Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features: **Qwen3-4B** has the following features: - Type: Causal Language Models - Training Stage: Pretraining & Post-training - Number of Parameters: 4.0B - Number of Paramaters (Non-Embedding): 3.6B - Number of Layers: 36 - Number of Attention Heads (GQA): 32 for Q and 8 for KV - Context Length: 32,768 natively and 131,072 tokens with YaRN. For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our blog, GitHub, and Documentation.
翻訳待ち・原文を表示
分類準備中
入力 未確認→ 出力 未確認
See our collection for all versions of Qwen3 including GGUF, 4-bit & 16-bit formats. Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants. - Fine-tune Qwen3 (14B) for free using our Google Colab notebook here! - Read our Blog about Qwen3 support: unsloth.ai/blog/qwen3 - View the rest of our notebooks in our docs here. - Run & export your fine-tuned model to Ollama, llama.cpp or HF.
翻訳待ち・原文を表示
入力 テキスト→ 出力 テキスト
Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants. We introduce the updated version of the **Qwen3-4B non-thinking mode**, named **Qwen3-4B-Instruct-2507**, featuring the following key enhancements: - **Significant improvements** in general capabilities, including **instruction following, logical reasoning, text comprehension, mathematics, science, coding and tool usage**. - **Substantial gains** in long-tail knowledge coverage across **multiple languages**. - **Markedly better alignment** with user preferences in **subjective and open-ended tasks**, enabling more helpful responses and higher-quality text generation. - **Enhanced capabilities** in **256K long-context understanding**.
翻訳待ち・原文を表示
入力 テキスト→ 出力 テキスト
We introduce the updated version of the **Qwen3-4B non-thinking mode**, named **Qwen3-4B-Instruct-2507**, featuring the following key enhancements: **Qwen3-4B-Instruct-2507** has the following features: - Type: Causal Language Models - Training Stage: Pretraining & Post-training - Number of Parameters: 4.0B - Number of Paramaters (Non-Embedding): 3.6B - Number of Layers: 36 - Number of Attention Heads (GQA): 32 for Q and 8 for KV - Context Length: **262,144 natively**. **NOTE: This model supports only non-thinking mode and does not generate `` `` blocks in its output. Meanwhile, specifying `enable_thinking=False` is no longer required.**
翻訳待ち・原文を表示
入力 テキスト→ 出力 テキスト
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。
基本情報を収録・詳細は順次追加
入力 テキスト→ 出力 テキスト
See our collection for all versions of Qwen3 including GGUF, 4-bit & 16-bit formats. Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants. - Fine-tune Qwen3 (14B) for free using our Google Colab notebook here! - Read our Blog about Qwen3 support: unsloth.ai/blog/qwen3 - View the rest of our notebooks in our docs here. - Run & export your fine-tuned model to Ollama, llama.cpp or HF.
翻訳待ち・原文を表示
分類準備中
入力 未確認→ 出力 未確認
Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants. We introduce the updated version of the **Qwen3-4B non-thinking mode**, named **Qwen3-4B-Instruct-2507**, featuring the following key enhancements: - **Significant improvements** in general capabilities, including **instruction following, logical reasoning, text comprehension, mathematics, science, coding and tool usage**. - **Substantial gains** in long-tail knowledge coverage across **multiple languages**. - **Markedly better alignment** with user preferences in **subjective and open-ended tasks**, enabling more helpful responses and higher-quality text generation. - **Enhanced capabilities** in **256K long-context understanding**.
翻訳待ち・原文を表示
入力 テキスト→ 出力 テキスト
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。
入力 テキスト→ 出力 テキスト
Over the past three months, we have continued to scale the **thinking capability** of Qwen3-4B, improving both the **quality and depth** of reasoning. We are pleased to introduce **Qwen3-4B-Thinking-2507**, featuring the following key enhancements: **NOTE**: This version has an increased thinking length. We strongly recommend its use in highly complex reasoning tasks. **Qwen3-4B-Thinking-2507** has the following features: - Type: Causal Language Models - Training Stage: Pretraining & Post-training - Number of Parameters: 4.0B - Number of Paramaters (Non-Embedding): 3.6B - Number of Layers: 36 - Number of Attention Heads (GQA): 32 for Q and 8 for KV - Context Length: **262,144 natively**.
翻訳待ち・原文を表示
分類準備中
入力 未確認→ 出力 未確認
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。
入力 テキスト→ 出力 テキスト
See our collection for all versions of Qwen3 including GGUF, 4-bit & 16-bit formats. Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants. - Fine-tune Qwen3 (14B) for free using our Google Colab notebook here! - Read our Blog about Qwen3 support: unsloth.ai/blog/qwen3 - View the rest of our notebooks in our docs here. - Run & export your fine-tuned model to Ollama, llama.cpp or HF.
