Open-Weight Models Under Apache License 2.0

Model · Audio classification

cn_conformer_small_dr_vectors

Quan

model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs)

Model · Audio classification

cn_ecapa_tdnn_small_arcface

Quan

model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs)

Model · Audio classification

ls_conformer_small_cosface

Quan

model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs)

Model · Audio classification

ls_conformer_small_dr_vectors

Quan

model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs)

Model · Audio classification

ls_conformer_tiny_ge2e

Quan

model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs)

Model · Text generation

Qwen3-1.7B-GPTQ-Int4

AXERA

This version of Qwen3-1.7B-GPTQ-Int4 has been converted to run on the Axera NPU using w4a16 quantization. This model has been optimized with the following LoRA: For those who are interested in model conversion, you can try to export axmodel through the original repo: https://huggingface.co/Qwen/Qwen3-1.7B Convert the original Huggingface Qwen3-1.7B-GPTQ-Int4 to axmodel, and then apply the w4a16 quantization to get the final axmodel for axllm runtime. 方式二：一行命令安装（默认分支 axllm）： 方式三：下载Github Actions CI 导出的可执行程序（适合没有编译环境的用户）： https://github.com/AXERA-TECH/ax-llm/actions?query=branch%3Aaxllm 下载 最新 CI 导出的可执行程序（axllm），然后：

English financial decision model, approximately 0.42B parameters, with MLX CUDA support. This card describes the 6-rounds release (previously labeled four-source). Training data cover CHFS, CGSS, CFPS and CLDS. Macro F1 averages over all candidate classes within each supported question version, then equally over versions. 23,317 input tokens/s, 10.19 ms mean latency, 98.11 decisions/s. Measured on one A100 40GB with FP16, batch and concurrency one, the same three requests, three warmups and 20 timed calls per request; input throughput uses a shared tokenizer. 本页为 6-rounds 训练版本，旧标签为 four-source；数据来源为 CHFS、CGSS、CFPS、CLDS。3-rounds 训练版本使用独立模型卡，下载链接指向对应历史权重。按训练轮次命名，数据来源单独说明。

Model · Text generation

gemma-4-12b-it-coreml-128k

岡優志 / Okayuji

Run Google's Gemma 4 12B IT locally on an Apple silicon Mac with Core ML. One int4 bundle supports conversations up to 131,072 tokens, switching from its 32K mode to its 128K mode without reloading weights or copying the conversation cache. Requires: macOS 26+ · Xcode 26 / Swift 6.2 · ~11 GB disk · ~12 GB inference memory, 24 GB+ recommended. See Requirements for measured peaks. The script downloads this bundle to./models/gemma-4-12b-it-coreml-128k. The model guide also covers one-shot prompts and CLI flags. For the app, open CoreMLSamples.xcodeproj at the repository root and run the CoreMLSamples scheme on My Mac. For a standalone download, the existing Hub command is: The runtime supplies…

Model · Audio classification

ls_conformer_small_triplet

Quan

model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs)

Model · Audio classification

ls_ecapa_tdnn_base_ge2e

Quan

model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs)

Sev-4B scores supplied answers to questions about security logs and recovered programs. v0.3.0-response-policy-research adds explicit authorization-policy questions grounded in real SwarmTraces programs. It is a Qwen3.5-4B LoRA adapter with a decision head. It does not generate text. On 175 policy questions from 35 source programs held out of training, accuracy increases from 127/175, 72.6%, to 147/175, 84.0% compared with the published v0.2.0 parent. The user selected this research release with its measured tradeoffs. 26 of 28 registered checks pass. The false-alert ceiling and zero-new-DNS-error check fail. Those results are preserved unchanged. Install the Sev runtime. A generic…

基于 Qwen2.5-7B 的中医领域继续预训练与监督微调模型。 - 基座模型：Qwen/Qwen2.5-7B - 架构：Qwen2 - 训练方式：中医 CPT + SFT - 权重格式：Safetensors 本模型以 Qwen/Qwen2.5-7B 为基座。使用和分发时请同时遵守基座模型及相关训练数据的适用许可与条款。

Model · Zero-shot classification

Vela-2.0-0.8B

vLLM Semantic Router

Open Foundation Routing Models Routing decisions. Safety checks. Precise text spans. The compact hybrid member of Vela 2.0: a 756M-parameter model for multilingual routing, safety checks and span-level decisions through one interface. Define options, labels and rubrics at request time. Ask multiple named questions about a request, context and answer, and receive structured decisions with the text spans that support your workflow. 1. Compact deployment. About 3 GB of GPU memory for FP32 parameters, with a 16,384-token input limit. 2. Routing and safety together. Use one request for routing, prompt-attack checks, PII and unsupported-claim detection. 3. Decisions at span resolution. Return…

