Skip to content

Qwen3.6 Plus

AlibabaProprietaryReasoningVerifiedReleased Apr 2, 2026Input: text, image, video
Compare with Qwen3.7 PlusSource scores ↗OpenRouter ↗
Overall score
53.8
Rank 80 of 222 · 44.2–63.4
Context window
1M
1,000,000 tokens
Price per 1M tokens
$0.33 in · $1.95 out
Source: OpenRouter
Blended price
$0.73
3:1 input:output tokens
Output speed
—
Not measured yet

Category scores

Scores per capability area with the model's rank among ranked models in that category.

  1. 1Multilingual#493.8
  2. 2Instruction following#4785.8
  3. 3Multimodal#2176.1
  4. 4Reasoning73.9
  5. 5Math#669.5
  6. 6Knowledge#6863.3
  7. 7Agentic & tool use#8559.3
  8. 8Coding#6757.4
Higher is better. #N is the model's rank in that category, where it is eligible to be ranked.

Benchmark results (43)

As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.

Agentic & tool use

Gert Labs50.60#21 of 52
MCP Atlas48.2#36 of 42
Terminal-Bench 2.061.6#16 of 39
Claw-Eval58.8#24 of 36
τ³-bench results70.7#5 of 23
Toolathlon39.8#17 of 22
ResearchClawBench18.0#8 of 19
WideResearch74.3#9 of 16
VITA-Bench44.3#3 of 11
QwenClawBench57.2#4 of 10
DeepPlanning41.5#2 of 7
MCP-Tasks74.1#2 of 5

Multimodal

Design Arena Website1243#48 of 91
MMMU-Pro78.8#17 of 48
CharXiv81.5#21 of 42
MMMU86.0#3 of 21
MathVision88.0#9 of 20
ScreenSpot Pro68.2#12 of 15
VideoMMMU84.0#11 of 12
V*96.9#1 of 12

Knowledge

GPQA90.4#23 of 117
HLE28.8#46 of 69
MMLU-Pro88.5#3 of 67
SuperGPQA71.6#5 of 20
MMLU-Redux94.5#4 of 15
C-Eval93.3#2 of 10

Math

FrontierMath v2 (Tiers 1-3)38.246#46 of 64
AIME2695.3#11 of 29
HMMT Feb 202687.8#10 of 24
HMMT Feb 202596.7#3 of 13
HMMT Nov 202594.6#2 of 12
MMAnswerBench83.8#4 of 10

Coding

SWE-bench Verified78.8#23 of 98
SWE-bench Pro56.6#40 of 69
LiveCodeBench v687.1#13 of 46
NL2Repo37.9#25 of 31

Multilingual

SWE Multilingual73.8#26 of 46
MMLU-ProX84.7#4 of 16
NOVA-6357.9#4 of 7

Reasoning

LongBench v262.0#5 of 17
AI-Needle68.3#3 of 4

Instruction following

IFEval94.3#5 of 53
IFBench75.8#22 of 42

Pricing by source

USD per 1M tokens.

SourceInputOutputCached in
List price (OpenRouter)$0.33$1.95—
OpenRouter qwen/qwen3.6-plus$0.33$1.95—

Alibaba's public Model Studio page lists Qwen3.6 Plus with a public range of $0.50-$2.00 input and $3.00-$6.00 output per million tokens, but does not expose an exact per-tier table for the exact `qwen3.6-plus` SKU.

Arena ratings

Human-preference ratings from LMArena. Best-rated variant shown.

Text1436#84 · 48.2K votes
WebDev (coding)1462#73 · 18.6K votes

Speed

Measured by us through OpenRouter (median of trials).

Not measured yet.

Related models

Similar score