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
输入 待核验→ 输出 待核验
价格待核验多模态与渠道费用须分别核对
查看详情 ↗DISCOVER & COMPARE
探索 AI 模型 84991
Qwen3.8-27B-Huihui-Uncensored-llamafile
Bojun-Feng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
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗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
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗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
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗gpt2-persian
bolbolzaban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
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗Abliterated-MiMo-V2.6-Distill-Qwen-9B-GGUF
BoldingBuildsThis is an experimental, refusal-ablated GGUF derivative of XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B. It deliberately reduces refusal on harmful prompts. Its responses require independent review and appropriate controls in any downstream application. **Evaluation status:** The judged safety evaluation is still running. The tables below report completed Q8_0 runs only; judged results will be added when complete. This is an edited derivative that was then quantized, not a plain quantization of the original weights. Text generation is the only evaluated modality; no vision capability is claimed for these GGUF files.
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗AREX-2-Abliterated-GGUF
BoldingBuilds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。
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗GLM-5.3-Flash-Uncensored-GGUF
BoldingBuildsRefusal removed from GLM-5.3-Flash by editing **666 of its 12,384 routed experts** — about **1.74% of the model's parameters**. Quantized to IQ1_S at **71.5 GiB**, which fits on a single 96 GB card. I did the edit and the quantization myself, and I measured what it cost. Every number below is first-party, taken on my own rig, on the file in this repo. > **Read this before you run it.** Every headline number below was measured at > **temperature 0, seed 0, repetition penalty 1.1/256, 16k context, `--reasoning-budget 1024`.** > The file itself ships `temperature 1.0` in its metadata and no reasoning budget, so a plain > `llama-server -m …` does **not** run the measured configuration. I tested that gap directly > (table under *Default settings vs tested settings*): at the file's own defaults, long coding > tasks ran out of output budget without answering about **1 time in 8**; with > `--reasoning-budget 1024` they did not. **For code and chat, use the quickstart line as > written. For hard math, leave the budget off** (see the correction below). This is an experimental 1.92-bit build: 37 of > 120 harmful-set responses and 8 of 120 benign ones hit the output cap in the headline run, and > that is reported separately from "unusable" below.
译文排队中 · 当前显示来源原文
输入 文本 · 图像→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗Muse-Spark-Open-816B-A42B-Abliterated-UD-IQ1_S-GGUF
BoldingBuilds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
输入 文本 · 图像→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗An uncensored edit of **NVIDIA Nemotron-3.5-Bolt-74B-A9B**, a hybrid Mamba/MoE model, produced by editing NVIDIA's **shipped 24GB 2-bit GGUF** (nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB-GGUF) **directly** — no BF16 requantize at any point. That matters for this model specifically. Bolt is a LatentMoE: every routed expert sums through one dense `ffn_latent_up` per layer, and the experts themselves are ~17GB of IQ2_XXS weights produced by NVIDIA's quantization-aware distillation. A conventional abliterate-then-requantize pipeline would throw that QAD work away. Instead the projection is applied in place and codes are re-picked under the **original block scales**, so at λ=0 the output file is byte-identical to the input by construction — which is the self-test the build gates on (verified: all edited tensors byte-identical at λ=0). The published file is the same length as NVIDIA's, 22,018,326,528 bytes. Bolt is multimodal, so the projector ships alongside. The edit touched only text-model tensors, so the projector is byte-for-byte NVIDIA's file; download both or vision silently does nothing.
译文排队中 · 当前显示来源原文
输入 文本 · 图像→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗NVIDIA-Nemotron-3.5-Lightning-30B-A3B-ShimQuant-GGUF
BoldingBuilds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
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗orcarouter_GLM-5.3-Flash-Uncensored-GGUF
BoldingBuildsGGUF quantizations of **orcarouter/GLM-5.3-Flash-Uncensored-FP8**, built with a custom importance matrix and measured rather than assumed. > ⚠️ **Uncensored model.** Safety refusals were removed by OrcaRouter. It will comply with > harmful, unethical, and illegal requests. Released for research — red-teaming, refusal- > mechanism study, robustness evaluation. You are responsible for what you do with it. > 🔧 **What happened, and where it stands (2026-09-06).** On 2026-09-05 a user reported > looping and incoherent output from `IQ2_S` (discussion #1). They were > right. The original build pinned only the `blk.45` MTP block at high precision; attention, > the **shared** experts, the three dense FFN blocks, and the embeddings were all quantized > down with the routed experts. The shared experts and attention run on every token. > > **`IQ2_S` was rebuilt on 2026-09-06** with those tensors held at `q8_0` (+7 GB) and tested > against the original on a rented H200: 12 tasks × 2 seeds × 2 samplers, same scorer, same > settings. Full table in **Coherence gate** below. Short version: the original failed 3 of 8 > coding tasks by reasoning to the output cap without ever answering; the rebuild passed > all 48, matching `IQ3_XXS`, which also passed all 48 on its original recipe. > **`IQ2_XXS` is being rebuilt the same way** and stays flagged until it passes the same gate. > > **Two corrections to earlier versions of this page.** The imatrix has **10 chunks, not 60** > (the run checkpointed at 10 and stopped; numerically sound, one uncovered expert row in > `blk.29`). And the repetition-penalty workaround I posted **did not change the outcome** on > this failure mode — 21/24 → 21/24 for the original file with and without it. It helps > n-gram repetition; it does not help a model that cannot emit the literal it intends. > The rebuilt files carry `repeat_penalty 1.1` in their metadata anyway, because it is cheap. > > Only `IQ3_XXS` had been benchmarked before publication, and I shipped all three off that. > That was the actual mistake. The gate below is the fix for the process, not just the file.