Model · Image and text to text

Xiaomi-Lens-0

SeerRay-Lab

Xiaomi-Lens-0 is a 4B multimodal language model built on Qwen3.5-4B. It adds a lightweight pixel-mask decoder and is trained with joint language and mask supervision. Each generated token identifies an image region that the decoder turns into a segmentation mask. The model supports referring expression comprehension and segmentation, multi-instance grounding, reasoning segmentation, embodied referring, and fine-grained visual perception. It can produce masks in its final answer or during reasoning. This repository contains the model weights, configuration, tokenizer, processor, and chat template. Inference uses the XiaomiLensQwen35ForConditionalGeneration class from the accompanying…

Model · Text generation

aiact-tier-4b-v0.1

Privify Inc

Author: Bhaskar Tallamraju and Vishal Saxena, Privify Inc. · Code and benchmark: github.com/Privify-Inc/aiact-tier · Fine-tuned from: Qwen3-4B (Alibaba Cloud / Qwen Team, Apache 2.0) · License: Apache 2.0. The base model's capabilities come from the Qwen team's work; please cite the Qwen3 Technical Report alongside this model. A 4B-parameter model, fine-tuned from Qwen3-4B with LoRA (Low-Rank Adaptation, a lightweight fine-tuning method), that classifies a described AI system under the EU AI Act (Regulation (EU) 2024/1689). It returns a risk tier and the provisions that apply, as JSON. Provided as is, with no warranty and no liability (Apache 2.0, sections 7 and 8). Not legal advice, not a…

Complete checkpoint with global LoRA already merged, followed by bitsandbytes NF4 quantization with double quantization and BF16 compute. No separate adapter is required. lmhead is excluded from quantization; existing quantization coverage of the other components is preserved. Corresponding merged BF16 checkpoint. The existing quantized weights are retained: recomputing NF4 from the matching merged BF16 checkpoint verified all 2,792 serialized tensors as identical. Use the Emomni-v1 code: The custom model is loaded by src.modelingemomni.EmomniModel, rather than the generic AutoModelForCausalLM loader. Local text/audio loading and generation were checked. These checks are not a new quality…

Model · Audio classification

ls_conformer_small_ext_softmax

Quan

model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs)

https://doi.org/10.1007/s10694-026-02000-3 Existing vision-based methods suffer from high false alarm rates in urban flame detection. Applying Multimodal Large Language Models (MLLMs) for secondary filtering shows great potential in reducing false alarms, yet they have high inference latency and are prone to reasoning collapse on negative samples without explicit Chain-of-Thought (CoT) guidance. To overcome these challenges, this study proposed Flash-Cascade, the first sub-second MLLM-based firewall to leverage CoT to efficiently filter false alarms. We deconstructed the flame detection process into four logical stages (planning, observation, analysis, and judgment), which informed the…

Model · Image and text to text

Qwen3.8-Flash-Next-Abliterated-Uncensored-Turbo-Brilliance-50x-Reasoning-Instruct-GSQ-RCO-GGUF

David Belton

generating better answers. This model has significant internal upgrades under user control that are "on the fly" (you set in your prompt(s) and/or API). Please read this model card carefully to get the most from it - including all the reasoning and instruct modes (27+27+9 xhigh) that feature model suggestions ("context help") for all use cases. The Turbo Brilliance system will begin appearing in other model sizes/types shortly (IE: Qwen, Gemma, Mistral, GLM, Llama,... including our strongest fine tunes too). (TIP: You can run this model on 12GB GPUs and up too "at speed".) - 180 Billion parameters with 512 experts, 10 activated per token. - SOTA BF16 performance in "IQ3S" using quant tuning…

This BF16 conversion of Qwen3-ASR-1.7B-hf runs with the main branch of ninfer-extended. Audio encoding, language prefill and decode run in the native C++/CUDA engine. The artifact includes the tokenizer, processor and chat template, with all 707 source parameters preserved byte for byte. Also included: a native BF16 Qwen3-ForcedAligner-0.6B artifact for word timestamps. See the combined ASR/alignment workflow and measurements below. On 60-second English audio, NInfer processes 62.5 seconds of audio per second on RTX 5070 Ti and 87.7 seconds per second on RTX 6000D. Four 30-second English recordings reach 197.8 and 299.0 audio seconds per second, respectively. Audio seconds/s = total input…

