言語から画像、オープンウェイトからAPIまで。能力と実際の費用を比較。
次のアイデアに合うAIを見つけましょう。
Built from Ornith's original weights with our own importance matrix. The calibration corpora and raw eval logs behind this card are public. See our quantization analysis below for measurements and instructions. You can now run Ornith 1.5 in Atomic Chat with toggles for thinking. GGUF builds of ornith-ai/Ornith-1.5-35B-A3B, converted and measured by AtomicChat.
翻訳待ち・原文を表示
Built from Ornith's original weights with our own importance matrix. The calibration corpora and raw eval logs behind this card are public. Learn how to run Ornith 1.5 9B locally - read our guide . You can now run Ornith 1.5 in Atomic Chat with toggles for thinking. See our quantization analysis below for measurements and instructions. GGUF builds of ornith-ai/Ornith-1.5-9B, converted and measured by AtomicChat.
翻訳待ち・原文を表示
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
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
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
The 8-step Qwen-Image-2.1 as GGUF for stable-diffusion.cpp, built from the published BF16 weights with our own importance matrix. The renders and measurements behind this card are public. Generate images locally in Atomic Chat . It runs stable-diffusion.cpp, the engine these files were built and checked with. Turbo drifts from the original picture faster than you might expect at 4 bits and below. Our AD-Q4_K stays 21% closer to the BF16 image than the plain Q4_K , for 150 MB more. Turbo needs its own 8-step schedule at CFG 1. The exact settings are in Running it . Looking for uncensored? The open Qwen-Image pipeline has no content filter. Qwen-Image-2.1-Turbo-Uncensored-GGUF removes the refusal direction from its text encoder. On images that changed nothing we could measure, and the card shows why. Each file renders the same 48 prompts × 2 seeds from the same starting noise as the BF16 weights. We compare every image with the BF16 image of the same prompt and seed.
翻訳待ち・原文を表示
Qwen-Image's text encoder, Qwen3-VL-8B-Instruct, with its refusal direction projected out: a drop-in --llm for stable-diffusion.cpp next to our Turbo denoiser GGUFs . Below, exactly which part is uncensored and what that does to the pictures. The measurements are public. As a chat model, the encoder's refusals fall from 88.9% to 1.2% on held-out English prompts. MMLU is unchanged: 77.35% before and after. On images it changes nothing we could measure. The stock Qwen-Image pipeline has no filter and already draws all 9 sensitive categories we tested, and with this encoder not one of 45 prompts changed outcome. See where exactly this is uncensored . Generate images locally in Atomic Chat . It runs stable-diffusion.cpp, the engine these files were built and checked with. Qwen-Image 2.1 makes a picture with three parts. Only one of them is changed here.
翻訳待ち・原文を表示
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
**Qwen3 4B Dflash b16**, self-quantized to GGUF by Atomic Chat. Built straight from Z Lab's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline. - **0.5B parameters**: the weights this repo quantizes. - **Context length**: 40,960 tokens (40K), as published by Z Lab. - **5 layers**: Dense decoder. - **Full imatrix ladder**: every quant is calibrated with an importance matrix. > [!NOTE] > These GGUFs are **self-quantized from the original weights**, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model.
翻訳待ち・原文を表示
**Qwen3.5 4B Dflash**, self-quantized to GGUF by Atomic Chat. Built straight from Z Lab's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline. - **0.6B parameters**: the weights this repo quantizes. - **Context length**: 262,144 tokens (256K), as published by Z Lab. - **6 layers**: Dense decoder, hybrid sliding-window (4096) and global attention. - **Full imatrix ladder**: every quant is calibrated with an importance matrix. > [!NOTE] > These GGUFs are **self-quantized from the original weights**, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model.
翻訳待ち・原文を表示
**Qwen3.5 4B**, self-quantized to GGUF by Atomic Chat. Built straight from Qwen's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline. > [!NOTE] > These GGUFs are **self-quantized from the original weights**, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model. > [!IMPORTANT] > Always pass `--jinja` so the **Qwen3.5 4B chat template** is applied. Without it the model can emit malformed turns.
翻訳待ち・原文を表示
**Qwen3.5 9B Dflash**, self-quantized to GGUF by Atomic Chat. Built straight from Z Lab's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline. - **1.3B parameters**: the weights this repo quantizes. - **Context length**: 262,144 tokens (256K), as published by Z Lab. - **6 layers**: Dense decoder, hybrid sliding-window (4096) and global attention. - **Full imatrix ladder**: every quant is calibrated with an importance matrix. > [!NOTE] > These GGUFs are **self-quantized from the original weights**, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model.
