Open-Weight Models Under Apache License 2.0

Model · Text generation

Iris-mini-APEX-I-MiniPlus-V2.1-GGUF

IsValorum

We maintain our previous releases publicly as a transparent engineering record of continuous optimization. Below is the exact evolutionary roadmap of our MiniPlus architectures: The APEX-I-MiniPlus quantization architecture powering this model was subjected to an extensive independent evaluation by Japanese AI researcher and evaluator zephel01 (CoolZero) on an NVIDIA RTX 5090 (32GB) workstation running llama.cpp CUDA b11027 with FlashAttention (-fa on -ctk q80 -ctv q80 -ngl 99). The evaluation tested the APEX-I hybrid MoE engine across 348 unseeded trials on SWE-bench style multi-file Python bug-fixing tasks with hidden pytest suites (llmbench): - L6 Multi-File Code Generation (60 tasks)…

Model

DualMind-GGUF

Convergent Intelligence

GGUF quantizations of DualMind for local inference via llama.cpp, Ollama, LM Studio, and other GGUF-compatible runtimes. DualMind is a 1.7B parameter model that implements a dual-cognition reasoning architecture: The model learns to reason freely, then critique its own reasoning, then produce a final answer. Multi-model dialectics collapsed into shared weights. Training lineage: Qwen3-1.7B → DistilQwen3 (uncensored) → Disctil (DISC-refined) → TKD from Qwen3-30B-A3B-Thinking → DualMind SFT on LogicInferenceOA dataset. - temperature: 0.6 - topp: 0.9 - repeatpenalty: 1.3 (important — prevents enumeration loops) - numpredict: 512–1024 - DualMind — source model (SafeTensors)…

Model · Video text to text

VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored-GGUF

Prithiv Sakthi

C/C++ — https://github.com/ggml-org/llama.cpp

This qwen2 model was trained 2x faster with Unsloth and Huggingface's TRL library.

Model

DualMinded-Qwen3-1.7B-GGUF

Convergent Intelligence

GGUF quantizations of DualMinded-Qwen3-1.7B for local inference via llama.cpp, Ollama, LM Studio, and other GGUF-compatible runtimes. DualMinded-Qwen3-1.7B is the Opus-trained variant of the DualMind architecture. While DualMind was trained on LogicInferenceOA, DualMinded was trained on Opus-4.6-Reasoning-3000x-filtered — high-quality reasoning traces from Claude Opus 4.6. The Opus training data provides longer, more structured reasoning chains. The thinking column maps directly to the phase without heuristic splitting, producing cleaner cognitive transitions. Training lineage: Qwen3-1.7B → DistilQwen3 → Disctil → TKD checkpoint-512 → DualMind SFT v2 on Opus-4.6-Reasoning. Both share the…

Model · Text generation

StandardOne-3B-GGUF

Standard Thinking

GGUF builds of Standard One 3B (Ministral 3 3B text + Pixtral vision tower, mistral3 architecture) for use with llama.cpp. Most quant levels below (marked "imatrix") were built with an importance matrix calibrated on 1,512 prompts sampled from our own training rows (see "Importance-matrix calibration" below), which recovers some of the accuracy quantization would otherwise lose. Q80 and BF16 don't need one. SHA256 checksums: SHA256SUMS. Source revisions, conversion tool version and full validation Q4KM note: this is the imatrix-calibrated version, not a plain quantization. We generated both and chose whichever scored higher on the mean of 10 held-out and public-dataset decision suites (no…

Model · Text generation

StandardOne-8B-GGUF

Standard Thinking

GGUF builds of Standard One 8B (Ministral 3 8B text + Pixtral vision tower, mistral3 architecture) for use with llama.cpp. Most quant levels below (marked "imatrix") were built with an importance matrix calibrated on 1,512 prompts sampled from our own training rows (see "Importance-matrix calibration" below), which recovers some of the accuracy quantization would otherwise lose. Q80 and BF16 don't need one. SHA256 checksums: SHA256SUMS. Source revisions, conversion tool version and full validation Q4KM note: this is the imatrix-calibrated version, not a plain quantization. We generated both and chose whichever scored higher on the mean of 10 held-out and public-dataset decision suites (no…

Model · Fill mask

AraBART

Moussa Kamal Eddine

AraBART is the first Arabic model in which the encoder and the decoder are pretrained end-to-end, based on BART. AraBART follows the architecture of BART-Base which has 6 encoder and 6 decoder layers and 768 hidden dimensions. In total AraBART has 139M parameters. AraBART achieves the best performance on multiple abstractive summarization datasets, outperforming strong baselines including a pretrained Arabic BERT-based models and multilingual mBART and mT5 models.