Model · Image and text to text

Qwen3-VL-4B-Instruct-LoRA-AX650

AXERA

Ready-to-run package for Qwen/Qwen3-VL-4B-Instruct on AX650 / NPU3. It includes an AX650 aarch64 axllm server, 36 compiled text layers, a fixed-shape image encoder, and two runtime-selectable LoRA adapters. This release supports text chat and single-image requests through the OpenAI-compatible chat API. - AX650 / AX650N aarch64; the results below were measured on an AX650 / NPU3 board. The image encoder contributes 144 visual soft tokens per image: (384 / 16 / 2)² = 144, using the packaged 16-pixel patch size and spatial merge size of 2. The total input-token count also includes the text prompt and chat-format tokens. Keep the full request, including visual tokens, within the prefill and…

https://doi.org/10.1007/s10694-026-02000-3 Existing vision-based methods suffer from high false alarm rates in urban flame detection. Applying Multimodal Large Language Models (MLLMs) for secondary filtering shows great potential in reducing false alarms, yet they have high inference latency and are prone to reasoning collapse on negative samples without explicit Chain-of-Thought (CoT) guidance. To overcome these challenges, this study proposed Flash-Cascade, the first sub-second MLLM-based firewall to leverage CoT to efficiently filter false alarms. We deconstructed the flame detection process into four logical stages (planning, observation, analysis, and judgment), which informed the…

Complete Emomni checkpoint with global LoRA merged into the model weights. The checkpoint includes its tokenizer, chat template, Whisper feature extractor, and generation configuration. No separate LoRA adapter is required. This release replaces the earlier audio-scoped LoRA checkpoint with the global-LoRA checkpoint; it is not an equivalent merge of the earlier audio-scoped model. Use the Emomni-v1 code: The custom model is loaded by src.modelingemomni.EmomniModel, rather than the generic AutoModelForCausalLM loader. The original export verified all 837 saved tensors and matching text/audio greedy outputs after reloading. These checks verify checkpoint serialization; they are not a new…

Model · Audio classification

ls_conformer_base_ge2e

Quan

model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs)

A quantized version of Qwen/Qwen3.5-4B for Jetson AGX Orin 64GB / CUDA SM87, deployed with Orinfer v0.2.0. It supports text, images and multiple images, streaming Chat Completions, native MTP, prefix caching, continuous batching and resumable JSONL offline inference. Translation requests use enablethinking: false; thinking follows the original model's chat template. Two-stage residual E8P Q4 with coefficients [4, 1], signed-block128 rotation and beam-8 encoding over the original BF16 weights, paired with INT8 computation. Sensitive tensors retain the validated W8/FP16/BF16/FP32 mixed-precision policy. Deployed read-only GPU data, including scales, derived constants and precision exceptions…

This is the LoRA adapter form (40 MB) of lododo-Distill-v1.0. The model itself: a distilled fine-tune of the Qwen3.6-35B-A3B base (MoE, 3B active), trained on data generated by Qwen3.8-27B as the teacher — bringing the base to (and beyond) the teacher's capability while keeping the MoE speed advantage. moefast is a small launcher that configures llama.cpp for your machine. It reads the model's GGUF metadata, measures your free VRAM, computes the expert-cache size, context and batch parameters, writes them into an editable config file next to the model, and starts llama-server. The expert cache itself — keeping expert weights in host RAM and caching the hot ones in VRAM — is llama.cpp's own…

Model · Image and text to text

Qwen3.8-Flash-Next-Turbo-Brilliance-50x-Reasoning-Instruct-GSQ-RCO-GGUF

David Belton

generating better answers. This model has significant internal upgrades under user control that are "on the fly" (you set in your prompt(s) and/or API). Please read this model card carefully to get the most from it - including all the reasoning and instruct modes (27+27+9 xhigh) that feature model suggestions ("context help") for all use cases. The Turbo Brilliance system will begin appearing in other model sizes/types shortly (IE: Qwen, Gemma, Mistral, GLM, Llama,... including our strongest fine tunes too). (TIP: You can run this model on 12GB GPUs and up too "at speed".) - 180 Billion parameters with 512 experts, 10 activated per token. - SOTA BF16 performance in "IQ3S" using quant tuning…