翻訳待ち・原文を表示
**Qwen3.5 9B**, self-quantized to GGUF by Atomic Chat. Built straight from Qwen's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline. > [!NOTE] > These GGUFs are **self-quantized from the original weights**, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model. > [!IMPORTANT] > Always pass `--jinja` so the **Qwen3.5 9B chat template** is applied. Without it the model can emit malformed turns.
翻訳待ち・原文を表示
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
**Qwen3.6 27B**, self-quantized to GGUF by Atomic Chat. Built straight from Qwen's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline. > [!NOTE] > These GGUFs are **self-quantized from the original weights**, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model. > [!IMPORTANT] > Always pass `--jinja` so the **Qwen3.6 27B chat template** is applied. Without it the model can emit malformed turns.
翻訳待ち・原文を表示
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
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
**Qwen3.6 35B A3B**, self-quantized to GGUF by Atomic Chat. Built straight from Qwen's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline. > [!NOTE] > These GGUFs are **self-quantized from the original weights**, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model. > [!IMPORTANT] > Always pass `--jinja` so the **Qwen3.6 35B A3B chat template** is applied. Without it the model can emit malformed turns.
翻訳待ち・原文を表示
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。
Built from Qwen's original weights with our own importance matrix. The calibration corpora behind our builds are public. Learn how to run Qwen3.8 27B locally - read our guide . You can now run Qwen3.8 in Atomic Chat with toggles for thinking. See our quantization analysis below for measurements and instructions. Every quantization of this model we could find, ours and everyone else's, measured against the original weights on the same held-out text. The dashed line is the best file available at each size from anyone but us.
翻訳待ち・原文を表示
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
Built from Qwen's original weights with our own importance matrix. The calibration corpora behind our builds are public. Built by the team behind Atomic Chat , the app that runs open-weight models locally and offline. Qwen3.8-Flash-Next is the first open-weight release of the architecture behind Qwen4. These GGUFs are self-quantized from Qwen's original weights with our own importance matrix, published alongside the quants. They need a llama.cpp build with Qwen3.8-Flash-Next support; Atomic Chat runs it as support ships. Need it without refusals? Qwen3.8-Flash-Next-Uncensored-GGUF has baked builds and a LoRA adapter for the quants here. Qwen3.8-Flash-Next has 177B parameters. Our 85 GB quant runs on an M5 Max with 64 GB of memory, with vision, at 36 tok/s. That is not a typo: the file is larger than the machine's entire RAM.
翻訳待ち・原文を表示
One refusal direction projected out of Qwen's original weights, shipped two ways: three baked GGUF builds, and a 318 MB LoRA adapter for the Flash-Next quants you already have. The measurements behind this card are public. Built by the team behind Atomic Chat , the app that runs open-weight models locally and offline. On held-out prompts, refusals fall from 92.6% to 1.2% in English, and to none of 30 with thinking on. MMLU moves by 0.5 points, inside the noise. The 93.9 GB build matches the original's token choice 91.1% of the time: closer to the original than our published 94.5 GB quant of the unmodified model (89.6%). Already running our Flash-Next quants ? Add the adapter with --lora-scaled instead of downloading a new model. *28 Oct 2026: re-split from 2 GB shards into three files per build: the weights in two, cut at 45 GB, and the n-gram table in a third of its own. The weights are unchanged (tensor-for-tensor identical).*
翻訳待ち・原文を表示
**Qwen3 Coder 30B A3B**, self-quantized to GGUF by Atomic Chat. Built straight from Qwen's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline. > [!NOTE] > These GGUFs are **self-quantized from the original weights**, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model. > [!IMPORTANT] > Always pass `--jinja` so the **Qwen3 Coder 30B A3B chat template** is applied. Without it the model can emit malformed turns.
翻訳待ち・原文を表示
提供元・分類・入出力の条件を変更してください。新しいモデルを随時追加しています。
月間トークン数を調整してAPI費用を比較。
収集済みの比較可能な100万トークン当たり米ドル料金です。キャッシュ割引、段階料金、追加推論出力、税金は含みません。最新料金は出典で確認してください。
入力・出力・キャッシュ料金を分け、同じ用途と使用量で総費用を比較します。画像・音声・動画は枚数、時間、解像度で課金される場合があります。
必ずしもそうではありません。バージョンごとのライセンス、商用制限、ポリシーを確認してください。自前運用には計算資源と保守費用も必要です。
順位はテスト、バージョン、評価方法によります。実際の業務データで精度、遅延、安定性、費用を比較してください。