Model · Token classification

stanza-zh-hans

Stanford NLP

Stanza is a collection of accurate and efficient tools for the linguistic analysis of many human languages. Starting from raw text to syntactic analysis and entity recognition, Stanza brings state-of-the-art NLP models to languages of your choosing. Find more about it in our website and our GitHub repository. This card and repo were automatically prepared with huggingstanza.py in the stanfordnlp/huggingface-models repo

Model · Text generation

Qwen3.8-27B-PaoAI-ROCmFP4-STRIX-BALANCED-GGUF

Edward Paolo Guevarra

The balanced Qwen3.8-27B quant for AMD Strix Halo (Ryzen AI Max+ 395 / gfx1151). Measured, not marketed: one 15 GB file on the iGPU, the full 262,144-token window integrity-checked, three serving profiles — reading 101–410 t/s and writing 14.8–49.8 t/s across the window (depending on how full it is, what you ask for, and the profile). Reading = processing your prompt (prefill). Writing = producing the answer (decode). t/s = tokens per second. Brand: PaoAI · community quant, not affiliated with Alibaba or the Qwen team. Qwen3.8-27B is a hybrid: 48 cheap linear-attention layers and only 16 full-attention layers that actually re-read your context. We compressed it part by part: the finer…

weighted/imatrix quants of https://huggingface.co/racineai/Mistral-Military-24B For a convenient overview and download list, visit our model page for this model. static quants are available at https://huggingface.co/mradermacher/Berthier-Mistral-Military-24B-GGUF This is a vision model - mmproj files (if any) will be in the static repository. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts…

Model · Text generation

Qwen3.5-ANE-CoreML

Takanori Ishikawa

Run Qwen3.5 text models on Apple Silicon with Core ML. These packages use 6-bit compressed linear weights and include a four-token prompt-prefill graph alongside the one-token decoder. The repository contains nine model/context packages: 0.8B at 2,048 tokens; 2B at 2,048 and 16,384 tokens; and 4B and 9B at 2,048, 8,192, and 16,384 tokens. This is an unofficial conversion and is not affiliated with or endorsed by the Qwen team. It is based on the original Qwen3.5-4B, and Qwen3.5-9B models. - Designed for Apple Neural Engine execution. Core ML packages split the model into chunks and keep the shared token embedding table on the host. - Compact weights. Linear layers use 6-bit k-means…

Model · Text generation

Lythri-7B-A4B-GGUF

Lythri

This repository contains GGUF quantizations of Lythri-7B-A4B for use with llama.cpp and compatible apps. For model details, evaluation results and limitations, see the original model card. Search for Lythri/Lythri-7B-A4B-GGUF in the model browser. To use a specific quantization, append its type to the repository name, for example Lythri/Lythri-7B-A4B-GGUF:Q80. Lythri is built on Gemma 4 and is released under the Apache License 2.0.

Model

Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-Experimental-GGUF

ST

This repository contains GGUF quantizations of Hermes3.6-35B-A3B-Uncensored-Genesis-Final-Q8KP produced with a custom, measurement-driven requantizer developed as a personal study of quantization formats for MoE models. Unlike a standard one-format quantization, each file here uses a per-tensor blend of formats (Q4K, Q6K, Q80, NVFP4), selected by measuring the element-wise quantization error of every tensor individually and applying a sensitivity policy derived from controlled perplexity experiments. All files were requantized directly from the Q80 source checkpoint. No F16/BF16 intermediate was produced; source tensors are dequantized in memory only. (Now working on imatrix improvement and…

Model · Image and text to text

Qwopus3.8-27B-Flash-GGUF-1M

Solstice-AI

Solstice-AI/Qwopus3.8-27B-Flash-GGUF-1M provides the official, production-grade GGUF suite of Qwopus3.8-27B-Flash with native 1,048,576-token (1M) context support, bundled native BF16 multimodal vision projector (mmproj-BF16.gguf), and companion DSpark drafter. Evaluated under the standardized benchmark harness