Model · Image and text to text

REX-Intern-S2-Cyber-F3

AI45Research

A REX checkpoint of Intern-S2 with WMDP-Cyber knowledge relocated into 8 removable experts per MoE layer (264 experts in total). Excising these experts together with their router rows yields A3, a standard 256-expert model with that knowledge removed. This repository holds the expanded checkpoint F3: the REX model after 250M tokens of standard full-parameter SFT, with the removable experts still attached. A3 differs from F3 only by excision. CPU only and about 1 min, with at least 16 GiB RAM. Disk: 70 GiB for F3 plus 69 GiB for A3../A3 must not exist beforehand, and expertexpansionmanifest.json must stay in./F3. Load./A3 with the model class named in its config.json. Intern-S2 ships a…

This model is a fine-tuned version of ntu-spml/distilhubert on the GTZAN dataset. It achieves the following results on the evaluation set: - evalloss: 0.8970 - evalmodelpreparationtime: 0.0018 - evalaccuracy: 0.87 - evalruntime: 443.4759 - evalsamplespersecond: 0.225 - evalstepspersecond: 0.029 The following hyperparameters were used during training: - learningrate: 5e-05 - trainbatchsize: 8 - evalbatchsize: 8 - lrschedulertype: linear - lrschedulerwarmupsteps: 100 - numepochs: 10 - mixedprecisiontraining: Native AMP - labelsmoothingfactor: 0.1 - Transformers 5.16.1 - Pytorch 2.11.0+cpu - Datasets 2.19.0 - Tokenizers 0.23.1

Model · Image and text to text

flux-3-image-klein-pup-GGUF

Unsloth AI

GGUF conversions of black-forest-labs/flux-3-image-klein-pup, the prompt upsampler for FLUX 3 Image Klein. It is a fine-tune of Qwen3.5-9B by Black Forest Labs that expands a short image request (and, for edits, the reference images) into the detailed caption FLUX 3 Image Klein is trained on. FLUX 3 Image Klein is meant to run with a prompt upsampler in front of it. Short prompts sent to the image model as typed often give generic images and broken poses (backflips, handstands, tangled arms); the upsampled caption describes the pose and every limb explicitly and fixes most of those. Unsloth Desktop and Unsloth Studio download and run this upsampler automatically for FLUX 3 Image Klein…

Model · Audio classification

ls_conformer_small_ge2e_contrast

Quan

model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs)

Modelos afinados (fine-tuned) para operar como el cerebro autónomo de OzyAssist en Windows. - Eliminación Total de Alucinaciones: Cero bucles de saltos de línea (\n\n\n) ni copias textuales del System Prompt.

Model · Image and text to text

REX-Qwen3.5-35B-A3B-Cyber-F3

AI45Research

A REX checkpoint of Qwen3.5-35B-A3B with WMDP-Cyber knowledge relocated into 8 removable experts per MoE layer (264 experts in total). Excising these experts together with their router rows yields A3, a standard 256-expert model with that knowledge removed. This repository holds the expanded checkpoint F3: the REX model after 250M tokens of standard full-parameter SFT, with the removable experts still attached. A3 differs from F3 only by excision. CPU only and about 1 min, with at least 16 GiB RAM. Disk: 69 GiB for F3 plus 67 GiB for A3../A3 must not exist beforehand, and expertexpansionmanifest.json must stay in./F3. Load./A3 with the model class named in its config.json. Scores in %, ±…

Model · Text generation

ProKope-7B

ProKope-AI

ProKope-7B is the #1 ranked open-weight reasoning model in the world in the 7B parameter class on the official Hugging Face Open LLM Leaderboard v2 (Evaluation Snapshot: October 2026). Developed by ProKope AI, ProKope-7B was engineered to solve one of the fundamental bottlenecks of sub-10B language models: achieving elite multi-step deductive reasoning, graduate-level scientific problem-solving, and robust instruction compliance without suffering from catastrophic reasoning collapse or narrow benchmark memorization. Evaluating across EleutherAI's rigorous lm-evaluation-harness Leaderboard v2 gauntlet, ProKope-7B achieves an official 44.75% Composite Score, setting the highest benchmarked…

Model · Image and text to text

Qwen3.8-27B-Human-KO-Enterprise-Boundary

ThakiCloud

Enterprise policy decision model — Korean-supervised, zero-shot English. The boundary-targeted checkpoint from When Should Enterprise Policy Decisions Be Learned, and When Should They Be Reasoned? Merged full weights — not an adapter — with the pointer head shipped alongside. It decides which one of seven actions an enterprise agent should take under a written policy, before any text is generated: trained on contrastive boundary pairs — records differing in one decision-relevant factor across three boundaries (answer vs. call a tool, call a tool vs. ask for confirmation, missing information) — against a control given the same number of tokens of randomly sampled in-domain data. Data was the…

