# POCKET-26B **On-device Korean AI, based on Google Gemma4-26B-A4B.** Runs on your PC or phone with **no GPU** — in Ollama, LM Studio, PocketPal, or any llama.cpp app. ## Run
EXAONE Deep exhibits superior capabilities in various reasoning tasks including math and coding benchmarks, ranging from 2.4B to 32B parameters developed and released by LG AI Research.
TheDrummer/Cydonia-24B-v4.3
TheDrummer/Gemmasutra-Pro-27B-v1
TheDrummer/Cydonia-22B-v1.1
TheDrummer/Cydonia-24B-v3.1
ReadyArt/Fever-Dream-22B-v1.1-Q6_K-GGUF
TheDrummer/Big-Tiger-Gemma-27B-v1
TheDrummer/Cydonia-24B-v3.1 (Q8_0)
Phi-2 is a Transformer with 2.7 billion parameters. It was trained using the same data sources as Phi-1.5, augmented with a new data source that consists of various NLP synthetic texts and filtered websites (source: Microsoft).
ReadyArt/Fever-Dream-22B-v1.1-Q6_K (Q6_K)
TheDrummer/Rocinante-12B-v1.1
This is from https://huggingface.co/TheDrummer/Rocinante-12B-v1.1-GGUF
Verified configuration + measured results for abliterated (uncensored) Qwen3.8-27B running as an autonomous redteam/sysadmin agent, served via llama-swap. This card documents OUR tested configuration and results on live tasks
TheDrummer/Behemoth-123B-v1.1
Lumma-fev-27b-v2 is a decision model from Frontiersmind.ai, fine-tuned from Qwen3.8-27B.
Dolphin Mistral 24B Venice Edition is a collaborative project with Venice.ai
Quantized version of DeepSeek-R1-32B optimized for tool usage with Cline / Roo Code and complex problem solving.
DeepSeep V3 from March 2025 Merged from Unsloth's HF - 671B params - Q8_0/713 GB & Q4_K_M/404 GB
Quantized Q4_K_S version of Devstral small 2 24B FP16 model