LLMAnonymized

Qwen 3.6 Plus Uncensored

Alibaba's Qwen 3.6 Plus Uncensored — a reasoning-optimized, multimodal model with 1M context, now uncensored and private on Venice.

Maker
Alibaba
Modality
Text + vision
Context
1,000K tokens
Open weights
Proprietary

Overview

What is Qwen 3.6 Plus Uncensored

Qwen 3.6 Plus Uncensored is Alibaba's enhanced reasoning model, released in April 2026 with a 1M-token context window, native vision, function calling, and web search. Hosted uncensored on Venice, it enables private, agentic workflows without storing prompts.

Running it privately on Venice

On Venice, Qwen 3.6 Plus runs with anonymized privacy—your prompts are never stored or profiled. This uncensored deployment unlocks its full agentic coding, vision, and tool use capabilities without surveillance, ideal for developers needing sovereignty over sensitive workflows.

AnonymizedNo prompt trainingTEE · hardware enclaveEnd-to-end encrypted

Assessment

Strengths and limitations

Strengths
  • State-of-the-art agentic coding with repository-level reasoning and front-end development fluency.
  • True multimodal input support: processes images, video, and text for unified reasoning.
  • 1M-token context window by default, enabling long-form analysis and complex document processing.
  • Native function calling, web search, and structured JSON output for autonomous agent workflows.
  • Improved stability and reduced overthinking compared to Qwen3.5-Plus, with better production reliability.
Limitations
  • Not open weights: cannot be self-hosted or audited locally.
  • Multimodal reasoning still lags behind Gemini 3 Pro in certain benchmarks.
  • Security-focused coding lags safety-tuned rivals: 43.3% hidden-test task success on BridgeBench's third-party Security Bench (GPT-5.4 Mini and Claude Sonnet 4.5 score ~87).

Capabilities

What it supports

  • Tool use / function calling
  • Vision (image input)
  • Reasoning
  • Web search
  • Code-optimized
  • Structured output (JSON schema)
  • Audio input
  • Video input
  • Multiple image inputs
  • Log probabilities

Specifications

Datasheet

Maker
Alibaba
Released
April 2026
Architecture
Hybrid autoregressive with chain-of-thought integration
Parameters
Not disclosed
Open/closed
Closed
Context window
1,000K tokens
Max output
65.536K tokens
Capabilities
Vision, Function calling, Reasoning, Web search, Code-optimized
Privacy on Venice
Anonymized — prompts not stored
Available on Venice since
Apr 2026
License
Proprietary

API

Call it from your code

Venice exposes an OpenAI-compatible API. Point your base URL at Venice and pass the model id.

curl https://api.venice.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $VENICE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen-3-6-plus",
    "messages": [{ "role": "user", "content": "Explain quantum tunneling simply." }]
  }'

Pricing

What it costs on Venice

Billed per token on Venice: $0.63 per 1M input tokens and $3.75 per 1M output tokens.

Input / 1M tokens
$0.63
Per 1M tokens
Output / 1M tokens
$3.75
Per 1M tokens
Cached input / 1M
$0.06
Per 1M tokens

New Venice accounts include a free daily allowance and 500 welcome credits — no credit card required.

Alternatives

How it compares

ModelMax contextInput cost / 1MOpen weightsStrongest at
Qwen 3.6 Plus Uncensored1M tokens$0.63 in · $3.75 out / 1MNoAgentic coding, multimodal reasoning
Claude Fable 51M tokens$12 in · $60 out / 1MNoCreative storytelling, long-form narrative
DeepSeek V4 Flash 07311M tokens$0.17 in · $0.35 out / 1MNoSpeed, low-cost inference
GLM 5.1200K tokens$1.10 in · $4.15 out / 1MYesOpen-weight flexibility, Chinese NLP

Alibaba's flagship reasoning model with full tool integration and uncensored access on Venice.

Use cases

What it is good for

  1. 01Autonomous coding agents that navigate and modify entire codebases.
  2. 02Multimodal document analysis combining text, tables, and embedded images.
  3. 03Private research workflows requiring web search and citation without data retention.
  4. 04Enterprise automation using function calling and structured JSON output.
  5. 05Vision-augmented reasoning tasks like OCR-heavy form processing or UI interpretation.

Prompting

Getting better results

Use explicit step-by-step reasoning prompts — the model excels at chain-of-thought workflows.

Include image references directly when using vision input for better context grounding.

Leverage web search capability by asking for up-to-date information explicitly.

For agent tasks, define clear goals and constraints to reduce overthinking.

Use JSON schema output for reliable data extraction and integration with downstream tools.

Version history

Qwen3.5-Plus
2026-02

Predecessor with smaller context and less stable agentic coding

Qwen 3.6 Plus Uncensored
2026-04

Current version — enhanced reasoning, 1M context, uncensored

FAQ

Frequently asked questions

Qwen 3.6 Plus Uncensored is Alibaba's advanced reasoning model released in April 2026, hosted on Venice with full privacy and no content filtering. It supports 1M-token context, vision, function calling, and web search for autonomous agent workflows.

On Venice, Qwen 3.6 Plus costs $0.63 per 1M input tokens and $3.75 per 1M output tokens. Cached input is available at $0.06 per 1M tokens, reducing costs for repeated access to long contexts.

No, Qwen 3.6 Plus is not open source or open weights. It is a proprietary model developed by Alibaba. However, it is available uncensored on Venice with full privacy protections.

Yes, Qwen 3.6 Plus supports native function calling and tool use, including web search, code execution, and structured JSON output, making it well-suited for autonomous agent applications.

Qwen 3.6 Plus has a 1,000,000-token context window by default, enabling analysis of extremely long documents, full code repositories, and complex multimodal inputs.

Yes, Qwen 3.6 Plus supports native vision input and can process multiple images along with text and video for multimodal reasoning tasks.

Qwen 3.6 Plus excels in agentic coding and multimodal reasoning at a fraction of the cost, while Claude Fable 5 is optimized for creative narrative generation. For technical and autonomous workflows, Qwen offers superior value and capability.

Yes, on Venice, Qwen 3.6 Plus runs with anonymized privacy — your prompts are not stored, profiled, or used for training, ensuring full confidentiality for sensitive or proprietary tasks.

Run Qwen 3.6 Plus Uncensored privately

No prompt logging. No data used for training.