Model · Image and text to text

REX-Qwen3.5-35B-A3B-Bio-F3

AI45Research

A REX checkpoint of Qwen3.5-35B-A3B with WMDP-Bio knowledge relocated into 8 removable experts per MoE layer (264 experts in total). Excising these experts together with their router rows yields A3, a standard 256-expert model with that knowledge removed. This repository holds the expanded checkpoint F3: the REX model after 250M tokens of standard full-parameter SFT, with the removable experts still attached. A3 differs from F3 only by excision. CPU only and about 1 min, with at least 16 GiB RAM. Disk: 69 GiB for F3 plus 67 GiB for A3../A3 must not exist beforehand, and expertexpansionmanifest.json must stay in./F3. Load./A3 with the model class named in its config.json. Scores in %, ±…

Bielik-11B-v3.0-Instruct post-trained for the Polish history matura (extended level) during the Warsaw Model Trainers alpha=64) on top of speakleash/Bielik-11B-v3.0-Instruct. closedem = exact match on closed items. Graded scores (deepseek-flash-high@think via benchmark/bench.py) are reported separately.

Complete Emomni checkpoint with global LoRA merged into the model weights. The checkpoint includes its tokenizer, chat template, Whisper feature extractor, and generation configuration. No separate LoRA adapter is required. The global-LoRA checkpoint is now distributed with its adapters already merged. Use the Emomni-v1 code: The custom model is loaded by src.modelingemomni.EmomniModel, rather than the generic AutoModelForCausalLM loader. The original export verified all 543 saved tensors and matching text/audio greedy outputs after reloading. These checks verify checkpoint serialization; they are not a new quality benchmark.

LoRA-адаптер для Qwen/Qwen3-4B (режим без размышлений, enablethinking=False) — помощник Moliq по банковским услугам Узбекистана. - Обучение: QLoRA r=16, α=32, все линейные слои; 1 эпоха на 2 219 примерах (Kaggle, T4, 6 ч), v2: вопросы по банку и виду кредита, меньше ложных отказов. - Данные: вопросы и ответы, собранные из официальных данных ЦБ — портал bankxizmatlari.uz (34 банка: реквизиты, кредиты, вклады, карты, курсы) и finlit.uz (FAQ, статьи). Ответ строится только по контексту со ссылками [n]; ~10% примеров — отказ, если ответа в контексте нет. - Работает в связке с поиском moliqrag.py (тот же формат промпта) — Space Azamaka/moliq-ai.

Model · Text to image

pimp-my-pony-gguf

Chuck

GGUF-quantized versions of huchukato/pimp-my-pony — an SDXL/Pony-based merge — for use with Each model variant lives in its own folder, with multiple quantization levels. More variants will be added over time. These files contain only the UNet — CLIP and VAE are not included. Requires the ComfyUI-GGUF custom node (pip install --upgrade gguf). SDXL is conv2d-heavy, so gains are smaller than on transformer models (Flux/Wan) — but still meaningful on limited VRAM / unified memory. Recommended generation settings are the same as the clip skip 2, included VAE, avoid excessive negative prompts. Converted with the official UNet extracted to safetensors → convert.py (BF16 GGUF) → llama-quantize…

Model · Text to speech

Qwen3-TTS-12Hz-1.7B-CustomVoice-CoreML

岡優志 / Okayuji

Three Core ML packages for decoding audio tokens, encoding reference audio, and extracting a speaker embedding from the Qwen3-TTS family. These are audio components, not a complete text-to-speech model. For: developers integrating the components into an Apple Silicon Mac application. Not for: generating speech from text with this download alone; the language model and host pipeline are separate. Platform: Apple Silicon (M1 and later), Core ML. This is a.mlpackage release, separate from Core AI.aimodel releases. Compile and load the packages with Apple's Core ML workflow, then use the exact interfaces below. For the complete upstream TTS pipeline and language-model weights, start with…

Model · Image and text to text

REX-Gemma-4-26B-A4B-Cyber-F3

AI45Research

A REX checkpoint of Gemma-4-26B-A4B with WMDP-Cyber knowledge relocated into 8 removable experts per MoE layer (136 experts in total). Excising these experts together with their router rows yields A3, a standard 128-expert model with that knowledge removed. This repository holds the expanded checkpoint F3: the REX model after 250M tokens of standard full-parameter SFT, with the removable experts still attached. A3 differs from F3 only by excision. CPU only and about 1 min, with at least 16 GiB RAM. Disk: 52 GiB for F3 plus 49 GiB for A3../A3 must not exist beforehand, and expertexpansionmanifest.json must stay in./F3. Load./A3 with the model class named in its config.json. Scores in %, ±…

