Toggle navigation
shortly

Alibaba

Available in Shortly AI

Qwen3 Next 80B A3B Thinking

Released September 11, 2025

ReasoningTool Use

Qwen3-Next uses a highly sparse MoE design: 80B total parameters, but only ~3B activated per inference step. Experiments show that, with global load balancing, increasing total expert parameters while keeping activated experts fixed steadily reduces training loss.Compared to Qwen3’s MoE (128 total experts, 8 routed), Qwen3-Next expands to 512 total experts, combining 10 routed experts + 1 shared expert — maximizing resource usage without hurting performance. The Qwen3-Next-80B-A3B-Thinking excels at complex reasoning tasks — outperforming higher-cost models like Qwen3-30B-A3B-Thinking-2507 and Qwen3-32B-Thinking, outpeforming the closed-source Gemini-2.5-Flash-Thinking on multiple benchmarks, and approaching the performance of our top-tier model Qwen3-235B-A22B-Thinking-2507.

Try out now
Context window
262.1K tokens
Maximum output
262.1K tokens
Input pricing
$0.15/M
Output pricing
$1.20/M

Independent benchmark

LiveBench historical snapshot

54.9overall

Historical benchmark data from LiveBench, release 2025-12-23. Scores are pinned and are not fetched on page load.

Qwen3 Next 80B A3B Thinking LiveBench category scores
CategoryScore
Reasoning58.164
Coding60.656
Agentic Coding8.333
Mathematics86.347
Data Analysis73.164
Language56.312
Instruction Following41.542

Benchmarks measure selected evaluation tasks and do not guarantee performance for every prompt or workflow.

FAQ

Similar AI models

GLM 5.3 FlashZ.aiGLM-5.3-Flash is Z.ai’s native multimodal coding model, featuring 320B total parameters, 18B activated parameters, and a 1M-token context window. Its efficient hybrid attention architecture supports visual coding, tool use, and end-to-end professional workflows across code, browsers, documents, and graphical interfaces.DeepSeek V4 Flash Vision ExpDeepSeekDeepSeek-V4-Flash-Vision-Exp is the first experimental multimodal model in the DeepSeek-V4 family. It builds on DeepSeek-V4-Flash by adding visual modules and continued training for visual understanding, with substantially improved multimodal agent capabilities while remaining comparable on text-only agent tasks.Nemotron 3.5 Lightning 30BNVIDIANVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 is a large language model (LLM) trained by NVIDIA. The model employs a hybrid Mixture-of-Experts architecture, utilizing interleaved Mamba-2 and MoE layers, along with select Attention layers. The Lightning 3.5 model is released alongside a number of speculative decoding methods for faster text generation. The model has 3B active parameters and 30B parameters in total.

Available with one subscription

Start with Qwen3 Next 80B A3B Thinking in Shortly AI

Keep your conversations, files and model choices together in one workspace.

Try Qwen3 Next 80B A3B Thinking