翻訳待ち・原文を表示
入力 テキスト→ 出力 テキスト
> [!Note] > This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format. > > These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, KTransformers, etc. > > In light of its parameter scale, the intended use cases are prototyping, task-specific fine-tuning, and other research or development purposes. Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Qwen3.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to empower developers and enterprises with unprecedented capability and efficiency. - **Unified Vision-Language Foundation**: Early fusion training on multimodal tokens achieves cross-generational parity with Qwen3 and outperforms Qwen3-VL models across reasoning, coding, agents, and visual understanding benchmarks.
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
> [!Note] > This repository contains model weights and configuration files for the pre-trained only model in the Hugging Face Transformers format. > > These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, etc. > > The intended use cases are fine-tuning, in-context learning experiments, and other research or development purposes, not direct interaction. > However, the control tokens, e.g., ` ` and ` ` were trained to allow efficient LoRA-style PEFT with the official chat template, mitigating the need to finetune embeddings, a significant optimization given Qwen3.5's larger vocabulary. Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Qwen3.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to empower developers and enterprises with unprecedented capability and efficiency.
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants. - You can now also fine-tune the model locally with Unsloth. - Read our Qwen3.5 fine-tuning guide here. > [!Note] > This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format. > > These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, KTransformers, etc. > > In light of its parameter scale, the intended use cases are prototyping, task-specific fine-tuning, and other research or development purposes.
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
NEW: MTP speculative decoding for ~1.5-2x faster generation - build llama.cpp from the MTP PR branch - has been merged 16th May 2026! Developer Role Support so Qwen3.5 can work in Codex, OpenCode and more! Tool calling improvements: Makes parsing nested objects to make tool calling succeed more. > [!Note] > This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format. > > These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, KTransformers, etc. > > In light of its parameter scale, the intended use cases are prototyping, task-specific fine-tuning, and other research or development purposes.
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
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。
基本情報を収録・詳細は順次追加
入力 テキスト · 画像→ 出力 テキスト
Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants. Disable thinking via `--chat-template-kwargs '{"enable_thinking":false}'`. Read our guide. - All GGUFs now use our new imatrix data. See some improvements in chat, coding, long context, and tool-calling use-cases. - GGUFs updated with an improved quantization algorithm. - Updated with fixed chat template for improved tool-calling & coding performance! - See our new benchmarks for 122B-A10B here.
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants. NEW: MTP speculative decoding for ~1.5-2x faster generation - build llama.cpp from the MTP PR branch - has been merged 16th May 2026! > **Note:** llama.cpp renamed `--spec-type mtp` to `--spec-type draft-mtp` on 2026-05-13
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
Do not use the Marlin backend (around 2x slower); let vLLM auto-select the NVFP4 kernel. > [!Note] > This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format. > > These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, KTransformers, etc. Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Qwen3.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to empower developers and enterprises with unprecedented capability and efficiency.
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
> [!Note] > This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format. > > These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, KTransformers, etc. Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Qwen3.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to empower developers and enterprises with unprecedented capability and efficiency. GPT-5-mini 2025-08-07 Claude-Sonnet-4.5 Qwen3-VL-235B-A22B Qwen3.5-122B-A10B Qwen3.5-27B Qwen3.5-35B-A3B
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants. Disable thinking via `--chat-template-kwargs '{"enable_thinking":false}'`. Read our guide. - All GGUFs now use our new imatrix data. See some improvements in chat, coding, long context, and tool-calling use-cases. - Rest of variants like UD_K_XL are now uploaded. Please follow the correct instructions / settings in our guide here.
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
NEW: MTP speculative decoding for ~1.5-2x faster generation - build llama.cpp from the MTP PR branch - has been merged 16th May 2026! Developer Role Support so Qwen3.5 can work in Codex, OpenCode and more! Tool calling improvements: Makes parsing nested objects to make tool calling succeed more. > [!Note] > This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format. > > These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, KTransformers, etc.
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
> [!Note] > This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format. > > These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, KTransformers, etc. > > In light of its parameter scale, the intended use cases are prototyping, task-specific fine-tuning, and other research or development purposes. Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Qwen3.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to empower developers and enterprises with unprecedented capability and efficiency. - **Unified Vision-Language Foundation**: Early fusion training on multimodal tokens achieves cross-generational parity with Qwen3 and outperforms Qwen3-VL models across reasoning, coding, agents, and visual understanding benchmarks.
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
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。
入力 テキスト · 画像→ 出力 テキスト