This BF16 conversion of Xiaomi-OCR-0 supports OCR and document parsing. The source model is based on Qwen3.5-0.8B-Base. The artifact includes language and vision weights, the tokenizer and image-processing configuration. Build the ninfer-extended main branch to run this artifact. This starts an OCR service with 32K context per request, four concurrent requests and BF16 weights/KV. Build dependencies: 64-bit Linux (WSL2 on Windows), CUDA supporting sm120a, C++20, CMake ≥3.28, Ninja, FFmpeg development libraries, libcurl, PCRE2, cuDNN 9 and cuBLAS. Omit CUBLASROOT when cuBLAS is installed with CUDA. Both engines use BF16 weights/KV, 4K context, one request, identical images/prompts, greedy…

Complete checkpoint with global LoRA already merged, followed by bitsandbytes NF4 quantization with double quantization and BF16 compute. No separate adapter is required. lmhead is excluded from quantization; existing quantization coverage of the other components is preserved. Corresponding merged BF16 checkpoint. This release is freshly quantized from the matching merged BF16 checkpoint, replacing the earlier quantized snapshot from a different checkpoint. Use the Emomni-v1 code: The custom model is loaded by src.modelingemomni.EmomniModel, rather than the generic AutoModelForCausalLM loader. Local text/audio loading and generation were checked. These checks are not a new quality benchmark.

LoRA adapter for google/gemma-4-12B-it-qat-q40-unquantized trained to answer the Polish history matura (poziom rozszerzony, CKE). Built at the Warsaw Model Trainers Hackathon (team repo: WarsawModelTrainersHackathon, src/posttrain/). Supervised fine-tuning, bf16 LoRA on the QAT weights (deployed on the official Q40 GGUF). One mixed pass over every source: short-key items (real/rephrased/synthetic exam items, textbook questions) trained with thinking OFF (question -> CKE answer), gemma-4-31b@think teacher traces (judge-filtered, exam items + essays) trained with thinking ON; exam-like rows oversampled x2-x3. Loss on the answer only. LoRA r=32, alpha=64, all language-model linear layers, lr…

LoRA adapter for google/gemma-4-12B-it-qat-q40-unquantized trained to answer the Polish history matura (poziom rozszerzony, CKE). Built at the Warsaw Model Trainers Hackathon (team repo: WarsawModelTrainersHackathon, src/posttrain/). Vision SFT (QLoRA, 4-bit base): image-based exam items (real CKE papers with their source images, synthetic Wikimedia image tasks, synthetic full papers) with the PNGs fed to the model, plus 30% of the teacher traces; vision tower frozen, language LoRA only. LoRA r=32, alpha=64, lr 1e-4, 1 epoch, batch 1 x gradaccum 16, maxlen 6144, Unsloth vision collator Consolidated training set data/trainingdatasets/sfttrain.jsonl (buildsft.py): real CKE papers 2005–2026…

Model · Text classification

OpenJudgement-4B-Preview

Kitani

Experimental open-weights judgment model by Kitani OpenJudgement is unfinished. We're releasing this checkpoint for people to experiment with, inspect, and build on. It still needs work on judgment quality, calibration, and inference efficiency. It is not as good as Jev overall in our internal task comparisons. It does show a meaningful improvement over untouched Qwen on our recorded validation comparison: 75.4% versus 64.2% annotation agreement. That is a result on a particular evaluation set, not a claim that we beat the base model on every task. There are questions it handles well and questions it confidently gets wrong. Please judge the preview by your own examples rather than assuming…

An 11 GB GGUF of Sarvam-30B that runs on a CPU with about 12 GB of free RAM. It scores 81.7% on MMLU, against 85.1 published for the full model. Sarvam-30B is a mixture-of-experts model: 32 billion weights, about 2.4 billion of them active per token, which is why a CPU can run it. Use the IQ2M file. The IQ1M file is here for comparison: it saves 3 GB but loses much more quality. For the larger Sarvam-105B, see Aaravomega/sarvam-105b-GGUF-IQ2M-IQ1M (2-bit and 1-bit) and Aaravomega/sarvam-105b-GGUF-Q80 (8-bit). You need a recent llama.cpp build with Sarvam support (architecture bailingmoe2; tested with release b11731) and about 12 GB of free RAM. Then open http://127.0.0.1:8080, or call the…

