找到适合你的 AI 模型。

加入对比来源 / 文档 ↗

This is a Japanese DeBERTa V2 base model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR. The input text should be segmented into words by Juman++ in advance. Juman++ 2.0.0-rc3 was used for pre-training. Each word is tokenized into subwords by sentencepiece. Note that we filtered out documents annotated with "header", "footer", or "noisy" tags in OSCAR. Also note that Japanese Wikipedia was duplicated 10 times to make the total size of the corpus comparable to that of CC-100 and OSCAR. As a result, the total size of the training data is 171GB.

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

加入对比来源 / 文档 ↗

deberta-v2-base-japanese-char-wwm

ku-nlp

This is a Japanese DeBERTa V2 base model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR. This model is trained with character-level tokenization and whole word masking. There is no need to tokenize texts in advance, and you can give raw texts to the tokenizer. The texts are tokenized into character-level tokens by sentencepiece. - Japanese Wikipedia (as of 20221020, 3.2GB, 27M sentences, 1.3M documents) - Japanese portion of CC-100 (85GB, 619M sentences, 66M documents) - Japanese portion of OSCAR (54GB, 326M sentences, 25M documents)

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

加入对比来源 / 文档 ↗

deberta-v2-large-japanese

ku-nlp

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

加入对比来源 / 文档 ↗

deberta-v2-large-japanese-char-wwm

ku-nlp

This is a Japanese DeBERTa V2 large model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR. This model is trained with character-level tokenization and whole word masking. There is no need to tokenize texts in advance, and you can give raw texts to the tokenizer. The texts are tokenized into character-level tokens by sentencepiece. - Japanese Wikipedia (as of 20221020, 3.2GB, 27M sentences, 1.3M documents) - Japanese portion of CC-100 (85GB, 619M sentences, 66M documents) - Japanese portion of OSCAR (54GB, 326M sentences, 25M documents)

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

加入对比来源 / 文档 ↗

deberta-v2-tiny-japanese

ku-nlp

This is a Japanese DeBERTa V2 tiny model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR. The input text should be segmented into words by Juman++ in advance. Juman++ 2.0.0-rc3 was used for pre-training. Each word is tokenized into subwords by sentencepiece. Note that we filtered out documents annotated with "header", "footer", or "noisy" tags in OSCAR. Also note that Japanese Wikipedia was duplicated 10 times to make the total size of the corpus comparable to that of CC-100 and OSCAR. As a result, the total size of the training data is 171GB.

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

加入对比来源 / 文档 ↗

deberta-v2-tiny-japanese-char-wwm

ku-nlp

This is a Japanese DeBERTa V2 tiny model pre-trained on Japanese Wikipedia, the Japanese portion of CC-100, and the Japanese portion of OSCAR. This model is trained with character-level tokenization and whole word masking. There is no need to tokenize texts in advance, and you can give raw texts to the tokenizer. The texts are tokenized into character-level tokens by sentencepiece. - Japanese Wikipedia (as of 20221020, 3.2GB, 27M sentences, 1.3M documents) - Japanese portion of CC-100 (85GB, 619M sentences, 66M documents) - Japanese portion of OSCAR (54GB, 326M sentences, 25M documents)

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

加入对比来源 / 文档 ↗

deberta-v3-base-japanese

ku-nlp

The tokenizer of this model is based on huggingface/tokenizers Unigram byte-fallback model. The vocabulary entries were converted from `llm-jp-tokenizer v2.2 (100k)`. Please refer to README.md of `llm-jp/llm-ja-tokenizer` for details on the vocabulary construction procedure. Note that, unlike ku-nlp/deberta-v2-base-japanese, pre-segmentation by a morphological analyzer (e.g., Juman++) is no longer required for this model. We shuffled the corpora, which has 270B tokens in total, and trained the model for 2 epochs. Thus, the total number of tokens fed to the model was 540B.

译文排队中 · 当前显示来源原文

分类待补充

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

加入对比来源 / 文档 ↗

gpt2-large-japanese-char

ku-nlp

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

加入对比来源 / 文档 ↗

gpt2-medium-japanese-char

ku-nlp

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

加入对比来源 / 文档 ↗

gpt2-small-japanese-char

ku-nlp

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

加入对比来源 / 文档 ↗

kudzueye

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。

图像生成

输入文本→ 输出图像

价格待核验多模态与渠道费用须分别核对

加入对比来源 / 文档 ↗

Marco-Mini-Instruct-REAP-GGUF

kueizen

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

加入对比来源 / 文档 ↗

Marco-Nano-Instruct-REAP-GGUF

kueizen

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

加入对比来源 / 文档 ↗

MiniCPM-V-4.7-35B-A3B-MLX-4bit-TensorFold

kuepker

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

视觉问答

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

加入对比来源 / 文档 ↗

kugelaudio-0-open

kugelaudio

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。

语音合成

输入文本→ 输出音频

价格待核验多模态与渠道费用须分别核对

加入对比来源 / 文档 ↗

Underdog-Saluki-27B-1.0-MTP-Abliterated-GGUF

Kujira

Same as the MTP release: stock llama.cpp with MTP drafting (`--spec-type draft-mtp`). - Recommended sampling with thinking on: `--temp 1.0 --top-p 0.95 --top-k 20 --presence-penalty 1.1` (Qwen3.8's own thinking-mode values plus a light presence penalty). At temperature 0.6 with presence penalty 1.5, long agent sessions sometimes got stuck repeating whole paragraphs in the thinking; see "Sampling" below for the comparison. - 12 GB cards: 40K context fits (11.4 GB with a 22K-token prompt). 64K starts at 11.85 GB and grows about 0.45 GB with long prompts, spills out of VRAM and slows down sharply. RTX 3080 12 GB, WSL2, llama.cpp (PrismML fork `prism-b10754`). Refusal = the reply contains a refusal phrase. Harmful prompts: the 104-prompt test split of a public harmful-behaviors set.

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

加入对比来源 / 文档 ↗

Underdog-Saluki-27B-1.0-MTP-GGUF

Kujira

Nothing was trained or re-quantized. Saluki is built from Qwen3.8-27B, so the head is the one that belongs to its base model. Works with stock llama.cpp (tested on upstream `03aa006`, 2026-10-08, just after release `b11493`) and any build that has MTP drafting (`--spec-type draft-mtp`). Sampling and thinking settings are the same as for Saluki (see its card). Without the `--spec-type` flags the file runs like the original Saluki: llama.cpp logs the MTP tensors as unused, skips them, and uses the same VRAM as the original 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