Qwen3.5 27B (Non-reasoning)
Intelligence
37.2
#108/523
Coding
33.4
#108/429
Math
—
Speed
91 tok/s
TTFT: 1.42s
Pricing
$0.30 / $2.40
per 1M tokens (in/out)
Qwen3.5-27B is a multimodal dense foundation model with 27 billion parameters. It combines strong reasoning, coding, multilingual, long-context, and visual understanding performance in a production-friendly open-weight package with a native 262K context window.
Qwen3.5 27B (Non-reasoning) is Alibaba’s model designed for various NLP tasks. It processes at 90.953 tokens per second and is priced at $0.3 per million input tokens, targeting professional users.
When to Use Qwen3.5 27B (Non-reasoning)
✓ Best For
- Text generation and summarization.
- Data analysis and reporting.
- Basic coding assistance.
✗ Not Ideal For
- Complex reasoning tasks.
- High-level mathematical computations.
How Qwen3.5 27B (Non-reasoning) Compares
Intelligence Index · Higher is better
KimiCohereAlibabaAnthropic
Benchmark Profile
Coding Index
XiaomiAnthropicAlibabaInclusionAIDeepSeek
Output Speed · tok/s
OpenAITencentAlibabaMeta
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (9)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 37.2 | |
| Coding Index | 33.4 | |
| GPQA | 842% | |
| HLE | 132% | |
| SciCode | 36.7% | |
| IFBench | 46.9% | |
| LCR | 55.7% | |
| TerminalBench Hard | 31.8% | |
| Tau2 | 87.1% |
Data: Artificial Analysis · Updated: March 27, 2026
Frequently Asked Questions (15)
When was Qwen3.5 27B (Non-reasoning) released?
Qwen3.5 27B (Non-reasoning) was released on February 24, 2026.
Who created Qwen3.5 27B (Non-reasoning)?
Qwen3.5 27B (Non-reasoning) was created by Alibaba.
How intelligent is Qwen3.5 27B (Non-reasoning)?
Qwen3.5 27B (Non-reasoning) scores 37 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 12).
How fast is Qwen3.5 27B (Non-reasoning)?
Qwen3.5 27B (Non-reasoning) generates output at 86.2 tokens per second (based on Alibaba's API), which is below average compared to other open weight non-reasoning models of similar size (median: 98.0 t/s).
What is the latency of Qwen3.5 27B (Non-reasoning)?
Qwen3.5 27B (Non-reasoning) has a time to first token (TTFT) of 5.60s (based on Alibaba's API), which is at the higher end compared to other open weight non-reasoning models of similar size (median: 1.79s).
How much does Qwen3.5 27B (Non-reasoning) cost?
Qwen3.5 27B (Non-reasoning) costs $0.30 per 1M input tokens (at the higher end, median: $0.16) and $2.40 per 1M output tokens (at the higher end, median: $0.40), based on Alibaba's API.
What is Qwen3.5 27B (Non-reasoning) API pricing?
Qwen3.5 27B (Non-reasoning) costs $0.30 per 1M input tokens and $2.40 per 1M output tokens (based on Alibaba's API). For a blended rate (3:1 input to output ratio), this is $0.82 per 1M tokens. Pricing may vary by provider.
How verbose is Qwen3.5 27B (Non-reasoning)?
When evaluated on the Intelligence Index, Qwen3.5 27B (Non-reasoning) generated 25M output tokens, which is at the higher end compared to other open weight non-reasoning models of similar size (median: 5.3M).
Is Qwen3.5 27B (Non-reasoning) a reasoning model?
No, Qwen3.5 27B (Non-reasoning) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
What input modalities does Qwen3.5 27B (Non-reasoning) support?
Qwen3.5 27B (Non-reasoning) supports text and image input.
What output modalities does Qwen3.5 27B (Non-reasoning) support?
Qwen3.5 27B (Non-reasoning) supports text output.
Can Qwen3.5 27B (Non-reasoning) process images?
Yes, Qwen3.5 27B (Non-reasoning) supports image input and can analyze, describe, and answer questions about images.
Is Qwen3.5 27B (Non-reasoning) multimodal?
Yes, Qwen3.5 27B (Non-reasoning) is multimodal. It can process text and image input and generate text output.
What is the context window of Qwen3.5 27B (Non-reasoning)?
Qwen3.5 27B (Non-reasoning) has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.
Is Qwen3.5 27B (Non-reasoning) open source?
Yes, Qwen3.5 27B (Non-reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