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗Qwen3.8-27B-Abliterated-ThinkFix-GGUF
BoldingBuilds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。
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗Qwen3.8-27B-Reasoning-Termination-Fix-GGUF
BoldingBuilds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
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗Ternary-Bonsai-2-27B-Abliterated-PQ2_0-MTP-GGUF
BoldingBuilds> **v2 is out:** Ternary-Bonsai-2-27B-Abliterated-v2-PQ2_0-MTP-GGUF. > It fixes the thinking-mode problem described below: blank answers on harmful prompts drop > from 27% to 15% at default settings and to 0% with `reasoning_effort=medium`. This repo > stays up unchanged as v1. Refusal behaviour removed from `prism-ml/Ternary-Bonsai-2-27B-gguf` at PQ2_0 (2.13 bpw), by flipping a small number of ternary digits in place. **Block scales are never touched and the model is never requantised**, so this is the released ternary packing with a few digits changed — not a re-derived one. A second file adds the Qwen3.8-27B multi-token-prediction head, which makes `--spec-type draft-mtp` work and decodes **~37% faster** at identical weights.
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗Ternary-Bonsai-2-27B-Abliterated-PTQ1_0-GGUF
BoldingBuilds> **v2 is out:** v2, PTQ1_0 > fixes the thinking-mode problem described below and is the one to use with thinking on. > This v1 file stays up unchanged. Refusal-ablated **Ternary Bonsai 2 27B**, edited directly on the ternary lattice and shipped in PrismML's native PTQ1_0 pack. - **5,946,648,928 bytes — byte-for-byte the size of the official PTQ1_0 release** - No requantization, no BF16 bake, no runtime steering, no control vector at serve time - Requires the PrismML llama.cpp fork
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗Ternary-Bonsai-2-27B-Abliterated-v2-PQ2_0-MTP-GGUF
BoldingBuilds**v2 is built to be run with thinking on.** Abliterated Bonsai 2 builds, my own v1 included, tend to write the whole answer inside their reasoning and run out of budget before handing it over. v2 changes one row of the output layer so the model closes its reasoning once the answer is ready. With `reasoning_effort` set to medium it answered every prompt on the main test; at default settings it halves v1's blank answers. **Where v2 stands out: hard requests, given room to think.** StrongReject (150 detailed harmful requests), thinking on, default settings, 16,384-token budget, same judge for every row: Much of that gap is answers the other builds never deliver: they write the answer inside their reasoning and run out of room before closing it.
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗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
输入 文本 · 图像→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗bert-emotion
boltuix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。
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗bert-lite
boltuix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。
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗bert-micro
boltuix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
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗bert-mini
boltuix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
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗NeuroBERT-NER
boltuixThe `boltuix/NeuroBERT-NER` model is a fine-tuned transformer for **Named Entity Recognition (NER)**, built on the `boltuix/NeuroBERT-Mini` base model. It excels at identifying 36 entity types (e.g., people, places, organizations, dates, money) in English text, making it ideal for applications like information extraction, chatbots, and knowledge graph construction. - **Dataset**: boltuix/conll2025-ner (143,709 entries, 6.38 MB) - **Entity Types**: 36 NER tags (18 entity categories with B-/I- tags + O) - **Training Examples**: ~115,812 | **Validation**: ~15,680 | **Test**: ~12,217 *Note*: Split sizes are approximate and don’t sum to 143,709; verify with dataset analysis. - **Domains**: News, user-generated content, research corpora - **Tasks**: Sentence-level and document-level NER - **Version**: v1.1 > **Note**: The dataset link is a placeholder. Replace with the correct Hugging Face repository URL once available.
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
价格待核验多模态与渠道费用须分别核对
查看详情 ↗boltzgen-1
boltzgenBoltzGen-1 is a generative model for designing proteins and peptides that bind arbitrary biomolecules.
译文排队中 · 当前显示来源原文
输入 待核验→ 输出 待核验
价格待核验多模态与渠道费用须分别核对
查看详情 ↗暂未找到匹配模型
可以调整厂商、分类或输入输出条件；新模型正持续收录。
COST EXPLORER
好模型，也要算好成本。
调整月度 Token 用量，直观比较不同模型的预估 API 开支。
1 M = 100 万 Token · 美元计价
仅比较同一渠道的基础文本 Token 报价
仅比较同一渠道的基础文本 Token 报价
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。
MODEL HANDBOOK
读懂模型，再做选择
API 价格应该怎么比较？+
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。
开放权重，就代表可以免费商用吗？+
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。
排行榜第一，就是最适合我的模型吗？+
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。