Open models to add in one click

Collected from Hugging Face, GitHub and npm: open LLMs in GGUF, speech models like Whisper, embedding models. Every card links to its source, repeats the license and the numbers, and adds a short summary. One button adds it to LogiShell; without the app it leads to the install and brings you back.

Qwen3-Coder-30B-A3B-Instruct-GGUF

model · LLM · by unsloth

See our collection for all versions of Qwen3 including GGUF, 4-bit & 16-bit formats.

apache-2.017 GB~21 GB RAMsource alive

8.6M downloads · 1.1K likes · numbers as of 2026-10-02
Ornith-1.5-9B-GGUF

model · LLM · by ornith-ai

Chirp Chirp! 🐦 We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement.

mit5.4 GB~7 GB RAMsource alive

5.1M downloads · 479 likes · numbers as of 2026-10-02
Ternary-Bonsai-2-27B-gguf

model · LLM · by prism-ml

Prism ML Website Whitepaper Demo & Examples Discord

apache-2.050 GB~59 GB RAMsource alive

3.9M downloads · 2.4K likes · numbers as of 2026-10-02
Ornith-1.5-35B-A3B-GGUF

model · LLM · by ornith-ai

Chirp Chirp! 🐦 We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement.

mit20 GB~24 GB RAMsource alive

3.6M downloads · 473 likes · numbers as of 2026-10-02
Ornith-1.0-9B-GGUF

model · LLM · by ornith-ai

Aloha! 🌺 Today, we are releasing Ornith-1.0, a self-improving family of open-source models for agentic coding.

mit5.2 GB~7 GB RAMsource alive

2.6M downloads · 673 likes · numbers as of 2026-10-02
JiRackUltra_1b

model · LLM · by CMSManhattan

JiRack Ultra 1B (CPU) A fast and efficient ~1.5B model optimized for CPU inference. The model was refactored with BitNet features and an updated tokenizer that includes new Routing, Tool call, and Ro…

mit1.0 GB~2 GB RAMsource alive

2.5M downloads · 0 likes · numbers as of 2026-10-02
mxbai-embed-large-v1

model · Embeddings · by mixedbread-ai

embedding model by mixedbread-ai, gguf files on Hugging Face.

apache-2.0639 MB~2 GB RAMsource alive

2.0M downloads · 826 likes · numbers as of 2026-10-02
nemotron-3.5-asr-streaming-0.6b-gguf

model · Speech · by handy-computer

nemotron-3.5-asr-streaming-0.6b: transcribe.cpp GGUF

other473 MB~2 GB RAMsource alive

1.8M downloads · 12 likes · numbers as of 2026-10-02
Ornith-1.0-35B-GGUF

model · LLM · by ornith-ai

Aloha! 🌺 Today, we are releasing Ornith-1.0, a self-improving family of open-source models for agentic coding.

mit20 GB~24 GB RAMsource alive

1.7M downloads · 1.1K likes · numbers as of 2026-10-02
parakeet-unified-en-0.6b-gguf

model · Speech · by handy-computer

parakeet-unified-en-0.6b: transcribe.cpp GGUF

cc-by-4.0455 MB~2 GB RAMsource alive

1.6M downloads · 9 likes · numbers as of 2026-10-02
deepseek-v4-gguf

model · LLM · by antirez

This quants are specific for the DS4 inference engine. They may work with other inference engines or not (they should, but not the MTP model which requires a specific loader).

mit3.5 GB~5 GB RAMsource alive

1.5M downloads · 477 likes · numbers as of 2026-10-02
Ornith-1.5-397B-GGUF

model · LLM · by ornith-ai

Chirp Chirp! 🐦 We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement.

mit228 GB~263 GB RAMsource alive

1.3M downloads · 40 likes · numbers as of 2026-10-02
nemotron-3.5-asr-streaming-0.6b

model · Speech · by nvidia

This model is the multilingual extension of nvidia/nemotron-speech-streaming-en-0.6b, adding language-ID prompt conditioning to support transcription across 40 language-locales from a single model.

