Browse AI skills, agents and automations

Meta-Llama-3.1-8B-Instruct-AWQ-INT4

hugging-quants · Kit

Runs Meta's Llama 3.1 8B Instruct chat model in 4-bit AWQ form, so it fits on a GPU with about 4 GiB of free memory for the weights. For developers who want a…

Phi-4-mini-instruct

microsoft · Kit

A 3.8B open instruction model from Microsoft with a 128K token context, built for reasoning, math and logic in tight memory or latency budgets. It suits develo…

Ornith-1.5-9B-GGUF by AtomicChat

AtomicChat · Kit

GGUF builds of Ornith 1.5 9B, a model that reads images and text, in sizes from 2.81 GB to 17.9 GB plus a vision projector. Pick the file that fits your graphi…

Hy-Memory-30B-A3B

tencent · Kit

Writes an assistant's long-term memory, choosing what to keep and how to update it inside Mem0, MemOS or Hy-Memory, with about 3B active parameters. For develo…

Ornith-1.5-122B-NVFP4-DFlash2-RTX-Spark

Sherlock-Comms · Kit

Serves the 122B Ornith 1.5 model through vLLM with DFlash2 speculative decoding, using an unquantized output head. Made for RTX Spark laptops with 128 GB of me…

Sarvam-105B GGUF

sarvamai · Kit

The GGUF version of Sarvam-105B, a large mixture-of-experts model with a focus on Indian languages, split into 9 shards at 4-bit for llama.cpp. For people with…

Chatterbox-Nano

ResembleAI · Kit

Speaks English text aloud on a plain CPU with a 110M-parameter model, and adds laughs and coughs through tags in the text. Suits developers who need fast on-de…

Qwen3.8-27B-mixed-3.5bpw-MLX

rapid-mlx · Kit

A 27B vision-language model compressed to 3.5 bits per weight for Apple Silicon Macs with 24 GB of memory, served locally through Rapid-MLX. Suits Mac users an…

Qwen2.5-7B-Instruct-AWQ

Qwen · Kit

A 4-bit AWQ build of Qwen2.5-7B-Instruct, a 7.6B chat model that follows instructions, handles tables and returns JSON. For developers who want a capable chat…

Nanbeige4.3-33B-INT4

Nanbeige · Kit

INT4 quantized version of Nanbeige4.3-33B, a 33B model for coding and professional agent tasks, made with quantization-aware training. For developers who want…

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

mradermacher · Kit

GGUF quantized files of a 27.3B coding-distilled Qwen3.8 model, from 2-bit to 8-bit with an optional multi-modal supplement. They suit developers who run a cod…

JevK5-chat-mlx-4bit

mlx-community · Kit

A 4-bit MLX copy of the 4.2B JevK5 chat model for Apple Silicon, with a 262,144 token context and a small memory footprint. Made for Mac users who want a quick…

DeepSeek-V4.1-Flash-1017-oQ4e-ngram-mtp

mlx-community · Kit

A mixed-precision MLX copy of DeepSeek-V4.1-Flash-1017 that keeps the Engram tables at FP8. Made for people with hundreds of GB of memory who want to run the m…

Qwen3.8-27B-oQ4e-mtp

Jundot · Kit

Runs the 27B Qwen3.8 vision-language model on MLX in about 17.5 GB, reading text, images and video with a 262,144-token context. A mixed-precision quantization…

Qwen3.6-35B-A3B-2Bit-GSQ

ISTA-DASLab · Kit

A 2-bit quantized copy of Qwen3.6-35B-A3B made with GSQ, a Gumbel-Softmax method, that keeps most reasoning and coding ability at a much smaller memory footpri…

SmolVLM2-2.2B-Instruct

HuggingFaceTB · Kit

Answers questions about videos, images and text together, with a 2.2B model that needs about 5.2 GB of GPU memory for video. For developers building visual ass…

svd-linear-attention-roberta-base-adasplash

Hooshaai · Kit

Classifies English text, such as SST-2 sentiment, with a RoBERTa Base model whose attention is made sparse by AdaSplash, trading some accuracy for lower latenc…

roberta-base-fwsvd-rank32

Hooshaai · Kit

Classifies English text with a RoBERTa Base model shrunk to about 82M parameters by Fisher-weighted SVD at rank 32, trading some accuracy for lower latency and…

roberta-base-adasplash

Hooshaai · Kit

A RoBERTa Base sentiment classifier whose attention is swapped for a sparse AdaSplash version, which zeroes out 40% of attention weights to cut time and memory…

hs-rerank-1

hindsight-ai · Kit

Scores how well each document answers a query, in eight languages and for inputs up to 8,192 tokens, so a search or memory system can put the best matches firs…