SShogo Bench

Qwen3.5 Plus

Qwen vision-language model for visual reasoning, documents, and agent tasks

Shogo Agent Index

Estimated#65 of 92
32.7range 21.5–44 · 11% of weight measured
Agentic coding
36
estimated
Long-horizon autonomy
37
30% measured
Knowledge-work agents
28
estimated
Computer use
23
estimated
Security & context learning
37
50% measured

Price

Input
$0.40/M
Output
$2.40/M
Cache read
$0.040/M
3:1 blend
$0.90/M
Agent-effective
$0.20/M

Cheapest host for agents: Alibaba at $0.37/M.

Agent-effective = 20 input tokens per output token, 85% cache hits. Why.

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 69
Terminal-Bench 2.0est.8%predicted 42
DeepSWEest.6%predicted 29
FrontierSWEest.5%predicted 6
CursorBenchest.4%predicted 29
GSO-Benchest.4%predicted 11
METR Time Horizonsest.10%predicted 67
Vending-Bench 2$16%source
PostTrainBenchest.4%predicted 19
APEX-Agentsest.8%predicted 28
GDPvalest.7%predicted 34
Remote Labor Indexest.5%predicted 1
DeepResearch Benchest.5%predicted 45
OSWorld 2.0est.6%predicted 3
OSWorldest.4%predicted 52
Cybenchest.5%predicted 53
CL-bench19.8%5%source

Other benchmarks

BenchmarkCategoryScoreSource
Chess PuzzlesReasoning9.0%source
DTBenchKnowledge80.5%source
GPQA DiamondReasoning84.8%source
LMCAKnowledge36.4%source
OTIS Mock AIMEMath86.7%source
SimpleQA VerifiedKnowledge25.4%source
WebDev ArenaCoding1399source

API hosts

Live prices and uptime for every host serving Qwen3.5 Plus via OpenRouter.

HostInputOutputCache readAgent $/MUptime 24hQuant
Alibaba$0.30$1.80—$0.3799.7%—