Trying to help at https://discord.gg/ollama
-
LightOnOCR-2
Best OCR model. LightOnOCR-2-1B is our flagship OCR model, refined with RLVR training for maximum accuracy. We recommend this variant for most OCR tasks.
vision tools 1b16.3K Pulls 2 Tags Updated 9 months ago
-
hy-mt2
HY-MT2 1.8B, 7B, and 30B-A3B (MoE) multilingual translation models designed for complex real-world scenarios, it supports translation among 33 languages and effectively follows translation instructions.
1.8b 7b 30b6,937 Pulls 7 Tags Updated 5 months ago
-
fara
Fara-7B is Microsoft's first agentic small language model (SLM) designed specifically for computer use. (Template supports tools)
vision tools 7b5,702 Pulls 4 Tags Updated 11 months ago
-
mimo-v2.6
MiMo-V2.6-Distill-9B (Qwen3.5-9B SFT): coding, agent tasks, visual coding, cybersecurity. Thinking + instruct quants, multimodal.
vision 9b4,205 Pulls 9 Tags Updated 1 month ago
-
spark-x2.5-heretic
Spark-X2.5-4B abliterated (uncensored) - efficient AI model for chat, writing, coding, reasoning, tools, and agents, 1M-token context, 200+ languages, with refusal behavior removed.
4b4,083 Pulls 5 Tags Updated 1 month ago
-
mimo-v2.6-heretic
Uncensored/Abliterated MiMo-V2.6-Distill-Qwen-9B: coding, agent tasks, visual coding, cybersecurity. Thinking+instruct variants.
vision 9b3,845 Pulls 9 Tags Updated 1 month ago
-
Qianfan-OCR
Qianfan-OCR is a 4B VL model that replaces multi-step OCR pipelines with a single system, enabling direct image-to-Markdown conversion and advanced document understanding.
vision tools thinking 4b2,677 Pulls 3 Tags Updated 8 months ago
-
ling-3.0-tiny
Ling-3.0-tiny is a lightweight 7.9B-parameter hybrid reasoning MoE with just 1.3B active parameters per token, offering strong reasoning and agentic capabilities at low inference cost.
8b2,480 Pulls 10 Tags Updated 2 months ago
-
lfm2
LFM2-8B-A1B is the best on-device MoE in terms of both quality (comparable to 3-4B dense models) and speed (faster than Qwen3-1.7B). Updated LFM2.5-8B-A1B is now available!
tools 8b2,293 Pulls 4 Tags Updated 8 months ago
-
minicpm5
MiniCPM5-2B is a dense 2B Transformer for on-device deployment. It achieves 2B-class SOTA, remains competitive with 4B models overall, and excels in coding, math, long-context, tool use, and agentic tasks.
2b2,292 Pulls 4 Tags Updated 2 months ago
-
lfm2.5
LFM2.5-350M & LFM2.5-8B-A1B are fast, memory-efficient hybrid MoE language models. Unsloth Dynamic (UD) quants available.
tools thinking 350m 8b2,109 Pulls 15 Tags Updated 4 months ago
-
minicpm-v4.7-heretic
MiniCPM-V4.7 1B (Abliterated/Uncensored) - excellent multimodal vision understanding, reasoning with a low 0.0122 KL divergence.
vision 1b1,727 Pulls 5 Tags Updated 1 month ago
-
strand-rust-coder
The model fine-tunes Qwen2.5-Coder-14B for Rust-specific programming tasks using a 191K-example synthetic dataset built via multi-model generation and peer-reviewed validation.
tools 14b1,539 Pulls 3 Tags Updated 8 months ago
-
mai-ui
MAI-UI is a family of large-scale GUI foundation agents (2B,8B) built for real-world, interactive human-computer control across desktop and mobile Uls. (Template supports tools)
vision tools 2b 8b1,525 Pulls 8 Tags Updated 10 months ago
-
spark-x2.5
Spark-X2.5-4B is an efficient, open-source AI model built for chat, writing, coding, reasoning, tools, and agents, supporting 1M-token context and 200+ languages.
4b1,442 Pulls 4 Tags Updated 1 month ago
-
minicpm5-heretic
Uncensored/Abliterated MiniCPM5-2B with only 10/100 refusals. It achieves 2B-class SOTA, remains competitive with 4B models overall.
2b1,335 Pulls 5 Tags Updated 1 month ago
-
AgentCPM-Explore
AgentCPM-Explore is a small, open‑source deep research model designed for long‑horizon, complex reasoning tasks on devices. (Template supports tools)
tools thinking 4b949 Pulls 1 Tag Updated 10 months ago
-
NuMarkdown-Thinking
It is the first reasoning OCR VLM. It is specifically trained to convert documents into clean Markdown files, well suited for RAG applications.
vision thinking 8b866 Pulls 2 Tags Updated 9 months ago
-
LightOnOCR-3
Best OCR model. LightOnOCR-3 is our flagship OCR model family 4B, 1B, and 0.8B with layout, grounding, and visual understanding. We recommend the 4B for most OCR tasks.
vision tools thinking 0.8b 1b 4b789 Pulls 12 Tags Updated 1 month ago
-
spark-x2.6
Spark-X2.6-4B keeps X2.5 architecture and gains from post-training: coding, tool use, long-context work, multilingual tasks, reasoning, supporting 1M-token context and 200+ languages.
1.7b 4b787 Pulls 10 Tags Updated 2 weeks ago
-
qwen3.5-3d-coder
3D focused Qwen3.5 4B fine tune by ccore. Generates self contained Three.js HTML scenes from natural language prompts: voxel architecture, gardens, mechanical objects, animated structures. General coding and tool use. Vision, MTP.
vision tools thinking 4b779 Pulls 7 Tags Updated 2 weeks ago
-
nemotron-3.5-embed
Nemotron-3.5-Embed 1B and 8B: multilingual text + image embedding models for retrieval and RAG. 32,768 tokens context, 34 languages, bundled vision projector.
vision embedding 1b 8b622 Pulls 8 Tags Updated 3 weeks ago
-
manim-coder
Fine-tuned manim model using Qwen2.5-Coder-14B-Instruct trained on 3blue1brown-manim dataset with 2,407 examples.
tools 14b599 Pulls 3 Tags Updated 9 months ago
-
sweep-next-edit-1.5B
Sweep Next-Edit a 1.5B parameter model for next-edit autocomplete, quantized to Q8_0 GGUF format. (Template supports tools)
tools501 Pulls 1 Tag Updated 9 months ago
-
sage-a1.5
Sage‑A1.5 is a Qwen3.6-35B-A3B MoE agentic model based on Qwen3.6, built to advance long-horizon scientific problem solving. IQ3_M, Q4_K_M. MTP GGUF.
vision tools thinking 35b154 Pulls 4 Tags Updated 2 weeks ago
-
context-1
Context-1 is a 20B agentic search model that breaks complex queries into steps, retrieves relevant info iteratively, and manages its context to support multi-hop reasoning alongside a larger model.
tools thinking 20b138 Pulls 3 Tags Updated 7 months ago
-
neohorse-1
NeoHorse-1-9B is a 9B causal language model and an initial prototype on the path toward recursive self-improvement (RSI).
9b131 Pulls 4 Tags Updated 1 month ago