Models
Updated Apr 2026
A curated selection of open models — every one runs in llama.app.
Qwen 3.6 27B35B-A3B
Alibaba's next-gen natively multimodal reasoning models. Dense and MoE variants that rival models many times their size on coding and vision tasks.
Gemma 4 E2BE4B26B-A4B31B
Google's most capable open models, built from Gemini 3 technology. Supports multimodal reasoning, agentic workflows, and 140+ languages.
GLM 4.7 Flash
Zhipu AI's agentic reasoning and coding models. Built for software engineering, browser automation, and multi-turn tool use.
Qwen 3.5 Small 0.8B
Alibaba's compact natively multimodal reasoning models. Supports thinking/non-thinking modes for text and vision tasks across 201 languages.
Devstral 2 24B123B
Mistral AI's agentic coding models for software engineering tasks. Excels at exploring codebases, multi-file editing, and powering code agents.
Ministral 3 3B3B Reasoning8B8B Reasoning14B14B Reasoning
Mistral AI's compact edge models with vision capabilities. Offers best cost-to-performance ratio for on-device deployment in 3B, 8B, 14B sizes.
GPT-OSS 20B120B
OpenAI's first open-weight models since GPT-2. Built for reasoning, agentic tasks, and developer use with function calling and tool use capabilities.
Gemma 3 270M1B4B12B27B
Google's multimodal models built from Gemini technology. Supports 140+ languages, vision, and text tasks with up to 128K context for edge to cloud deployment.
Find thousands more GGUF models on Hugging Face ↗