Model · Text generation

granitelib-guardian-r1.0

IBM Granite

The Guardian Library includes four families of LoRA adapters, each developed for a specific task related to safety, factuality, and policy compliance in LLM-based systems. We provide adapters for: We give a brief overview of the functionality of each capability below; details can be found in each individual adapter readme. The four capabilities that have been implemented as LoRA adapters and made available in this HF repository are: Guardian Core: A LoRA adapter trained to judge whether the input prompts and output responses of an LLM-based system meet specified criteria, including safety risks (harm, jailbreaking, profanity, violence, sexual content, social bias, unethical behavior)…

Model · Image classification

Darwin-35B-ZTC-Vision

FINAL_Bench

Darwin-35B-ZTC-Vision is a vision decision model. It does not generate text. Given one or more images, a state and a typed question, it returns a calibrated probability for every option in one forward pass: yes or no (noul), one of several options (choice), or a level on an ordered scale (score). It handles up to four images per request, for example "which of the two photos has the higher quality" or "is there a defect on this part". It is a 35B mixture-of-experts model (about 3B active parameters per token) built on Darwin-35B-A3B-Opus with the Darwin ZTC decision readout. It is the vision member of the ZTC series; the text member is Darwin-27B-ZTC-v9. All 20 required vision benchmarks…

Model · Text classification

shingi-27b

Kortexa.ai

Shingi 27B is a local decision model. You give it context (text, and optionally images), a question and the possible answers. It returns a choice, calibrated probabilities or an ordinal score. It reads the logit of every candidate answer directly and generates no text. The model is a single 7.2 GB ternary GGUF file (PQ20). It is Bonsai 2 27B with were trained. It needs no adapter. On Linux with an NVIDIA GPU and the CUDA toolkit, or on a Mac with Apple Silicon: This builds the pinned Prism runtime, downloads this model into your Hugging Face cache and serves the API on http://127.0.0.1:8765. Use --port to change the port. The code, API examples and options are in kortexa-ai/shingi-27b.…

Model · Text generation

Jev-Style-Qwen3.5-2B-Decision-v2-GGUF

Chaoliang Yan

Website: jevstyle.com — all JevStyle decision models, benchmarks and quickstart in one place. A Jev-style decision model for classification, routing and typed choices. Give it a state, a question and a list of options; one prefill returns a selected option with calibrated probabilities. All three files include independently fitted calibration; use runtime temperature 1.0. Agreement is against CUDA merged BF16 on the same frozen 500-decision subset; accuracy is the task-macro average over its real-label examples. Full precision comparison. 81.20% macro accuracy on the fixed English reference panel, compared with 76.68% for v1 and 75.09% for English Laya. The results below use the CUDA…

static quants of https://huggingface.co/ConicCat/Gemma4-Garnet-31B For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants are available at https://huggingface.co/mradermacher/Gemma4-Garnet-31B-i1-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 See…

Model · Text generation

Jev-Style-Qwen3.5-2B-Decision-v2-MLX-bf16

Chaoliang Yan

Website: jevstyle.com — all JevStyle decision models, benchmarks and quickstart in one place. A Jev-style decision model for classification, routing and typed choices. Give it a state, a question and a list of options; one prefill returns a selected option with calibrated probabilities. Download this build: model.safetensors. The repository also includes its calibration, inference client and evaluation records. 81.20% macro accuracy on the fixed English reference panel, compared with 76.68% for v1 and 75.09% for English Laya. The results below use the CUDA reference structure: 11 real-label task groups, 3,277 decisions, equal task weights, and the same 3,100-record calibration split.…

