GGUF quantizations of ThinkingCap-Qwen3.8-27B-abliterated, the uncensored variant of bottlecapai/ThinkingCap-Qwen3.8-27B, for llama.cpp and apps built on it. Nine files from 13.0 GB (IQ3_M) to 29.0 GB (Q8_0): six with llama.cpp's standard mix for their type and three (`UD-*`) with Unsloth's per-tensor layouts. Every file is quantized directly from the BF16 GGUF, never from another quant, and every file below Q8_0 uses an importance matrix computed on this model. In every file the MTP head is pinned to Q8_0 (see Tensor pins). **Which build?** Q5_K_M (19.7 GB) is the recommended file. Below it, by memory: UD-Q4_K_XL (17.7 GB), IQ4_XS (15.5 GB), UD-IQ4_XS (14.4 GB), UD-Q3_K_XL (13.2 GB); Q6_K (22.5 GB) or Q8_0 (29.0 GB) when the memory is there. UD-Q4_K_XL and UD-Q3_K_XL have lower divergence than Q4_K_M and IQ3_M at a slightly larger size.
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
A 4-bit weight-only quantization (NVFP4 weights, bf16 activations) of ThinkingCap-Qwen3.8-27B-abliterated, the uncensored variant of bottlecapai/ThinkingCap-Qwen3.8-27B. 20.6 GB, down from 55.6 GB in bf16, at KL 0.0155 against the bf16 model. The weights fit a 24 GB GPU with a little room for KV cache; 32 GB and up is comfortable. vLLM serves it on Ampere and newer (sm80+) with a weight-only kernel: Marlin FP4 on most GPUs (Ampere, Ada, Hopper, RTX 50, DGX Spark) and a dedicated BF16×FP4 kernel on B200/B300. Activations stay bf16, so no GPU computes in FP4 here. **Which build?** The W4A16 build (19.5 GB, KL 0.0260) runs on the same GPUs (vLLM: Marlin, Machete on Hopper) and is 1.1 GB smaller; this build has the lower KL. Speed was not measured for either. With memory to spare, the FP8-BLOCK build (31.3 GB, KL 0.0046) has the lowest KL of the published builds; on Ada GPUs use FP8-DYNAMIC (31.3 GB, KL 0.0068) instead.
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 音声→ 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 音声→ 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 音声→ 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 音声→ 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 音声→ 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 音声→ 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 音声→ 出力 テキスト
NVIDIA Parakeet TDT 0.6B V3 (Multilingual) model converted to ONNX format for onnx-asr.
翻訳待ち・原文を表示
入力 音声→ 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。
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。
基本情報を収録・詳細は順次追加
入力 音声→ 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 埋め込み
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
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
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
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
**Independently computed using dedicated cloud compute clusters, funded 100% out-of-pocket.** If you consider these releases to be of high quality and they help you in your day-to-day workflow, consider supporting on Ko-fi. Every contribution directly sustains future cloud compute hours and helps support my university studies. > *Independently computed on private cloud clusters. If this handcrafted release saves you VRAM and runs faster on your GPU, consider fueling the community compute fund on Ko-fi.*
翻訳待ち・原文を表示
入力 テキスト · 画像→ 出力 テキスト
If you consider these releases to be of high quality and they help you in your day-to-day workflow, consider supporting on Ko-fi. Every contribution directly sustains future cloud compute hours and helps support my university studies. > These are complementary APEX-I releases, not alternate downloads of the same model. Each receives the same surgical tensor-by-tensor approach and a design suitable for full or partial system-RAM inference: > - **Cyber-Tiel-Coder-35B-A3B APEX-I-MiniPlus-V2.1** — specialist uncensored agentic coding MoE tuned for high-speed local development (14.75 GB / Q5–Q6 quality).
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→ 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
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
> > The legacy build suffered from severe recurrent state instability, context degradation, and unintended language mixing during generation. > > **The current release completely resolves these issues:** > - Complete long-context associative recall and conversational stability restored. > - Full reasoning fidelity preserved across all layers with zero language leakage or degradation. > - Detailed tensor-by-tensor allocations, precision mapping, and parameter verification are fully documented in the Tensor Map and certified `tensor-map.tsv` file. > > **Action Required:** Please delete any older `.gguf` file downloaded prior to today and download the newly published certified build: **`Huihui-Qwen3.8-27B-Abliterated-VAL-APEX-I-MiniPlus-V3.gguf`**. If you consider these releases to be of high quality and they help you in your day-to-day workflow, consider supporting on Ko-fi. Every contribution directly sustains future cloud compute hours and helps support my university studies. VAL-APEX-I stands for: **Vector-calibrated Asymmetric Layer-wise Outlier-preserving Recurrent-aware Unified Matrix-quantization**
翻訳待ち・原文を表示
入力 テキスト→ 出力 テキスト
> > The legacy build suffered from severe recurrent state instability, context degradation, and unintended language mixing during generation. > > **The current release completely resolves these issues:** > - Complete long-context associative recall and conversational stability restored. > - Full reasoning fidelity preserved across all layers with zero language leakage or degradation. > - Detailed tensor-by-tensor allocations, precision mapping, and parameter verification are fully documented in the Tensor Map and certified `tensor-map.tsv` file. > > **Action Required:** Please delete any older `.gguf` file downloaded prior to today and download the newly published certified build: **`Huihui-Qwen3.8-27B-Abliterated-VAL-APEX-I-NanoPlus-V3.gguf`**. If you consider these releases to be of high quality and they help you in your day-to-day workflow, consider supporting on Ko-fi. Every contribution directly sustains future cloud compute hours and helps support my university studies. VAL-APEX-I stands for: **Vector-calibrated Asymmetric Layer-wise Outlier-preserving Recurrent-aware Unified Matrix-quantization**
翻訳待ち・原文を表示
入力 テキスト→ 出力 テキスト
