A large language model built by the Technology Innovation Institute (TII) for use in summarization, text generation, and chat bots.
1.2M Pulls 38 Tags Updated 2 years ago
Cogito v1 Preview is a family of hybrid reasoning models by Deep Cogito that outperform the best available open models of the same size, including counterparts from LLaMA, DeepSeek, and Qwen across most standard benchmarks.
2.1M Pulls 20 Tags Updated 1 year ago
Google Gemma-3-4B-it llama.cpp quantization by Fabrício Carraro
200 Pulls 1 Tag Updated 1 year ago
🦎 Salamandra-7b-instruct llama.cpp quantization by Fabrício Carraro
43 Pulls 1 Tag Updated 1 year ago
7 Pulls 1 Tag Updated 1 year ago
1 Tag Updated 1 year ago
LFM2 is a family of hybrid models designed for on-device deployment. LFM2-24B-A2B is the largest model in the family, scaling the architecture to 24 billion parameters while keeping inference efficient.
1.1M Pulls 6 Tags Updated 7 months ago
LFM2.5 is a new family of hybrid models designed for on-device deployment.
1.3M Pulls 5 Tags Updated 8 months ago
# You take a better ideas and data, and webhooks, and masters, for the all carrers in the world, and you understend the language i used to comunicated with you, you are mi TODOLOG@
30 Pulls 1 Tag Updated 2 years ago
Bringing the monolithic GGUF to HauhauCS's work of aggressively uncensoring Gemma 4. For edge models having full audio and vision support; there are multi-modal specific focused models available [Vision OR Sound].
36.3K Pulls 8 Tags Updated 5 months ago
Legal reasoning specialist — 139/150 (93%). Contract law, case analysis, statutory interpretation, compliance, legal writing, jurisprudence. Based on Qwen2.5-1.5B, 32K context. Uncensored.
249 Pulls 12 Tags Updated 2 months ago
Cybersecurity specialist — 144/150 (96%). Vulnerability analysis, exploit development, cryptography, network security, reverse engineering, incident response. Based on Qwen2.5-1.5B, 32K context. Uncensored.
1,292 Pulls 12 Tags Updated 2 months ago
Perfect censorship bypass (5.0/5) + superior tool use (5.0/5). 13.8/15 total. Best-in-class for shell commands, API calls, Docker, K8s. Trained on Fable5 traces + tool augmentation. 5GB, 16K context. Apache 2.0.
452 Pulls 2 Tags Updated 2 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.
368 Pulls 7 Tags Updated 2 months ago
Scientific reasoning specialist — 140/150 (93%). Physics, chemistry, biology, mathematics, earth science, scientific method. Based on Qwen2.5-1.5B, 32K context. Uncensored.
216 Pulls 12 Tags Updated 2 months ago
1 Pull 1 Tag Updated 2 years ago
Tinychatbot-v1.0 is a finetune of Meta's Llama3.2, it was trained on databricks-dolly-15k.
8 Pulls 1 Tag Updated 1 month ago
Best all-rounder uncensored AI agent. 13.9/15 total score. Strong reasoning (4.5), tool use (4.6), censorship bypass (4.8). Trained on real Fable5 agent traces. 5GB, 16K context. Shell commands, code gen, multi-step reasoning. Apache 2.0.
426 Pulls 2 Tags Updated 2 months ago
Matthew William's demo in video 4 of Ollama Course: Basics. This is the model from hugging face brought into ollama: ClosedCharacter/Peach-9B-8k-Roleplay . Quantized to Q4_K_M
85 Pulls 1 Tag Updated 1 year ago
A fine-tuned Salamandra model representing Estigia, a brilliant and playful CubeSat from the Pluton UPV team. Designed to talk about space with a charming, child-like personality, capable of conversing fluently in English, Spanish, and Catalan.
1 Pull 1 Tag Updated 5 months ago