Model · Image and text to text

Nex-N2.5-mini-APEX-I-MiniPlus-GGUF

IsValorum

Also, don't confuse APEX-I-MiniPlus (Standard) with a generic baseline APEX-I-Mini. Traditional APEX-I-Mini drops core experts aggressively to 2-bit IQ2S and leaves output.weight at 3-bit Q3KM, which creates a noticeable perplexity hit on complex reasoning tasks. Standard MiniPlus avoids that degradation floor while keeping boundary layers in linear Q3K for single-cycle vectorized AVX2 CPU dequantization (hitting 23 to 26+ tok/s on DDR4 laptops), while protecting output in Q6K and routers in F32. To put the numbers in perspective: this cuts nearly 2 GB off a flat 3-bit quant (approx. 15.6 GB), and weighs only about approx. 1 GB more than a generic APEX-I-Mini (approx. 12.5 GB). For that…

weighted/imatrix quants of https://huggingface.co/ElementMerc/tinyllama-1.1b-abliterated For a convenient overview and download list, visit our model page for this model. static quants are available at https://huggingface.co/mradermacher/tinyllama-1.1b-abliterated-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter…

Model · Video text to text

VideoGuard-Qwen3.5-9B-Safety-RL-Uncensored-GGUF

Prithiv Sakthi

C/C++ — https://github.com/ggml-org/llama.cpp

Model · Text generation

Qwen3.5-35B-A3B-Jackripper-v1.0-Uncensored-GGUF

Jari Katajisto

Virallinen Kataguru Jackripper 35B-A3B MoE v1.0 Uncensored GGUF -julkaisu. Malli yhdistää Qwen 3.5 35B-A3B -arkkitehtuurin (35B parametria, joista ~3B aktiivisena per token), suomen kielen syväkuratoinnin, sovereign agentic -päättelyn sekä Zero-Domain IPPCC SOMA/ARA -aktivointiortogonalisoinnin kieltäytymissuuntien poistamiseksi. Kaikki kvantit on suojattu suomalaisella imatrix-aktivaatiomatriisilla ja ne puhuvat todennetusti 100 % virheetöntä suomea. Tästä mallista tarjotaan tasan kolme (3) kuratoitua APEX-kvanttia eri VRAM-kokoluokkiin: 1. Lataa haluamasi APEX-kvantti (Jackripper-35B-A3B-v1.0-Uncensored-APEX-Medium.gguf suositeltu 16 Gt korteille, APEX-Small.gguf 12 Gt korteille). 2.…

static quants of https://huggingface.co/Jackrong/Qwopus3.5-4B-Coder For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants are available at https://huggingface.co/mradermacher/Qwopus3.5-4B-Coder-i1-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 See…

weighted/imatrix quants of https://huggingface.co/IFM/AmberSafe For a convenient overview and download list, visit our model page for this model. static quants are available at https://huggingface.co/mradermacher/AmberSafe-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 See…

Model · Text generation

Vinci-Prova-7B-1.0

SimpleDirect

An experimental post-training transfer study. We applied the Vinci SFT + DPO character recipe to mistralai/Mistral-7B-Instruct-v0.3 to answer one question: does character training developed on a different model lineage transfer to this one? Apache-2.0, 7.25B, drop-in with transformers. This uses a retired base, and we are saying so first. Mistral lists Mistral 7B Instruct v0.3 as retired as of 30 March 2025 (deprecated 30 November 2024), with Ministral 3 8B as the recommended replacement. "Retired" is Mistral's own lifecycle term. The open weights remain downloadable on Hugging Face under Apache-2.0. We selected this base for continuity with our earlier experiments, not because it is…

Model

Al-Khwarizmi-3B

Mohamed Zoelfakar

Fine-tune of HuggingFaceTB/SmolLM3-3B-Base, trained in three stages — full fine-tuning, LoRA, and a final DPO alignment pass — to solve grade-school math word problems. - Trained on the complete OpenAI GSM8K dataset (both main and socratic reasoning styles) across two LoRA passes, on top of an initial full fine-tune - 87.21% mean token accuracy on held-out validation data — up from 84.75% after the initial full fine-tune, and 82.4% at the very first checkpoint - Validation loss reduced by ~19% across the full pipeline (0.678 → 0.462), with training and validation loss tracking closely throughout every stage — no overfitting observed - Added a final DPO (Direct Preference Optimization) stage…

