Qwen 3.5 Models Are Now Available in Puter.js

Puter.js now supports the full Qwen 3.5 model series from Alibaba's Qwen team—six models spanning from a 3B-active ultra-efficient Flash model to a 397B-parameter flagship that competes with GPT-5.2 and Claude Opus 4.5.

What is Qwen 3.5?

Qwen 3.5 is the latest model family from Alibaba's Qwen team. All models use a hybrid Gated DeltaNet + MoE architecture with native multimodal support, up to 1M token context, and coverage of 201 languages.

• Qwen3.5-397B-A17B: The open-weight flagship with 397B total parameters and 17B active per token. Competes with GPT-5.2 and Claude Opus 4.5 on reasoning, coding, and multimodal benchmarks.
• Qwen3.5-Plus: The hosted flagship API with a 1M token context window and built-in tool use including code interpreter. Designed for agentic workflows.
• Qwen3.5-122B-A10B: The largest medium MoE model (122B total, 10B active). Leads the lineup on tool-use and agent benchmarks like BFCL-V4 and BrowseComp.
• Qwen3.5-27B: The only dense model in the series. All 27B parameters active on every pass. Ties GPT-5 mini on SWE-bench Verified at 72.4 and runs well on consumer hardware.
• Qwen3.5-35B-A3B: A sparse MoE model activating just 3B of 35B parameters. Outperforms the previous-gen 235B flagship and runs on GPUs with as little as 8 GB VRAM.
• Qwen3.5-Flash: The production API version of 35B-A3B with 1M context and native function calling at ~$0.10/M input tokens.

Examples

General reasoning

puter.ai.chat("Explain the trade-offs between event sourcing and traditional CRUD for a banking application",
  { model: 'qwen/qwen3.5-397b-a17b' }
);

Long-context analysis

puter.ai.chat("Analyze this codebase and identify potential memory leaks:\n" + largeCodebase,
  { model: 'qwen/qwen3.5-plus-02-15' }
);

Code generation

puter.ai.chat("Write a Rust implementation of a lock-free concurrent hash map",
  { model: 'qwen/qwen3.5-27b' }
);

Streaming

const response = await puter.ai.chat(
  "Write a step-by-step guide to deploying a Next.js app on AWS Lambda",
  { model: 'qwen/qwen3.5-flash-02-23', stream: true }
);

for await (const part of response) {
  puter.print(part?.text);
}

Get Started Now

// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

Or add one script tag to your HTML:

<script src="https://js.puter.com/v2/"></script>

No API keys needed. Start building with Qwen 3.5 models immediately.

Learn more:

• Free, Unlimited Qwen API
• Puter.js AI Documentation