从语言到影像，从开源到 API。读懂模型能力、比较真实成本，
为你的下一个想法，找到合适的智能引擎。
> **Join the Discord** for updates, roadmaps, projects, or just to chat. No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals. Stronger uncensoring — model is fully unlocked and won't refuse prompts. Disclaimers that were present in previous releases have been significantly reduced in this version.
译文排队中 · 当前显示来源原文
> **Join the Discord** for updates, roadmaps, projects, or just to chat. No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals. Stronger uncensoring with more thorough refusal removal. If this variant is too loose for your use case, a Balanced variant may follow.
译文排队中 · 当前显示来源原文
> **Join the Discord** for updates, roadmaps, projects, or just to chat. No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals. Stronger uncensoring with more thorough refusal removal. If this variant is too loose for your use case, a Balanced variant may follow.
译文排队中 · 当前显示来源原文
> **Join the Discord** for updates, roadmaps, projects, or just to chat. No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals. Stronger uncensoring — model is fully unlocked and won't refuse prompts. May occasionally append short disclaimers (baked into base model training, not refusals) but full content is always generated.
译文排队中 · 当前显示来源原文
> **Join the Discord** for updates, roadmaps, projects, or just to chat. No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals. Stronger uncensoring with more thorough refusal removal. If this variant is too loose for your use case, a Balanced variant may follow.
译文排队中 · 当前显示来源原文
> **Join the Discord** for updates, roadmaps, projects, or just to chat. No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals. Stronger uncensoring with more thorough refusal removal. If this variant is too loose for your use case, a Balanced variant may follow.
译文排队中 · 当前显示来源原文
> **Join the Discord** for updates, roadmaps, projects, or just to chat. > **Not sure which variant to pick?** 99.9%+ of users should use **Balanced** — same 0/465 refusal rate, more stable sampling, great for agentic coding / tool-use / reasoning / creative writing. Pick **Aggressive** only if you specifically want the model to skip its preamble on hardcore prompts. > **HuggingFace's "Hardware Compatibility" widget doesn't recognize K_P quants** — it may show fewer files than actually exist. Click **"View +X variants"** or go to **Files and versions** to see all available downloads.
译文排队中 · 当前显示来源原文
> **Join the Discord** for updates, roadmaps, projects, or just to chat. > **HuggingFace's "Hardware Compatibility" widget doesn't recognize K_P quants** — it may show fewer files than actually exist. Click **"View +X variants"** or go to **Files and versions** to see all available downloads. No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended — just without the refusals.
译文排队中 · 当前显示来源原文
> **Join the Discord** for updates, roadmaps, projects, or just to chat. > **HuggingFace's "Hardware Compatibility" widget doesn't recognize K_P quants** — it may show fewer files than actually exist. Click **"View +X variants"** or go to **Files and versions** to see all available downloads. No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals.
译文排队中 · 当前显示来源原文
> **HauhauCS FastMTP: up to 3.02x document TG and 1.93x reasoning TG versus non-MTP — plus up to 35.2% more document TG and 21.1% more reasoning TG than standard embedded MTP.** > **Join the Discord** for updates, roadmaps, projects, or just to chat. This is the **Aggressive variant**: direct answers, no refusal behavior, and minimal preamble on hard prompts.
译文排队中 · 当前显示来源原文
> **Join the Discord** for updates, roadmaps, projects, or just to chat. No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals. This is the **Aggressive** variant with stronger uncensoring. Use this when the Balanced variant refuses too much.
译文排队中 · 当前显示来源原文
No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals. This is the **Balanced** variant with moderate uncensoring. Best for agentic coding and reliability-critical tasks.
译文排队中 · 当前显示来源原文
Quantized GGUF weights of DreamShaper 8 LCM, optimized for fast local CPU and low-memory GPU image generation using `stable-diffusion.cpp`. These models are designed specifically to run inside the **Haven AI Companion** local portrait generation engine.