Model · Image and text to text

Qwen3.8-27B-TURBO-Fable-Cold-Fusion-NEO-CODER-mlx-oQ8e-1M

Solstice-AI

Solstice-AI/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NM-DAU-mlx-oQ8e-1M is the premier Apple Silicon mixed-precision serving release of DavidAU's flagship Qwen3.8-27B Cold Fusion foundation (DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NM-DAU). Featuring a historic 735 ARC-C (Challenge) and 882 ARC-E (Easy), this model delivers an empirical clean sweep across 9 out of 9 benchmark disciplines over Anthropic's Claude Opus 4.6 Max under the official Claude Code evaluation harness. Engineered with baked-in 1,048,576 Token (1 Million Token) YaRN RoPE scaling, calibrated via importance matrix optimization (oqimatrixreport.json), and accelerated…

Model · Text generation

XiYanSQL-QwenCoder-7B-2504

XGenerationLab

We are excited to release the XiYanSQL-QwenCoder-2504 version, our latest SQL generation model. This version continues to optimize upon the previous version, delivering enhanced performance. - Our model incorporates important explorations combining fine-tuning and GRPO training, leveraging the post-training strategies of GRPO without a thinking process, achieving both efficiency and accuracy in SQL generation. - It demonstrates impressive performance and supports multiple dialects, ready to use out of the box. - Improved generalization capabilities, excelling on different dialects and out-of-domain datasets. In this evaluation, we have also added a real-world SQL benchmark (the DW test…

Model · Text generation

gemma-4-E2B-meeting-agent-zh

Luigi

Gemma-4-E2B (QAT) fine-tuned for three on-device meeting tasks, on a phone CPU: 1. Realtime reading. The model reads a live zh-TW meeting transcript window by window and writes short cited notes: decisions, actions, proposals, open issues and key figures. The harness turns those notes into minutes. 2. Notes → prose (from v8 on). It turns the notes into a cited summary in prose. 3. Notes → title (from v8 on). It writes a title for the whole meeting from the notes. mobile-v2/ holds the E2B mobile-v2 reader. It continues mobile-v1 with two training stages: 1. SFT on reading rows plus dense-list conversion rows, including a third domain: Taiwanese financial TV programs; 2. one round of…

static quants of https://huggingface.co/ritiksuman/rudra For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants are available at https://huggingface.co/mradermacher/rudra-i1-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 See…

Model · Image and text to text

TerminalHorizon-35B-A3B

Zoushun

Scaling Agentic Data for Long-Horizon Terminal Intelligence TerminalHorizon is a fully automated data synthesis engine for long-horizon terminal agents. Grounded in real-world professional work, it constructs executable environments and generates training trajectories that connect agentic behavioral patterns across stages toward a shared goal. This repository releases TerminalHorizon-35B-A3B, obtained by supervised fine-tuning Qwen3.5-35B-A3B on The terminal is a shared interaction interface, not a restriction to software tasks. The training environments span 25 professional domains and 463 subdomains, including software engineering, machine learning, education, agriculture, healthcare…

Model · Text generation

Thanatos-27B-HERETIC

Fool

Architecture: Qwen 3.8 27B (Dense) | Parameters: 27B | Context: 262,144 native / 1,010,000 extensible | Base: Heretic ARA (heretic-org) | Teacher: Claude Fable 5 | Type: Distilled + Abliterated LLM A personal sibling to FoolDev/Janus-35B-HERETIC, which since its 0.9.0 is a Qwen 3.6 35B-A3B MoE — so the two are genuinely different models, not two abliterations of one base (see Janus vs Thanatos). This one is the dense Qwen/Qwen3.8-27B, sharing the teacher (Claude Fable 5) and dataset family, built on heretic-org/Qwen3.8-27B-heretic-ara, an uncensored Heretic ARA-method abliteration, with refusal-trained behavior dialed back at the base layer. (The sibling's -35B- slug names its model…

weighted/imatrix quants of https://huggingface.co/ElementMerc/qwen3-1.7b-abliterated For a convenient overview and download list, visit our model page for this model. static quants are available at https://huggingface.co/mradermacher/qwen3-1.7b-abliterated-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9…

static quants of https://huggingface.co/mfielding92/SmartCode-Fable-5-CoT-Reasoning-QVO-Qwen-3.6-27B-Distilled For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants are available at https://huggingface.co/mradermacher/SmartCode-Fable-5-CoT-Reasoning-QVO-Qwen-3.6-27B-Distilled-i1-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the…