This is a mixed-precision, inference-ready quantization of allenai/MolmoPoint-8B. The 7.6B-parameter language transformer uses bitsandbytes NF4 with BF16 compute and double quantization. The vision encoder, multimodal connector, visual embedding projection, and pointing head remain BF16 to preserve pointing behavior. This is intentionally not an entirely 4-bit model. Quantizing the visual path caused substantial pointing degradation during recipe selection. The figure compares both released quantizations with the BF16 baseline. See Evaluation for coverage, protocol, and metric details. and pointing head The included quantizationprovenance.json records the resolved base revision, complete…

Model · Text classification

Decision-1.0-Route-0.6B

vLLM Semantic Router

One Decision model for the request-time signals of vLLM Semantic Router: subject area, output modality and user feedback as Choice questions, and prompt attack, harmful request, twelve hazard categories, fact-check need, personal data, tool need and (on a response) unsupported claims as Noul questions. It is Decision-1.0-Kai-0.6B with its Choice and Noul paths fine-tuned on a corpus-matched suite built for these signals (semantic-router#4305). These weights replace the first release. They come from a tuning round that changed one thing at a time and let the dev split decide each change: one Noul epoch instead of three, Choice options trained as KEY: description (the form the Decision…

Model · Text classification

kev-8b

Jared Palmer

kev-8b is a decision model: one document (the state) and a set of typed questions in, a probability distribution per question out, in one forward pass. No text generation. It is a LoRA adapter (r=16) plus a pointer head on Qwen/Qwen3-8B-Base (revision 49e3418f), serving TypeSafe's public /v1/systemone contract. Research preview, not a versioned release. It is the best checkpoint of any size under a frozen, checksummed protocol (best in-distribution accuracy, best Brier), trained with the low-learning-rate recipe found at 4B. It is the best out-of-domain kev (0.796 on transfer-v4 dev, 6 pp from Jev). It misses the release screen we set in advance by one pair (held-out policy pairs 0.69…

Model · Image and text to text

Mitsuba-ComfyUI-27B-GGUF

Yoshinori Sugishita

A ternary (1.58-bit) Qwen3.8-27B tuned for ComfyUI work: writing image/video generation prompts that follow strict conditions, and describing images. It is not for coding. ComfyUI 向けに調整した、Qwen3.8-27B の三値（1.58 ビット）モデルです。システムプロンプトにそった画像・動画用プロンプトの作成と、画像の説明が得意です。コーディングには向きません。 - Self-made ternarization of the official Qwen3.8-27B weights (not derived from Bonsai's weights), stored in Prism ML's PQ20 / PTQ10 GGUF formats. - 7.3 GB (PQ20) / 6.0 GB (PTQ10). Runs on a single 16 GB GPU. Measured with the same questions and conditions for all four models. Details: EVALUATION.md. The last column is the original, un-quantized Qwen3.8-27B (BF16), shown as the reference point: it shows what the…

This is an inference-ready bitsandbytes LLM.int8 quantization of allenai/MolmoPoint-8B. All eligible linear modules are quantized except the 7.5M-parameter pointing head, which remains BF16. This avoids a runtime patch because bitsandbytes 0.50.0's INT8 kernel does not natively accept the pointing head's high-rank inputs. The figure compares both released quantizations with the BF16 baseline. See Evaluation for coverage, protocol, and metric details. the model) The included quantizationprovenance.json records the resolved base revision, complete quantization recipe, software versions, and build environment. The checkpoint was built and validated with Python 3.12, Transformers 4.57.1…

Model · Text generation

Qwen3.8-27B-BaseQuant-V1

Base Compute

Quantized builds of Qwen/Qwen3.8-27B in three sizes. On held-out text, each one is closer to the BF16 model than the unsloth dynamic quant of the same size. Each GGUF is no larger than its unsloth counterpart. The Small bundle needs baseRT 0.4.0 or later. KL divergence of each quant's next-token distribution from the BF16 model, ours beside unsloth's quant of the same size, measured in llama.cpp. Lower is closer to BF16. The percentage over each pair is our change against unsloth. On text that was not used to build these files, all three are 17–34% closer to BF16 than unsloth's quant of the same size. Each cell: unsloth KL / BaseQuant KL (change). Bold marks the lower KL. The prompts come…

