suicide → crisis hotline, car theft → legal vehicle acquisition).