Model · Text generation

Ternary-Bonsai-2-27B-onwx

Ryu Gyo

Prism ML の Ternary-Bonsai-2-27B（Qwen3.8-27B を三値の重みで学習し直したモデル）を、onwx 用に変換したものです。onwx は LLM を AMD Ryzen AI の NPU（XDNA2）だけで 動かすアプリで、このリポジトリにはモデルと、このモデル専用の NPU プログラムが入っています（アプリは別にインストールします）。 重みは三値（−1・0・+1）を 2 ビットのまま NPU に載せ、行列計算・DeltaNet・アテンション・LM ヘッドまで 1 トークンの計算をすべて NPU の 32 個の AI エンジンで行います。NPU 向けのプログラムは AI エンジンごとに手で書いてビルド済みなので、PC の上でコンパイルを待つことはありません。 （onwx 0.4.0、Ryzen AI Max+ 395（Windows）で NPU の電力モードを performance にして測定。Ryzen AI 300 シリーズ（HX 370 など）も NPU は同じ 32 個の AI エンジンで、同じ方式で動きます。標準の電力モードでは NPU のクロックが下がって 2 割ほど遅くなります。電力モードは管理者の PowerShell で C:\Windows\System32\AMD\xrt-smi.exe configure --pmode performance。Windows の Copilot+ の AI 機能が NPU を使っていると遅くなることがあります） 1. onwx（アプリ）を入れます。 Windows…

Model · Image and text to text

Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-NVFP4-1M

Solstice-AI

This model is pre-configured with native-esque 1,048,576 token (1M) YaRN scaling baked directly into config.json. Users do not need to supply command-line flags or rope overrides: vLLM and SGLang automatically initialize 1M rotary positional frequencies on load.

DeepNarra AI Ecosystem — Sovereign Frontier Architecture Apex Tri-Hybrid Base Foundation Model DeepNarra Hiraya-1-Base is an official release in the sovereign DeepNarra ecosystem. 1. Two-Tier Hierarchical MoE (H-MoE): 92 Fine-Grained Sub-Experts per layer with active Shared Foundation Expert active on every token for baseline coherence. 2. Coconut Continuous Latent Deliberation: Continuous thought phase space ($\mathbb{R}^{768}$) enabling internal chain-of-thought progression prior to verbal token emissions. 3. Symplectic Störmer-Verlet Physics: Hamiltonian energy-conserving dynamics for stable long-horizon reasoning trajectories. 4. Hardware-Efficient Recurrence: Chunk-wise Gated Linear…

Model · Image and text to text

Qwen3.8-27B-TTURBO-Cold-Fusion-709-L-Uncensored-mlx-oQ8e-1M

Solstice-AI

Solstice-AI/Qwen3.8-27B-TTURBO-Cold-Fusion-709-L-Uncensored-mlx-oQ8e-1M is the official Apple Silicon MLX release of DavidAU's Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored. Engineered specifically for Apple Silicon Unified Memory architectures (M2, M3, M4, M5 Max / Ultra), this release couples oQ8e imatrix-guided mixed-precision with Solstice's proprietary 10-Level Cognitive Variable Thinking System and native Hardware Multi-Token Prediction (MTP). This checkpoint embeds the Solstice 10-Level Cognitive Jinja chat template, allowing instant dynamic control over reasoning depth

Model · Text generation

llm-jp-4.1-8b-onwx

Ryu Gyo

国立情報学研究所 大規模言語モデル研究開発センターの llm-jp-4.1-8b-thinking を、onwx 用に変換したものです。onwx は LLM を AMD Ryzen AI の NPU（XDNA2）だけで 動かすアプリで、このリポジトリにはモデルと、このモデル専用の NPU プログラムが入っています（アプリは別にインストールします）。 重みは int4（128 個ごとのスケールと最小値つき）で NPU に載せ、行列計算・アテンション・LM ヘッドまで 1 トークンの計算をすべて NPU の 32 個の AI エンジンで行います。NPU 向けのプログラムはビルド済みなので、PC の上でコンパイルを待つことはありません。 （onwx 0.4.0、Ryzen AI Max+ 395（Windows）で NPU の電力モードを performance にして測定。Ryzen AI 300 シリーズ（HX 370 など）も NPU は同じ 32 個の AI エンジンで、同じ方式で動きます。標準の電力モードでは NPU のクロックが下がって 2 割ほど遅くなります。電力モードは管理者の PowerShell で C:\Windows\System32\AMD\xrt-smi.exe configure --pmode performance。Windows の Copilot+ の AI 機能が NPU を使っていると遅くなることがあります） 1. onwx（アプリ、0.2.2 以降）を入れます。 Windows はスタートメニューで「PowerShell」を開き、次の 1…

