I upload models and quants that I like, or would like people to have easy downloads for.
-
LFM2
Small models made to run on mobile devices, developed by Liquid AI
350m 700m 1.2b 2.6b 8b6,256 Pulls 19 Tags Updated 2 months ago
-
qwen3
Useful Unsloth-DQ2 quants of the smaller qwen3 models
tools thinking 1.7b 4b 8b2,348 Pulls 21 Tags Updated 5 months ago
-
lfm2.5
An upgraded version of LFM2 trained on over twice as many tokens
1.2b2,094 Pulls 3 Tags Updated 3 weeks ago
-
dolphin3-llama3.2
Dolphin 3.0 llama3.2 🐬 - A powerful, customizable AI model for local use.
1b 3b2,009 Pulls 9 Tags Updated 9 months ago
-
dolphin3-qwen2.5
Dolphin 3.0 Qwen 2.5 🐬 - A powerful, customizable AI model for local use.
tools 1.5b 3b1,801 Pulls 9 Tags Updated 1 year ago
-
deepseek-r1-0528-qwen3
Useful DQ2 quants for deepseek-r1-0528-qwen3-distill:8b
8b1,598 Pulls 3 Tags Updated 10 months ago
-
qwen3-rerankertools
1,553 Pulls 2 Tags Updated 9 months ago
-
granite-4.0
Efficient, intelligent, and tiny models from IBM
tools 350m 1b 3.2b 3.4b 7b1,032 Pulls 17 Tags Updated 4 months ago
-
qwen3-embeddingtools
832 Pulls 2 Tags Updated 9 months ago
-
granite-embedding-multilingual
Embedding models supporting multiple languages, made by the IBM-Granite team, in 2 (small) sizes
embedding732 Pulls 4 Tags Updated 1 year ago
-
exaone-4.0
LGAI has developed their latest version of the exaone model series, now with reasoning!
1.2b549 Pulls 4 Tags Updated 5 months ago
-
amoral-gemma3-1b-v2
gemma3 finetuned for neutral, non-judgemental responses
427 Pulls 6 Tags Updated 8 months ago
-
lucy
A compact model focused on agentic web search
tools 1.7b418 Pulls 6 Tags Updated 7 months ago
-
olmoe-1b-7b-0924
An MoE model developed by allenai competitive with llama2
405 Pulls 5 Tags Updated 1 year ago
-
phi4-mini
Phi-4-mini is the latest small LLM from microsoft. These are the quants that I like to use, that weren't uploaded to the main model by the ollama team.
tools 3.8b394 Pulls 3 Tags Updated 1 year ago
-
VibeThinkertools
362 Pulls 2 Tags Updated 4 months ago
-
jamba-reasoning
A small and efficient reasoning model, with a hybrid transformer and mamba architecture
tools 3b360 Pulls 4 Tags Updated 5 months ago
-
olmo2
Allen Institute's Latest tiny LLM in the Olmo family
1b206 Pulls 6 Tags Updated 7 months ago
-
gemma3
Google's new small model
270m 1b197 Pulls 10 Tags Updated 7 months ago
-
falcon-h11.5b
189 Pulls 5 Tags Updated 5 months ago
-
gemma3n
Google's MatFormer addition to the gemma3 family
e2b186 Pulls 6 Tags Updated 6 months ago
-
deepseek-r1-qwen-distill
Useful DQ2 quants for deepseek-r1-qwen-distill:1.5b
1.5b173 Pulls 6 Tags Updated 10 months ago
-
granite-embedding-english
Embedding models for English language, made by the IBM-Granite team, in 2 (small) sizes
embedding102 Pulls 4 Tags Updated 1 year ago
-
granite3.3
Additional quants of the granite3.3:2b model
tools 2b8 Pulls 5 Tags Updated 7 months ago
-
tulu3
A model finetuned from llama3.1, developed by allenai
5 Pulls 6 Tags Updated 1 year ago
-
ministral:8b
This is the Ministral 8b model released by MistralAI. I've had some fun using it, and honestly find it to be really fast and smart for a model of this size. Hope you enjoy pulling it early through here! (ps- this is q4_K_S)