A 60M-parameter GPT pretrained and instruction-tuned on a single 8GB-RAM NVIDIA Jetson, offering a native 2048-token context and real multi-turn conversation support.
12 Pulls 1 Tag Updated 3 weeks ago
Ollama Agent + Skills + Vision 26 t/s en 8 VRAM 2026
20 Pulls 1 Tag Updated 1 month ago
The DictaLM-2.0-Instruct Large Language Model (LLM) is an instruct fine-tuned version of the DictaLM-2.0 generative model using a variety of conversation datasets. For full details of this model please read this post: https://dicta.org.il/dicta-lm
1,856 Pulls 7 Tags Updated 2 years ago
Abliterated (refusal-direction removed, Arditi et al. 2024) variant of `Qwen/Qwen3.5-4B`. **Not** fine-tuned — no preference/instruction data added; only the refusal direction is orthogonalized out.
806 Pulls 1 Tag Updated 2 months ago
Specialized LLM for automated German IT job applications, ATS optimization, and cover letter generation for the 2026 German labor market.
12 Pulls 1 Tag Updated 1 month ago
Abstractive summarization using gpt-oss-20B as base model.
29 Pulls 1 Tag Updated 9 months ago
SHS AI Is an AI Made by a student of Sekolah Hamidah Sampurna (SHS) school in Indonesia in 2025.
7 Pulls 1 Tag Updated 1 year ago
A 60M-parameter GPT trained from scratch on a single 8GB-RAM NVIDIA Jetson, offering a native 2048-token context for raw text completion. Base pretrained checkpoint, not instruction-tuned.
2 Pulls 1 Tag Updated 3 weeks ago
Abliterated (decensored) version of google/gemma-4-E4B-it, created using Heretic v1.2.0 with Arbitrary-Rank Ablation (ARA) and row-norm preservation.
13.1K Pulls 5 Tags Updated 4 months ago
First uncensored reasoning model — trained on 27,699 real reasoning examples. 0% refusal rate. Chain-of-thought, verification, agent critique loops. Based on Qwen3-8B, 8 quants. Runs on phones.
322 Pulls 7 Tags Updated 1 month ago
ReadyArt/Broken-Tutu-24B-Transgression-v2.0
1,737 Pulls 3 Tags Updated 1 year ago
ReadyArt/Broken-Tutu-24B-Transgression-v2.0 (i1-Q6_K)
73 Pulls 1 Tag Updated 1 year ago
ReadyArt/Broken-Tutu-24B-Transgression-v2.0 (Q8_0)
14 Pulls 1 Tag Updated 1 year ago
Built a custom AI assistant using Ollama 3.2 that helps generate clean and useful code, with knowledge up to 2023.
14 Pulls 1 Tag Updated 6 months ago
Local-first AI tool router. 2B/4B/9B read images (vision); 27B text-only. 14B/32B retired. 99.1-100% routing accuracy (BFCL). 97% of traffic stays local.
386 Pulls 6 Tags Updated 1 week ago
Athenea-4B-Coding is a fine-tuned version of huihui-ai/Huihui-Qwen3-4B-Thinking-2507-abliterated, specialized in code reasoning, debugging, and problem solving.
521 Pulls 1 Tag Updated 6 months ago
2-bit Q2_K_XL quantized GGUF version of Qwen3-235B-A22B-Thinking-2507 (MoE, 22B active), optimized for deep reasoning with a 262K context window. Runs on Ollama with ~86.5 GiB RAM.
1,410 Pulls 1 Tag Updated 1 year ago
llava-NousResearch_Nous-Hermes-2-Vision-GGUF_Q4_0 with function calling
1,769 Pulls 1 Tag Updated 2 years ago
Abliterated version of mistralai/Mistral-Small-Instruct-2409
379 Pulls 1 Tag Updated 1 year ago
State-of-the-art OCR (Optical Character Recognition) vision language model based on [allenai/olmOCR-2-7B-1025](https://huggingface.co/allenai/olmOCR-2-7B-1025).
7,760 Pulls 1 Tag Updated 10 months ago