Qwen3.6-Plus

Released 2026-04 · hybrid · 1.0M tokens · 3 benchmarks · Open weight

Editorial notes

Lanzado 2 abril 2026. Flagship multimodal con 1M contexto. Linear attention + sparse MoE, chain-of-thought siempre activa, function calling nativo. SUPERA Claude Opus 4.5 en Terminal-Bench 2.0 (61.6 vs 59.3). #1 OmniDocBench (91.2). Pricing ~$0.29/M input.

Spec sheet

Company: Alibaba
Country: CN
Type: hybrid
Release: 2026-04
Context: 1.0M tokens
License: qwen-license
Quants: BF16, Q8_0, Q5_K_M, Q4_K_M
Pricing (alibaba): $0.29/$1/M
Slug: qwen3-6-plus

Quick install

2 tools

ollama.com

ollama run qwen3.6:plus

docs.vllm.ai

vllm serve Qwen/Qwen3.6-Plus --max-model-len 1000000

Benchmarks (3)

Coding 2

Reasoning 1

MMMU

Multimodal Multidiscipline Understanding — multimodal reasoning over academic im

86.0

Cite this model

BibTeX · APA

BibTeX

@misc{frontier-qwen3-6-plus,
  title  = {Qwen3.6-Plus},
  author = {{Alibaba}},
  year   = {2026},
  note   = {Frontier Benchmarks AI atlas. Accessed 2026-05-08},
  url    = {https://frontierbenchmarks.com/models/qwen3-6-plus}
}

APA

Alibaba (2026). Qwen3.6-Plus [Large language model]. Frontier Benchmarks AI. Retrieved 2026-05-08, from https://frontierbenchmarks.com/models/qwen3-6-plus

Citation reflects the atlas page, not the original model paper. For the paper, see the "Resources" section above.

⚔️ Battle vs another model ← All models More from Alibaba Methodology