HiNow LogoHiNow

Qwen 3.6 35B A3B

Model Information

Display Name: Qwen 3.6 35B A3B

API Model ID: qwen/qwen3.6-35b-a3b

Category: Image To Text

Description: Qwen 3.6 35B A3B is Alibaba''s latest multimodal Mixture-of-Experts model, combining agentic coding, advanced reasoning, and vision understanding in a highly efficient 35B parameter model with only 3B active per token. **Key Features:** - 262K native context window (extensible to 1M with YaRN) - Multimodal: processes text, images, and video input - Mixture-of-Experts architecture (35B total, 3B active) - Thinking mode enabled by default for complex reasoning - Function calling, tool use, and structured JSON outputs - Multi-Token Prediction (MTP) for faster inference **Capabilities:** - State-of-the-art agentic coding and software engineering - Image and video understanding - Complex multi-step reasoning with chain-of-thought - Function/tool calling for agentic workflows - Code generation, debugging, and repository-level reasoning - Bilingual excellence in English and Chinese **Best For:** - Autonomous coding agents and IDE assistants - Long-document analysis and retrieval (up to 1M tokens with YaRN) - Multimodal applications with vision + reasoning - Terminal automation and DevOps workflows - Research and knowledge-intensive tasks **Technical Specs:** - Architecture: Sparse Mixture-of-Experts (8 Routed + 1 Shared out of 256 experts) - Total Parameters: 35B - Active Parameters: 3B per token - Hidden Dimension: 2,048 - Layers: 40 - Token Embedding: 248,320 (padded) - License: Apache 2.0 **Benchmarks:** - SWE-bench Verified: 73.4% - Terminal-Bench 2.0: 51.5% - MMLU-Pro: 85.2% - C-Eval: 90.0% - MMBench (vision): 92.8% - RealWorldQA (vision): 85.3%

Context Window: 262,144 tokens

Max Output: 32,768 tokens

How to Use This Model

To use Qwen 3.6 35B A3B via the HInow.ai API, use the model ID: qwen/qwen3.6-35b-a3b

API Request Example (Chat/Text)


POST https://api.hinow.ai/v1/chat/completions
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "qwen/qwen3.6-35b-a3b",
  "messages": [
    {"role": "user", "content": "Your message here"}
  ]
}

API Request Example (Image Generation)


POST https://api.hinow.ai/v1/images
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "qwen/qwen3.6-35b-a3b",
  "prompt": "Your image description here"
}

Pricing

  • input: $0.35
  • output: $1.50

Available Parameters

  • temperature: Controls randomness (0-2). Higher = more creative, lower = more deterministic. Default: 0.7 (Options: 0, 0.3, 0.5, 0.7, 1.0, 1.5, 2.0)
  • top_p: Nucleus sampling threshold (0-1). Default: 0.9 (Options: 0.1, 0.5, 0.7, 0.9, 0.95, 1.0)
  • top_k: Limits sampling to top K tokens. Default: off (Options: 10, 20, 40, 50, 100)
  • max_tokens: Maximum tokens to generate. Default: 8192 (Options: 256, 512, 1024, 2048, 4096, 8192, 16384, 32768)
  • response_format: Output format for structured responses (Options: text, json_object, json_schema)
  • tools: Function/tool definitions for agentic workflows
  • enable_thinking: Enable thinking mode for step-by-step reasoning. Default: true (Options: true, false)
  • stop: Up to 4 sequences where the API stops generating further tokens

Quick Reference

To use this model, set: "model": "qwen/qwen3.6-35b-a3b"

Featured: No

Documentation: https://hinow.ai/models/qwen/qwen3.6-35b-a3b

API Endpoint: https://api.hinow.ai/v1

Back to Models
Qwen 3.6 35B A3B

Qwen 3.6 35B A3B

qwen/qwen3.6-35b-a3b

$0.350 / $1.50
per 1M tokens (in/out)

About

Qwen 3.6 35B A3B is Alibaba''s latest multimodal Mixture-of-Experts model, combining agentic coding, advanced reasoning, and vision understanding in a highly efficient 35B parameter model with only 3B active per token.

Key Features:

  • 262K native context window (extensible to 1M with YaRN)
  • Multimodal: processes text, images, and video input
  • Mixture-of-Experts architecture (35B total, 3B active)
  • Thinking mode enabled by default for complex reasoning
  • Function calling, tool use, and structured JSON outputs
  • Multi-Token Prediction (MTP) for faster inference

Capabilities:

  • State-of-the-art agentic coding and software engineering
  • Image and video understanding
  • Complex multi-step reasoning with chain-of-thought
  • Function/tool calling for agentic workflows
  • Code generation, debugging, and repository-level reasoning
  • Bilingual excellence in English and Chinese

Best For:

  • Autonomous coding agents and IDE assistants
  • Long-document analysis and retrieval (up to 1M tokens with YaRN)
  • Multimodal applications with vision + reasoning
  • Terminal automation and DevOps workflows
  • Research and knowledge-intensive tasks

Technical Specs:

  • Architecture: Sparse Mixture-of-Experts (8 Routed + 1 Shared out of 256 experts)
  • Total Parameters: 35B
  • Active Parameters: 3B per token
  • Hidden Dimension: 2,048
  • Layers: 40
  • Token Embedding: 248,320 (padded)
  • License: Apache 2.0

Benchmarks:

  • SWE-bench Verified: 73.4%
  • Terminal-Bench 2.0: 51.5%
  • MMLU-Pro: 85.2%
  • C-Eval: 90.0%
  • MMBench (vision): 92.8%
  • RealWorldQA (vision): 85.3%

Capabilities

Image To TextText To Text
Context262K tokens
Max Output33K tokens

Parameters

temperature

Controls randomness (0-2). Higher = more creative, lower = more deterministic. Default: 0.7

00.30.50.71.01.52.0
top_p

Nucleus sampling threshold (0-1). Default: 0.9

0.10.50.70.90.951.0
top_k

Limits sampling to top K tokens. Default: off

10204050100
max_tokens

Maximum tokens to generate. Default: 8192

25651210242048409681921638432768
response_format

Output format for structured responses

textjson_objectjson_schema
tools

Function/tool definitions for agentic workflows

enable_thinking

Enable thinking mode for step-by-step reasoning. Default: true

truefalse
stop

Up to 4 sequences where the API stops generating further tokens

Code Examples

curl -X POST https://api.hinow.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $HINOW_API_KEY" \
  -d '{
    "model": "qwen/qwen3.6-35b-a3b",
    "messages": [
      {
        "role": "user",
        "content": [
          {"type": "text", "text": "Describe this image"},
          {"type": "image_url", "image_url": {"url": "https://example.com/image.jpg"}}
        ]
      }
    ],
    "parameters": {
      "temperature": "0",
      "top_p": "0.1",
      "top_k": "10",
      "max_tokens": "256",
      "response_format": "text",
      "tools": "",
      "enable_thinking": "true",
      "stop": ""
    }
  }'