译文排队中 · 当前显示来源原文
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
> [!IMPORTANT] > ### MTP Layers Grafted on Unsloth UD XL Quantizations > > The GGUF files in this repo are **Unsloth Dynamic 2.0 XL (UD XL) quantizations** of Qwen3.6-35B-A3B > with **Multi-Token Prediction (MTP) layers grafted on top**. > > - **Base quantization:** Unsloth UD XL — see > Unsloth Dynamic 2.0 GGUFs for benchmarks. > - **MTP layers:** Taken from > am17an/Qwen3.6-35BA3B-MTP-GGUF > and stored in **Q8_0** precision, then merged into the UD XL GGUFs. > - **Why Q8 for MTP?** The draft heads are small relative to the base model, so Q8_0 keeps them > near-lossless while avoiding the overhead of full-requantizing the entire stack. > - **`convert.py`:** The script used to graft MTP layers into the UD XL GGUFs. Adapted from > this gist. > - **`35BA3B-MTP.gguf`:** The raw Q8_0 MTP layer source file (extracted only) used as input to `convert.py`. > > To run these files with MTP you need a custom build of **llama.cpp** that includes the MTP/speculative > decoding support from PR #22673. Follow the steps below. > > --- > > ### 🐳 Or Use Pre-built Docker Images (Recommended) > > Don't want to build from source? Ready-to-use Docker images with MTP support are available for multiple backends: > > > | Image | Backend | > |---|---| > | `havenoammo/llama:cuda13-server` | CUDA 13 | > | `havenoammo/llama:cuda12-server` | CUDA 12 | > | `havenoammo/llama:vulkan-server` | Vulkan | > | `havenoammo/llama:intel-server` | Intel (CPU/Xe) | > | `havenoammo/llama:rocm-server` | ROCm (AMD) | > > Quick start — running this model via Docker: > > The two flags that matter most for MTP are `--spec-type mtp` and `--spec-draft-n-max 3`. > See the Reddit post for more details and discussion. > > --- > > ### Building llama.cpp with MTP Support — Step by Step > > #### 1. Clone and enter the repo > > > #### 2. Fetch the latest remote changes > > This pulls all new refs from the upstream repository so you're working off the current tip of master. > > #### 3. Fetch PR #22673 as a local branch > > PR #22673 ("llama + spec: MTP Support") adds the speculative-decoding / MTP infrastructure that > lets `llama-server` consume multi-token prediction heads. We pull it directly without waiting for > i
译文排队中 · 当前显示来源原文
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
This is an uncensored version of Qwen/Qwen3.8-27B created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens. The newly added **Huihui-Qwen3.8-27B-abliterated-Underdog-Saluki** series come from ConwayResearch/Underdog-Saluki-27B-1.0 and ukisai/Swift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF. Only layers **22 to 52 (0-based indexing)** have been ablated, while the other layers remain unablated. It may come with a small disclaimer warning. Due to ConwayResearch/Underdog-Saluki-27B-1.0 not providing an imatrix file, we will attempt to perform ablation by mixing the GGUF weights of Swift-1.5 and ConwayResearch/Underdog-Saluki-27B-1.0.
译文排队中 · 当前显示来源原文
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
可以调整厂商、分类或输入输出条件；新模型正持续收录。
调整月度 Token 用量，直观比较不同模型的预估 API 开支。
使用已采集且单位可比的 USD / 百万 Token 渠道报价。报价为采集时快照；不计缓存优惠、阶梯价、推理额外输出与税费，请以来源的当前报价为准。
先区分输入、输出和缓存价格，再使用同一任务和用量估算总费用。图像、音频和视频可能按张、时长或分辨率计费，不能直接与 Token 单价比较。
不一定。必须逐一核对具体模型版本的许可证、商用限制及使用政策。自己部署还需要考虑显存、推理资源和运维费用。
榜单取决于测试任务、版本与评测方法。建议用实际业务样本比较准确性、时延、稳定性和成本，再选择模型。