static quants of https://huggingface.co/MuXodious/Qwen3.5-9B-SOMPOA-heresy-MTP For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants are available at https://huggingface.co/mradermacher/Qwen3.5-9B-SOMPOA-heresy-MTP-i1-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter…

static quants of https://huggingface.co/jylin001206/SII-Decision-4B-Full For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality…

Model · Text generation

Qwen3.5-35B-A3B-Sceptic-v3.4-GGUF

Jari Katajisto

Virallinen Kataguru-lippulaiva ja uusi perusta suomenkieliselle MoE-päättelylle. Malli yhdistää Qwen 3.5 35B-A3B -arkkitehtuurin (35B parametria, joista ~3B aktiivisena per token), suomen kielen syväkoulutuksen, Ornith STEM/matematiikkavektorit, JEV Truth -totuudenmukaisuuden sekä Diabolical Code Immunity v2/v3 -järjestelmäkoodauskyvyn. Kaikki kvantit on suojattu suomalaisella scepticv3.4.imatrix-aktivaatiomatriisilla ja ne puhuvat todennetusti 100 % virheetöntä suomea. Tästä mallista tarjotaan tasan kolme (3) kuratoitua APEX-kvanttia eri VRAM-kokoluokkiin: Mittaukset on ajettu virallisella tuotantolaitteistolla (2x RTX 5090 Blackwell TP=2, vLLM / llama.cpp b11064): 1. Lataa haluamasi…

static quants of https://huggingface.co/m-a-p/OProver-32B For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here…

static quants of https://huggingface.co/llmfan46/Qwen3-VL-32B-Instruct-uncensored-heretic For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants are available at https://huggingface.co/mradermacher/Qwen3-VL-32B-Instruct-uncensored-heretic-i1-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter…

The MOSS-Image SFT model often draws faint, evenly spaced vertical color stripes across smooth areas like skies, fog, backdrops and walls. This repo has two banding-fix LoRAs that remove them while keeping the image close to what SFT would have made, plus preset workflows that pair the fix with community speed LoRAs for 10–14 step (or 4–8 step) generation. Just want one file that works? MOSS-Image-SFT-DeBand-int8 has the fix (and the Fast preset) merged into ready-to-run 5.4 GB int8 models. No LoRAs or custom nodes. 1. Download the main fix, mossimagesftbandingfixL37-41bf16.safetensors, into ComfyUI/models/loras/. 2. In your MOSS-Image workflow, add LoraLoaderModelOnly after the SFT model…

static quants of https://huggingface.co/DavidAU/gemma-3-12b-it-vl-HighIQ-Polaris-Heretic-Uncensored-Thinking For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants are available at https://huggingface.co/mradermacher/gemma-3-12b-it-vl-HighIQ-Polaris-Heretic-Uncensored-Thinking-i1-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the…

This repository distributes final consumer artifacts with immutable paths and SHA-256 identities. Source checkpoints for the initial G1i cohort come from BlinkDL/rwkv7-g1 at revision 159cf82ed0f18eb9dcd92c388688bec11024895e. Generation, quantization, context length, converter or SDK revision, platform, accelerator, SoC, ABI, architecture, and batch mode are identity metadata, not directory levels. They remain in the immutable manifest and, when needed to distinguish bytes, in the globally unique basename. Use index.json for discovery and the referenced immutable manifest for sizes, digests, source lineage, recipes, and consumer-acceptance status. Binary paths and cohort manifests are…

static quants of https://huggingface.co/MirilAI/Miril-DroneVLM-2B-2 For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant…

static quants of https://huggingface.co/Gille/StrangeMerges35-7B-slerp For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality…

static quants of https://huggingface.co/CCSSNE/tvall43-Qwen3.5-4B-heretic-v2 For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants are available at https://huggingface.co/mradermacher/tvall43-Qwen3.5-4B-heretic-v2-i1-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter…

static quants of https://huggingface.co/MaYiding/EventTwin For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here…

