Alibaba's Qwen3.6 Plus reaches 78.8 on SWE-bench with 1M context window
Alibaba released Qwen3.6 Plus on April 2, 2026, featuring a 1 million token context window at $0.50 per million input tokens and $3 per million output tokens. The model combines linear attention with sparse mixture-of-experts routing to achieve a 78.8 score on SWE-bench Verified, with significant improvements in agentic coding, front-end development, and reasoning tasks.
Qwen 3.6 Plus at a glance
How Qwen 3.6 Plus ranks now
| Benchmark | Score | Rank | Source |
|---|---|---|---|
| SWE-bench Verified | 78.8% | 18 of 80 | openrouter.ai, Apr 2026 ↗ |
Rank is against every model in our database with a sourced score on that benchmark, each counted once with its best score.
Alibaba's Qwen3.6 Plus Reaches 78.8 on SWE-bench with Million-Token Context
Alibaba released Qwen3.6 Plus on April 2, 2026, introducing a model that combines efficient linear attention with sparse mixture-of-experts routing to handle complex reasoning and coding tasks at scale.
Key Specifications
Context and Pricing:
- Context window: 1,000,000 tokens
- Input pricing: $0.50 per million tokens
- Output pricing: $3.00 per million tokens
Performance: The model achieves a 78.8 score on SWE-bench Verified, positioning it alongside leading state-of-the-art models. Alibaba claims major improvements over the 3.5 series in agentic coding, front-end development, and overall reasoning capabilities.
Architecture and Capabilities
Qwen3.6 Plus uses a hybrid architecture combining:
- Efficient linear attention mechanisms for scalability
- Sparse mixture-of-experts routing for high-performance inference
According to Alibaba, the model excels at complex tasks including 3D scene generation, game development, and repository-level problem solving. The company describes particular improvements in "vibe coding experience," though this term lacks precise technical definition.
The model claims substantial performance gains in both pure-text and multimodal tasks, though specific benchmark comparisons to previous versions remain undisclosed.
Data Collection Notice
Alibaba explicitly states that Qwen3.6 Plus collects prompt and completion data for model improvement purposes. Users should review privacy implications before deploying in sensitive applications.
Deployment Availability
Qwen3.6 Plus is available through OpenRouter, which routes requests across multiple providers to optimize for context window support and uptime. The platform provides normalized API access across providers and supports reasoning-enabled inference with step-by-step thinking visibility.
What This Means
Qwen3.6 Plus positions Alibaba's Qwen line as a competitive option in the 1M-context segment, matching context window sizes offered by Claude 3.5 and other leaders. The 78.8 SWE-bench score places it in the tier of capable coding models, though detailed comparisons to other 1M-context models remain unavailable. Pricing at $0.50/$3.00 per million tokens is competitive with other high-context models. The explicit data collection policy requires careful consideration for enterprises handling proprietary code or sensitive information.
Related Articles
Paradigm Intelligence launches Phai-A1 language model and Phai-M1 video model, trained on Huawei Ascend
Paradigm Intelligence (范式智能, formerly 第四范式) has unveiled two in-house models: Phai-A1, a language model that runs on a single RTX 3090, and Phai-M1, a video model with synchronized audio. The company says both were trained on Huawei Ascend hardware, with open weights coming soon.
Cloudflare releases open-weight Clairic 27B and new Clairic-flash 9B for data extraction
Cloudflare announced on November 18, 2026 an updated 27-billion-parameter Clairic and a new 9-billion-parameter Clairic-flash. Both have open weights on Hugging Face and are available as hosted Workers AI endpoints.
OpenAI releases GPT-Live-1.1 in the API, with standby mode billing idle WebRTC time at $0.02 per minute
OpenAI has released GPT-Live-1.1 in the API. The full-duplex voice model improves multilingual accents and transcription, and a new standby mode bills idle WebRTC time at $0.02 per minute instead of $0.05.
Baseten releases open-weight bison-1 and groundhog-1 decision models; bison-1 scores 76 on AA Decision Index v2.1
Baseten released bison-1 and groundhog-1, two open-weight decision models built on Gemma 4. Bison-1 scores 76 on the AA Decision Index v2.1, and both models support 128k context and image input.
Comments
Loading...