BGE-M3 is a new model from BAAI distinguished for its versatility in Multi-Functionality, Multi-Linguality, and Multi-Granularity.
6.5M Pulls 3 Tags Updated 2 years ago
A new collection of open translation models built on Gemma 3, helping people communicate across 55 languages.
2.3M Pulls 13 Tags Updated 7 months ago
EXAONE 3.5 is a collection of instruction-tuned bilingual (English and Korean) generative models ranging from 2.4B to 32B parameters, developed and released by LG AI Research.
557.4K Pulls 13 Tags Updated 1 year ago
Agentic coding model for 24 GB GPUs: TeichAI's Gemma-4-31B Fable-5 agent distill (vision + thinking + tools) with a disciplined coding-agent system prompt baked in. Inspect → reproduce → smallest fix → re-verify.
162 Pulls 1 Tag Updated 1 month ago
Gemma 3-270M abliterated small size and redefined AI Model for faster tasks related to general purpose, cybersecurity, coding and research.
226 Pulls 2 Tags Updated 3 weeks ago
Gemma-3-270M small size and redefined AI Model for faster tasks related to general purpose, cybersecurity, coding and research.
59 Pulls 2 Tags Updated 3 weeks ago
49 Pulls 2 Tags Updated 3 weeks ago
10.6K Pulls 4 Tags Updated 5 months ago
Gemma 4 26B MoE (Google DeepMind) with thinking mode enabled. Mixture-of-Experts — 25.2B total / 3.8B active parameters, 256K context. Supports text and image input. Knowledge cutoff: January 2025.
2,783 Pulls 1 Tag Updated 5 months ago
Gemma-3-1b redefined AI Model for faster tasks related to general purpose, cybersecurity, coding and research.
23 Pulls 2 Tags Updated 3 weeks ago
Gemma 4 26B MoE (Google DeepMind) with thinking mode disabled. Mixture-of-Experts — 25.2B total / 3.8B active parameters, 256K context. Supports text and image input. Knowledge cutoff: January 2025.
1,074 Pulls 1 Tag Updated 5 months ago
Gemma 4 31B (Google DeepMind) with thinking mode enabled. Best for complex reasoning, math, coding, and multi-step analysis. Knowledge cutoff: January 2025. Sampling: temperature 1.0 / top_p 0.95 / top_k 64.
653 Pulls 1 Tag Updated 5 months ago
Gemma 4 31B (Google DeepMind) with thinking mode disabled. Best for quick questions, chat, and straightforward tasks. Knowledge cutoff: January 2025. Sampling: temperature 1.0 / top_p 0.95 / top_k 64.
583 Pulls 1 Tag Updated 5 months ago
Gemma 4 31B multimodal instruct model with vision support, quantized to Q3_K_S and optimized for 16 GB VRAM. Supports a practical context window of approximately 36K–40K tokens, depending on the Ollama version, GPU, backend and runtime configuration.
296 Pulls 1 Tag Updated 3 weeks ago
Gemma 4 31B text-only instruct model with tool-calling and thinking support, quantized to Q3_K_S and optimized for 16 GB VRAM. Supports up to 256K context, with approximately 64K tokens fitting fully within 16 GB VRAM in tested configurations.
85 Pulls 1 Tag Updated 1 month ago
sobre la base de gemma 3 13b, un modelo entrenado con mi lore personal del universo ark1, una novela narrativa. novela narrativa, juego interactivo, juego de rol, umbral, ritual, filosofia.
15 Pulls 1 Tag Updated 7 months ago
Google Gemma 3n | Edge AI with tool support. Designed for consumer devices: efficient, local, tool-enabled
1,305 Pulls 1 Tag Updated 1 year ago
Google Gemma 3n | Edge AI with tool support Designed for consumer devices: efficient, local, tool-enabled
192 Pulls 1 Tag Updated 1 year ago
1 Pull 1 Tag Updated 1 year ago
Jarvis Voice’s pinned EmbeddingGemma 300M BF16 artifact: 768-dimensional multilingual embeddings using one immutable model across cloud and local Ollama deployments.
47 Pulls 1 Tag Updated 3 weeks ago