other708 MB~2 GB RAMsource alive

1.2M downloads · 1.2K likes · numbers as of 2026-10-02
LFM2.5-2.6B-GGUF

model · LLM · by LiquidAI

LFM2.5 is a new family of hybrid models designed for on-device deployment. It builds on the LFM2 architecture with extended pre-training and reinforcement learning.

other1.6 GB~3 GB RAMsource alive

1.1M downloads · 360 likes · numbers as of 2026-10-02
GLM-5.3-Flash-GGUF

model · LLM · by unsloth

Unsloth Dynamic 3.0 achieves superior accuracy & outperforms other leading quants.

mit9 MB~1 GB RAMsource alive

1.0M downloads · 443 likes · numbers as of 2026-10-02
cohere-transcribe-03-2026-gguf

model · Speech · by handy-computer

cohere-transcribe-03-2026: transcribe.cpp GGUF

apache-2.01.5 GB~3 GB RAMsource alive

978K downloads · 3 likes · numbers as of 2026-10-02
JiRackUltra_14b

model · LLM · by CMSManhattan

JiRack Ultra 14B (CPU) A fast and efficient 14B model optimized for CPU inference. The model was refactored with BitNet features and an updated tokenizer that includes new Routing, Media, Vision, Sou…

mit8.4 GB~11 GB RAMsource alive

964K downloads · 2 likes · numbers as of 2026-10-02
Qwen3.8-27B-NVFP4-MTP-GGUF

model · LLM · by esatapedico

A family of nine GGUF files of Qwen3.8-27B (the native vision-language 27B dense model, Gated DeltaNet + Gated Attention hybrid layout, 262,144-token native context, MTP speculative head), converted…

apache-2.014 GB~17 GB RAMsource alive

752K downloads · 115 likes · numbers as of 2026-10-02
gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF

model · LLM · by yuxinlu1

💻🤖 Gemma4-12B v2 — Coding + Agentic Edition ✨ ### 🐣 Tiny footprint, big brain — a local coding & tool-using agent for everyone

apache-2.06.9 GB~9 GB RAMsource alive

710K downloads · 1.6K likes · numbers as of 2026-10-02
Qwen3.8-9B-Distill-GGUF

model · LLM · by empero-ai

GGUF quantizations of empero-ai/Qwen3.8-9B — a full-parameter distillation of Qwen3.8 2.4T A95B into the Qwen3.5-9B architecture — for llama.cpp, Ollama, LM Studio, Jan, KoboldCpp, and other stock GG…

apache-2.05.4 GB~7 GB RAMsource alive

676K downloads · 286 likes · numbers as of 2026-10-02
POCKET-35B-GGUF

model · LLM · by FINAL-Bench

🆕 POCKET-Qwen3.8-Flash-Next — a 180B model running on a laptop with 8 GB VRAM + 32 GB RAM · 4.17 tok/s measured. >

apache-2.020 GB~24 GB RAMsource alive

669K downloads · 81 likes · numbers as of 2026-10-02
Qwen3.8-4B-Distill-GGUF

model · LLM · by empero-ai

GGUF quantizations of empero-ai/Qwen3.8-4B — a full-parameter distillation of Qwen3.8 2.4T A95B into the Qwen3.5-4B architecture — for llama.cpp, Ollama, LM Studio, Jan, KoboldCpp, and other stock GG…

apache-2.02.6 GB~4 GB RAMsource alive

666K downloads · 161 likes · numbers as of 2026-10-02
Qwen3.8-Flash-Next-GGUF

model · LLM · by AtomicChat

Built from Qwen's original weights with our own importance matrix. The calibration corpora behind our builds are public.

other553 MB~2 GB RAMsource alive

634K downloads · 170 likes · numbers as of 2026-10-02
Ternary-Bonsai-27B-gguf

model · LLM · by prism-ml

Prism ML Website Whitepaper Demo & Examples Discord

apache-2.050 GB~59 GB RAMsource alive

628K downloads · 1.4K likes · numbers as of 2026-10-02