AI projects from Hugging Face

Qwen3.8-27B Splash 8-bit

dmeight · Kit

Runs Qwen3.8-27B at 8-bit through the Splash runtime in LM Studio on Apple silicon. An unofficial package that keeps the model closer to the original than the…

Cloudflare clef GGUF

bartowski · Kit

Static GGUF quantizations of Cloudflare's clef decision model, which scores every answer option to typed questions about a state in one pass instead of writing…

Qwen3.8-Whittle-MoE-27B-MLX-4bit

barozp · Kit

A 4-bit MLX build for Apple Silicon of a mixture-of-experts model carved from the dense Qwen3.8-27B, with 17.8B active of 27B total parameters. For people who…

Qwen3.8-27B-NVFP4-DGX-SPARK

autotrust · Kit

Speeds up Qwen3.8-27B NVFP4 decoding about 3.9 times on a DGX Spark, using the same weights and a tuned speculative-decoding setup. For teams serving it with v…

Qwen3.8-27B-Quark-AWQ-MXFP4

amd · Kit

Runs Qwen3.8-27B as a smaller MXFP4-quantized copy made with AMD Quark and AWQ, served through vLLM. Made for developers who want a compressed 27B vision-langu…

Qwen3.8-27B Agention Precision GGUF

agentionai · Kit

Packs Qwen3.8-27B as llama.cpp GGUF files in eight sizes from 8.95 to 16.35 GiB, plus a vision projector. The author reports less drift from the BF16 model tha…

Qwen3.8-Flash-Next-125B-UltraLite-37GiB-GGUF

0xKitkat · Kit

Squeezes the Qwen3.8-Flash-Next mixture-of-experts model into about 37 GiB of text-only GGUF files for experiments. It needs a patched llama.cpp build that com…

Laplace-1-Atlas

JuneoAI · Kit

Helps organizations decide who does a task, who sees information, what runs and what gets accepted, by scoring candidate options in one pass. A 27.4B model bui…

Qwen3.8-27B-QSA-GSQ-RCO-GGUF

pentacoxian-dev · Kit

Runs Qwen3.8-27B with sparse attention on a single 24 GB GPU at a 262K-token context with image input, as GGUF files for people with a custom llama.cpp engine…

Qwen3.8-27B-CodeForge-3xMerge i1 GGUF

mradermacher · Kit

Importance-matrix GGUF quants of the Qwen3.8-27B-CodeForge-3xMerge model, from about 11 GB to 22.5 GB, for local use. For people running 27B models on their ow…

Qwen3.6-35B-A3B lododo-Distill-v1.0 (GGUF)

lododoai · Kit

Runs as a local text model, a Qwen3.6-35B-A3B fine-tune trained on data written by Qwen3.8-27B and packed as one GGUF file for llama.cpp. For people who run mo…

Qwen3.8-27B-Fable-5-Coding-Distilled

khazarai · Kit

Debugs and reviews code by checking real files before it proposes a fix, a Qwen3.8-27B fine-tune trained on 195M tokens of reasoning traces. For developers who…

Qwen3.8-27B-SLQ-GGUF

kaitchup · Kit

Compresses Qwen3.8-27B into three GGUF files with different size and fidelity levels, using a custom statistically lossless quantization. For people who run th…

Qwen3.8-27B-oQ6e-mtp

Jundot · Kit

A mixed-precision MLX copy of Qwen3.8-27B at about 6.9 bits per weight, with the vision encoder and multi-token prediction head included, for running locally o…

Qwen3.8-27B Astra stop edit

jarrodwatts · Kit

A small weight edit for Qwen3.8-27B, with a script that applies it, that cuts the model's thinking length on open-ended tasks by changing a single matrix, with…

Qwopus3.8-27B-Flash-V2-GGUF

Jackrong · Kit

Trims wasted reasoning in long agent tasks, as a GGUF build of the second Flash fine-tune of Qwen3.8-27B with better Python formatting. For people who run agen…

Qwopus3.8-27B-Coder-Flash

Jackrong · Kit

Writes code, calls tools and works through software tasks as a coding agent. A 27.8B fine-tune of Qwen3.8-27B by Jackrong, trained on about 200,000 code and re…

Qwen3.8-27B-DFlash2 by incoai

incoai · Kit

Speeds up Qwen3.8-27B by drafting blocks of tokens that the main model then checks, with output unchanged. A 1.9B draft model for SGLang and vLLM servers, not…

Holo4-27B

Hcompany · Kit

Looks at screenshots and decides the next click, keystroke, code run or tool call, built on Qwen3.8-27B. A computer-use model for web, desktop and mobile agent…

Qwen3.8-27B Intel Arc Tuned

Greatjedi · Kit

Runs a 27B Qwen3.8 model in GGUF files retuned for Intel Arc GPUs, with pure decode near 22 tokens a second in the author's tests. For people running llama.cpp…