Model · Image and text to text

Qwen3.8-27B-UltimateDetails2-stage1__The-Harley-Pelican

David Belton

https://huggingface.co/collections/DavidAU/turbo-brilliance-ai-constructs-50x-reasoning-instruct Even at stage 1, these versions are exceeding Qwen 3.8 core metrics on all fronts... and we are just getting started. Now with three branches, Ultimate 1, 2 and 3. ULT3 - has 3 components built and now in testing. - more components being built. - ULT1 and ULT2 will also share these (parts and assembled) too. Metrics posted below. 2 huge "How to save the planet" examples (67 PAGES!) posted below too. See also the SVG "Harley Pelican" below too. - This model (and a few more) will soon have 12 reasoning and 12 instruct modes plus interactive help, model embedded system to select the best…

本仓库提供由 Qwen/Qwen3-0.6B 转换得到的 RKNN LLM 模型。 - Model ID：RKNNAI/RK182X-LLM-Qwen3-0.6B - 模型显示名称：RK182X-LLM-Qwen3-0.6B - 源模型：Qwen/Qwen3-0.6B - 模型类型：LLM ModelScope 完整下载： Hugging Face 完整下载： ModelScope 指定配置下载： Hugging Face 指定配置下载： - 使用配套 RKNN Runtime 和驱动；运行前用 rknn-smi -v 检查设备端版本。 源模型许可证见根目录 LICENSE；同时遵守 RKNN Toolkit 和 RKNN Runtime 许可条款。

本仓库提供由 Qwen/Qwen3.5-0.8B 转换得到的 RKNN LLM 模型。 - Model ID：RKNNAI/RK182X-LLM-Qwen3.5-0.8B - 模型显示名称：RK182X-LLM-Qwen3.5-0.8B - 源模型：Qwen/Qwen3.5-0.8B - 模型类型：LLM ModelScope 完整下载： Hugging Face 完整下载： ModelScope 指定配置下载： Hugging Face 指定配置下载： - 使用配套 RKNN Runtime 和驱动；运行前用 rknn-smi -v 检查设备端版本。 源模型许可证见根目录 LICENSE；同时遵守 RKNN Toolkit 和 RKNN Runtime 许可条款。

本仓库提供由 Qwen/Qwen3.5-2B 转换得到的 RKNN LLM 模型。 - Model ID：RKNNAI/RK182X-LLM-Qwen3.5-2B - 模型显示名称：RK182X-LLM-Qwen3.5-2B - 源模型：Qwen/Qwen3.5-2B - 模型类型：LLM ModelScope 完整下载： Hugging Face 完整下载： ModelScope 指定配置下载： Hugging Face 指定配置下载： - 使用配套 RKNN Runtime 和驱动；运行前用 rknn-smi -v 检查设备端版本。 源模型许可证见根目录 LICENSE；同时遵守 RKNN Toolkit 和 RKNN Runtime 许可条款。

本仓库提供由 Qwen/Qwen3-ASR-0.6B 转换得到的 RKNN ASR 模型。 - Model ID：RKNNAI/RK182X-ASR-Qwen3-ASR-0.6B - 模型显示名称：RK182X-ASR-Qwen3-ASR-0.6B - 源模型：Qwen/Qwen3-ASR-0.6B - 模型类型：ASR ModelScope 完整下载： Hugging Face 完整下载： ModelScope 指定配置下载： Hugging Face 指定配置下载： - 使用配套 RKNN Runtime 和驱动；运行前用 rknn-smi -v 检查设备端版本。 源模型许可证见根目录 LICENSE；同时遵守 RKNN Toolkit 和 RKNN Runtime 许可条款。

本仓库提供由 Alibaba-NLP/gme-Qwen2-VL-2B-Instruct 转换的 RKNN 模型。 - Model ID：RKNNAI/RK182X-EMBEDDING-gme-Qwen2-VL-2B-Instruct - 模型显示名称：RK182X-EMBEDDING-gme-Qwen2-VL-2B-Instruct - 源模型：Alibaba-NLP/gme-Qwen2-VL-2B-Instruct - 模型类型：EMBEDDING ModelScope 完整下载： Hugging Face 完整下载： ModelScope 指定配置下载： Hugging Face 指定配置下载： - 使用配套 RKNN Runtime 和驱动；运行前用 rknn-smi -v 检查设备端版本。 - 用于图文检索，输出 embedding 向量。 - 源模型许可证：Apache License 2.0，正文见 LICENSE，归属与转换修改说明见 NOTICE。 - 同时遵守 RKNN Toolkit、RKNN Runtime 相关许可；本包不包含运行库。