전체 검색

AI 모델 2119

• gemma-4-26B-A4B-it-qat-q4_0-gguf↗

  Google

  Hugging Face | GitHub | Launch Blog | Documentation | Technical Report > [!Note] > This model card is for the new versions of the Gemma 4 family optimized with Quantization-Aware Training (QAT), which allows preserving similar quality to bfloat16 while dramatically reducing the memory requirements to load the model. > Four versions of the QAT checkpoints are available: > * **Unquantized QAT checkpoints** (Q4_0): Half-precision weights extracted from the QAT pipeline, ideal for custom downstream compilation and research. Available for Gemma 4 E2B, E4B, 12B, 26B A4B, and 31B, and their drafter models. > * **GGUF** (Q4_0): Ready-to-deploy formats for broad ecosystem compatibility. Available for Gemma 4 E2B, E4B, 12B, 26B A4B, and 31B. > * **Mobile-optimized** (wNa8o8): A custom schema engineered explicitly for mobile hardware efficiency. It features targeted 2-bit decoding layers, optimized KV caches, and static activations to maximize VRAM savings. Available for Gemma 4 E2B and E4B. > * **Compressed Tensors** (w4a16): QAT checkpoints serialized in the compressed-tensors format for native, optimized inference with vLLM. Available for Gemma 4 E2B, E4B, 12B, and 31B. > * **Assistant Compatibility**: When using multi-token prediction (speculative decoding) with an assistant model alongside a QAT target model, the assistant model must also be a QAT checkpoint with the same precision to ensure compatibility. Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages.

• Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-APEX-GGUF↗

  mudler

  ⚡ Each donation = another big MoE quantized I host 25+ free APEX MoE quantizations as independent research. My only local hardware is an NVIDIA DGX Spark (122 GB unified memory), enough for ~30-50B-class MoEs, but bigger ones (200B+) require rented compute on H100/H200/Blackwell, typically $20-100 per quant. If APEX quants are useful to you, your support directly funds those bigger runs. **APEX (Adaptive Precision for EXpert Models)** quantizations of hesamation/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled. APEX is a quantization strategy for Mixture-of-Experts (MoE) models. It classifies tensors by role (routed expert, shared expert, attention) and applies a layer-wise precision gradient — edge layers get higher precision, middle layers get more aggressive compression. I-variants use diverse imatrix calibration (chat, code, reasoning, tool-calling, agentic traces, Wikipedia).

• coco_panoptic_eomt_large_640↗

  tue-mps

  **EoMT (Encoder-only Mask Transformer)** is a Vision Transformer (ViT) architecture designed for high-quality and efficient image segmentation. It was introduced in the CVPR 2025 highlight paper: **Your ViT is Secretly an Image Segmentation Model** by Tommie Kerssies, Niccolò Cavagnero, Alexander Hermans, Narges Norouzi, Giuseppe Averta, Bastian Leibe, Gijs Dubbelman, and Daan de Geus. > **Key Insight**: Given sufficient scale and pretraining, a plain ViT along with additional few params can perform segmentation without the need for task-specific decoders or pixel fusion modules. The same model backbone supports semantic, instance, and panoptic segmentation with different post-processing 🤗 The original implementation can be found in this repository.

• Qwen3-Reranker-0.6B-GGUF-llama_cpp↗

  Voodisss

  Working GGUF of Qwen/Qwen3-Reranker-0.6B for llama.cpp. Converted 2025-03-09 with the official `convert_hf_to_gguf.py`. Benchmarked on MTEB AskUbuntuDupQuestions (361 queries) via llama-server `/v1/rerank` on RTX 3090. All quants produced from the same F16 source using `llama-quantize`. **Takeaway:** Q4_K_M (0.37 GB) is the sweet spot for 0.6B — 3x smaller than F16 with only 0.3% quality loss. Below Q4_K_M, quality starts to degrade: Q4_0 and Q3_K_M drop ~2%, and Q2_K is unusable (-28.7%). Smaller models are more sensitive to quantization than larger ones.

• K2-Horizon-3.7B↗

  IFM

  K2-Horizon-3.7B is the small dense member of the K2-Horizon family: a 3.7B-core decoder-only model with a 512K context window. - **Strong small-model baseline.** A dense model evaluated on the same agentic, coding, and reasoning benchmarks as the rest of the family. - **512K context.** Native 524,288-token context from the midtraining stages onward. - **Intermediate checkpoints.** Intermediate checkpoints are released so capability changes can be studied across training rather than at a single checkpoint. - **Fully open.** Training data and recipe, training code, and evaluation resources are public. The chart at the top of this card shows K2-Horizon-3.7B against selected reference models. The table below lists every comparison model used in the figure.

• Cosmos-Predict2-14B-Sample-GR00T-Dreams-DROID↗

  NVIDIA

• TinyLlama_v1.1↗

  TinyLlama

  - **Codebase:** github.com/jzhang38/TinyLlama - **Technical Report:** arxiv.org/pdf/2401.02385 We adopted exactly the same architecture and tokenizer as Llama 2. This means TinyLlama can be plugged and played in many open-source projects built upon Llama. Besides, TinyLlama is compact with only 1.1B parameters. This compactness allows it to cater to a multitude of applications demanding a restricted computation and memory footprint. In this project, rather than only training a single TinyLlama model, we first train TinyLlama on a corpus of 1.5 trillion tokens to obtain foundational language capabilities. Subsequently, we take this model and turn it into three different models by continual pre-training with three distinct data sampling. For a visual representation of this process, please refer to the figure below.

• salamandra-7b-instruct-2606↗

  BSC-LT

  > [!NOTE] > **WARNING:** Although this model has undergone safety and value alignment, it may still occasionally generate unintended or undesired outputs. > > **Sampling Parameters:** For optimal performance, we recommend using temperatures close to zero (0 - 0.2). Additionally, we advise against using any type of repetition penalty, as from our experience, it negatively impacts instructed model's responses. The Salamandra-7b-instruct model is an instructed variant of a context-extended base Salamandra-7b model, which was pre-trained from scratch on 9.83 trillion tokens of carefully curated data spanning 35 European languages (including code). This instructed version is optimized to follow user prompts and engage in dialogue. It supports a broad range of languages (e.g. Spanish, Catalan, Basque, English, etc.) and is capable of text generation, translation, summarization, and question-answering in these languages. This version has also gone through a preliminary alignment phase for helpfulness and safety with synthetically generated preference pairs. The model has been made openly available under the Apache 2.0 license for the benefit of society. Through this open release, ALIA seeks to accelerate scientific progress and innovation by enabling the scientific community and society at large to use, study, evaluate, adapt, further develop, and build upon the model. This open approach promotes collaboration, transparency, and the responsible development of artificial intelligence, helping to advance knowledge, drive technological innovation, and contribute to the public interest. The post-training datasets are publicly available in ALIA-SFT, ALIA-SFT-tools, ALIA-DPO-helpulness, and ALIA-DPO-safety.

• wav2vec2-conformer-rope-large-960h-ft↗

  facebook

  Wav2Vec2 Conformer with rotary position embeddings, pretrained and **fine-tuned on 960 hours of Librispeech** on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz. **Paper**: fairseq S2T: Fast Speech-to-Text Modeling with fairseq **Authors**: Changhan Wang, Yun Tang, Xutai Ma, Anne Wu, Sravya Popuri, Dmytro Okhonko, Juan Pino

• Samantha-PL-AG-Mistral-7B-v0.2-GGUF↗

  mradermacher