qwen/qwen3.6-plus-2026-04-02

qwen/qwen3.6-plus-2026-04-02
VisionToolsJSONReasoning
by qwen

Qwen3.6 Plus snapshot 2026-04-02 — frozen version of qwen3.6-plus, same capabilities.

ctx1.05M tokens
Max output65.5K
Inputtext + image + video
Outputtext
p50 TTFT3.26 s
INPUT$0.50/ 1M tokens
OUTPUT$3.00/ 1M tokens
p50 TTFT3.26 s7d
p95 TTFT10.00 s7d
TRAFFIC4.6Mtokens / 7d

qwen/qwen3.6-plus-2026-04-02 is a specific checkpoint of the Qwen3.6 Plus series, hosted by Qwen and accessible through OrcaRouter's OpenAI-compatible API. It accepts text, image, and video inputs,…

What is qwen/qwen3.6-plus-2026-04-02?

Who should use this model and why?

What input modalities does it support and what are the context limits?

Code samples

Call from any SDK

OpenAI-compatible — keep the SDK you already use

  • OpenAI SDKhttps://api.orcarouter.ai/v1
from openai import OpenAI

client = OpenAI(
    base_url="https://api.orcarouter.ai/v1",
    api_key="$ORCAROUTER_API_KEY",
)

response = client.chat.completions.create(
    model="qwen/qwen3.6-plus-2026-04-02",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Supported parameters

  • enable_search
  • enable_thinking
  • include_reasoning
  • logprobs
  • max_tokens
  • n
  • parallel_tool_calls
  • presence_penalty
  • reasoning
  • repetition_penalty
  • response_format
  • seed
  • stop
  • stream
  • stream_options
  • temperature
  • thinking_budget
  • tool_choice
  • tools
  • top_k
  • top_logprobs
  • top_p

Pricing

TierInput / 1M tokensOutput / 1M tokens
256K$0.500$3.00
1.0M$2.00$6.00
Tier selected by input token count of each request

Cost calculator

Tokens / month10MM
Input share70%%
Estimated / month $12.50

Estimate based on list price

Tiered pricing — this estimate uses base-tier rates.

Token & cost estimator

Input tokens: 20Cost per request: $0.001510

Estimate only — actual token counts depend on the provider's tokenizer.

Performance

p50 TTFT
3.26 s
Output speed
63.0 tok/s
p95 TTFT
10.00 s
Error rate
3.0%

Public benchmarks

54.5
AA Coding
Better than 67% of models compared
#40 of 123
39.6
AA Intelligence
Better than 64% of models compared
#44 of 125
GPQA Diamond
88.2
Humanity's Last Exam
25.7
IFBench
75.2
Long-Context Recall
69.7
SciCode
40.7
tau_banking
16.5
TerminalBench Hard
43.9
terminalbench_v2_1
61.4
τ²-Bench
97.7
Source: artificialanalysis.ai

How it compares

qwen/qwen3.6-plus-2026-04-02qwen/qwen3-max-previewQwen3.5 397B A17Bqwen/qwen3.5-plus
Input $/M$0.50$0.86$0.17$0.12
Output $/M$3.00$3.44$1.03$0.69
Context1.0M262K33K1.0M
Quality8/108/108/108/10
Compare side-by-sideCompare side-by-sideCompare side-by-sideCompare side-by-side

FAQ

How much does it cost to use qwen/qwen3.6-plus-2026-04-02 on OrcaRouter?
Pricing details are not provided in the given facts. Costs are determined by OrcaRouter based on input and output token counts, with potential additional charges for image and video processing (per image or frame). Consult OrcaRouter's official pricing page for current rates. There is no mention of free credits or trial access for this model.
What is the context window size and max output?
The context window is 1,048,576 tokens (input + output tokens). The maximum output per generation is 65,536 tokens. Both figures are from the provided facts.
What are the model's main strengths?
Based on given facts, the main strengths are: a very large context window (1M tokens), support for image and video inputs, a high τ²-Bench score of 97.7 indicating strong agentic task performance, and a large maximum output (65K tokens). It is a recent checkpoint (April 2, 2026).
How does this model compare to other large language models?
Compared to models like GPT-4o or Claude 3.5 Sonnet, this model offers a significantly larger context window (1M tokens) and native video support. Its τ²-Bench score of 97.7 suggests strong agentic reasoning. Direct comparisons on other benchmarks are not provided. For text-only short tasks, smaller models may be more cost-effective.
What data handling and privacy policies apply?
Data handling is governed by OrcaRouter's privacy policy and terms of service, as well as Qwen's (Alibaba Cloud) data processing agreements. No specific data retention or training data details are provided in the given facts. Users should review OrcaRouter's privacy documentation and ensure compliance with their own data protection requirements.
How do I call this model via an OpenAI-compatible API?
Use OrcaRouter's base URL https://api.orcarouter.ai/v1 with the model identifier "qwen/qwen3.6-plus-2026-04-02". The API follows the OpenAI Chat Completions format. Set your authentication token from OrcaRouter. For multimodal inputs, include image or video data in the messages array using the appropriate content types (e.g., `image_url`). Refer to OrcaRouter's API documentation for details.
Can I use this model with streaming responses?
Yes, OrcaRouter's API supports streaming by setting the `stream` parameter to true in your request. This returns tokens as they are generated, reducing perceived latency. The model's maximum output of 65,536 tokens can be streamed incrementally.
Is there a cheaper way to use long contexts with this model?
To reduce cost, you can truncate parts of the input that are not relevant, use lower resolution images or fewer video frames, and set a low `max_tokens` limit. OrcaRouter may offer prompt caching for repeated prefixes (check documentation for this model). Alternatively, if you need long context but not multimodal capability, consider a cheaper text-only long-context model on OrcaRouter.
What types of tasks is this model best suited for?
It is best suited for tasks requiring long-context understanding combined with visual inputs, such as analyzing lengthy documents with diagrams, summarizing video content, and building autonomous agents that maintain state over many steps. The 97.7 τ²-Bench score highlights its capability for agentic workflows.
Does this model have any known limitations?
It may hallucinate, produce biased outputs, and struggle with precise math without external tools. Extremely long contexts (close to 1M tokens) may cause loss of attention to early content. Video understanding depends on frame selection and may not capture real-time dynamics. Latency and cost are higher than smaller models.

Embed this badge

qwen/qwen3.6-plus-2026-04-02$0.50/M in3258ms p50via OrcaRouter
HTML <a href="https://www.orcarouter.ai/models/qwen/qwen3.6-plus-2026-04-02" target="_blank"> <img src="https://www.orcarouter.ai/embed/qwen/qwen3.6-plus-2026-04-02.svg" alt="qwen/qwen3.6-plus-2026-04-02 on OrcaRouter" /> </a>
Markdown [![qwen/qwen3.6-plus-2026-04-02](https://www.orcarouter.ai/embed/qwen/qwen3.6-plus-2026-04-02.svg)](https://www.orcarouter.ai/models/qwen/qwen3.6-plus-2026-04-02)

Model card as data

GET /api/public/models/qwen/qwen3.6-plus-2026-04-02Open
Machine-readable:/llms.txt/llms-full.txt