Alibaba's performant long context models for agentic and coding tasks.
Qwen3-Coder-Next is a coding-focused language model from Alibaba's Qwen team, optimized for agentic coding workflows and local development.
Qwen3.6 delivers substantial upgrades in agentic coding and thinking preservation than previous Qwen models.
Qwen3.8 delivers substantial gains across coding, professional work, research, and long-horizon agentic tasks.
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.
Qwen3 Coder 30B A3B Unsloth Dynamic 2.0 GGUF MoE
Qwen3 Coder 30B A3B 1M Unsloth Dynamic 2.0 GGUF MoE
Validated runtime configuration + measured results for an abliterated (uncensored) Qwen3-Coder-Next 80B-A3B running as an agentic coding backend on CPU via ik_llama.cpp
Qwen3 Coder Next 80B A3B Unsloth Dynamic 2.0 GGUF MoE
Qwen3-Coder-30B-A3B-Instruct-GGUF from: https://huggingface.co/unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF
Qwen3-Coder Next, no-thinking, 256K context length
Verified 80B MoE (Unsloth GGUF) build of Qwen3-Coder-Next. Validated for Ollama v0.15.5+ with fixed chat templates
Qwen3-Coder-uncensored
LocoOperator-4B is a 4B-parameter tool-calling agent model trained via knowledge distillation from Qwen3-Coder-Next inference traces.
From qwen3-coder:30b . Compatible with opencode (both).