Qwen3.6 Plus

Qwen3.6 Plus is presented as a scalable hybrid model combining linear attention with sparse mixture-of-experts routing. The available description positions it for challenging software-development work, including agentic coding, front-end creation, 3D scenes, games, and repository-level problem solving.

Its practical emphasis is on stronger reasoning and multimodal handling than the preceding 3.5 series, along with an improved coding workflow. A reported 78.8 score on SWE-bench Verified and third-party coding, long-context reasoning, and instruction-following results suggest a strong fit for complex development and analytical tasks, though these figures come from third-party listings rather than primary technical documentation.

Ofoxbailian/qwen3.6-plusqwen

Open provider page Provider docs

Quick Info

Provider: Ofox
Model key: bailian/qwen3.6-plus
Release date: Apr 2, 2026
Last updated: Apr 2, 2026
Knowledge cutoff: 2025-04
Input modalities
Output modalities
Capabilities

Cost

Input token cost: $0.50
Output token cost: $3.00

Limits

Output tokens: 65,536 tokens
Context window: 1,000,000 tokens

Transparent token rates

Compare Qwen3.6 Plus pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about Qwen3.6 Plus

Ofox

Coverage

Alibaba officially launched Qwen3.6-Plus via its Model Studio API, positioning it as a major capability upgrade over the Qwen3.5 series released earlier in February. According to the Alibaba Cloud Community blog, the release delivers drastically enhanced agentic coding capabilities, spanning frontend web development th The announcement highlights comprehensive evaluation results showing Qwen3.6-Plus achieving top results in coding agents, long-horizon planning, and tool-calling benchmarks, while setting new records in STEM reasoning, ultra-long-context information extraction, and multilingual tasks. Alibaba emphasizes the model's dee

Ofox

Coverage

Tongyi Lab officially released Qwen3.6-Plus on 2 April 2026, after launching the Qwen3.5 series in February, targeting the core pain point of unstable task execution in agent programming. The model is available through the Aliyun BaiLian (Bailian) API and emphasizes deep logical reasoning, massive memory, and precise e Qwen3.6-Plus defaults to a 1 million character context window, improving long document parsing and multi-turn dialogue accuracy, and achieves strong cost-performance versus competing models. It integrates with developer tools including OpenClaw (formerly Moltbot) for terminal agent coding, Qwen Code for codebase unders