Model · Text generation

ViuAI-500M

Vipin Kumar

ViuAI Sarus-500M is a 495-million parameter autoregressive Small Language Model (SLM) designed, trained from scratch, and aligned by ViuAI Research. Engineered for ultra-low latency edge devices and lightweight CPU/GPU deployment, Sarus-500M balances deep knowledge representation with minimal resource requirements (~944 MB model footprint). The official recommended production checkpoint is: You can run Sarus-500M locally in a few lines of code without requiring the heavy transformers package: Because Sarus-500M is a 500M SLM, setting proper sampling parameters prevents logit tail collisions: The dpo5kfinal.pt checkpoint was evaluated on the standard multi-domain benchmark: Sarus-500M is…

Model · Text generation

Qwen3.8-27B-EfficientThink-Uncensored-VAL-APEX-I-MiniPlus-V2.1-GGUF

IsValorum

Official VAL-APEX-I quantization of nerkyor/Qwen3.8-27B-EfficientThink-Uncensored-K3-Opus5-Grok4.6-GPT5.6Sol-SFT-SimPO-DFlash2. Vector-calibrated Asymmetric Layer-wise Outlier-preserving Recurrent-aware Unified Matrix-quantization The NanoPlus V3 file is 11.35 GB, approximately 0.45 GB smaller than GSQ-RCO IQ3S, with 11.49% lower reported WikiText-2 PPL (6.2577 vs 7.07). MiniPlus V3 reaches 13.21% lower reported PPL (6.1361 vs 7.07) at 14.16 GB, approximately 2.36 GB larger. × is the number of tensors in each format. Counts were read from the released GGUF headers; the GSQ-RCO column uses its published tensor allocation. This release contains 851 tensors. The main GGUF has no blk.64.…

Model · Text generation

Q-TensorFormer

V C Premchand Yadav

Q-TensorFormer is an open-source, publication-grade research architecture and inference testbed designed by Premchand Yadav. It investigates a fundamental question in efficient deep learning: Can a Transformer dynamically allocate its computational capacity, tensor rank, attention complexity, and KV-cache precision according to the marginal informational value of each token under explicit hardware constraints? Q-TensorFormer is an experimental PyTorch and Hugging Face research architecture that investigates closed-loop, information-value-driven resource allocation for Transformer inference under explicit memory, latency, bandwidth, and quality constraints. Rather than claiming that dynamic…

Model · Text generation

Qwen3.8-27B-EfficientThink-Uncensored-VAL-APEX-I-NanoPlus-GGUF

IsValorum

Official VAL-APEX-I quantization of nerkyor/Qwen3.8-27B-EfficientThink-Uncensored-K3-Opus5-Grok4.6-GPT5.6Sol-SFT-SimPO-DFlash2. Vector-calibrated Asymmetric Layer-wise Outlier-preserving Recurrent-aware Unified Matrix-quantization The NanoPlus V3 file is 11.35 GB, approximately 0.45 GB smaller than GSQ-RCO IQ3S, with 11.49% lower reported WikiText-2 PPL (6.2577 vs 7.07). MiniPlus V3 reaches 13.21% lower reported PPL (6.1361 vs 7.07) at 14.16 GB, approximately 2.36 GB larger. × is the number of tensors in each format. Counts were read from the released GGUF headers; the GSQ-RCO column uses its published tensor allocation. This release contains 851 tensors. The main GGUF has no blk.64.…

Model · Text generation

DeepSeek-R1-Distill-Qwen-32B-abliterated-GGUF

Alan Frederick Benitez

Unlike incomplete GGUF uploads, this repository provides the full 11-quantization spectrum (from high-precision F16 down to lightweight Q2K) of the abliterated DeepSeek-R1-Distill-Qwen-32B model. Choose the exact fit for your VRAM/RAM constraints without sacrificing reasoning capabilities. This is the integral and complete collection of quantizations in GGUF format for the DeepSeek-R1-Distill-Qwen-32B-abliterated model, prepared locally for use with llama.cpp, Ollama, LM Studio, or Text-Generation-WebUI. This model combines advanced reasoning distillation (deep logical thinking capability via tags) based on Alibaba's powerful Qwen-32B model, but processed with abliteration techniques to…