VECTOR WIREAI INTELLIGENCE
PKT
Refresh Models Deals Regulatory Sources

Questions this page answers

Which is better, Qwen3.5 122B A10B or Qwen3.7 Plus Preview?
Across 44 shared benchmarks, Qwen3.5 122B A10B scores higher on 6 and Qwen3.7 Plus Preview on 38. The widest gap is OmniScience Non-Hallucination, where Qwen3.7 Plus Preview scores 74.5 against 12.9. Both cost $0.40 per million input tokens on tracked API pricing; Qwen3.7 Plus Preview is cheaper on output ($1.60 against $3.20).

Qwen3.5 122B A10B vs Qwen3.7 Plus Preview

Across 44 shared benchmarks, Qwen3.5 122B A10B scores higher on 6 and Qwen3.7 Plus Preview on 38. The widest gap is OmniScience Non-Hallucination, where Qwen3.7 Plus Preview scores 74.5 against 12.9. Both cost $0.40 per million input tokens on tracked API pricing; Qwen3.7 Plus Preview is cheaper on output ($1.60 against $3.20).

AlibabavsAlibaba44 shared benchmarks638 head-to-head
BenchmarkQwen3.5 122B A10BQwen3.7 Plus Preview
AA Agentic Index21.320.8
AA Intelligence32.839.4
AA-LCR70.369
AA-Omniscience-41.52.4
arena_vision12461281
Artificial Analysis Coding Index45.755.9
baby_vision40.270.4
bfcl_v472.272.9
critpt0.99.1
DeepPlanning24.162.3
ERQA6269.8
gdpval24.322.3
GPQA Diamond86.690.3
HLE47.535.6
IFBench76.179.1
ifeval93.494.6
include82.883
LiveCodeBench v678.989.6
LVBench74.476.2
mathvision86.290.3
MLVU87.387.4
mmlu_prox82.285.4
mmlu_redux9494.5
MMLU-Pro86.788.5
mmmlu86.789
MMMU-Pro76.980.5
OmniDocBench 1.589.891.4
OmniScience Accuracy24.422.5
OmniScience Non-Hallucination12.974.5
OSWorld-Verified5873.3
RealWorldQA85.186.9
scicode4251.3
screenspot_pro_no_tools70.479
SimpleVQA0.60.8
supergpqa67.171.4
SWE-bench Verified7277.7
TauBench V3 - Banking15.317.9
Terminal-Bench 2.049.470.3
Terminal-Bench 2.147.661
Terminal-Bench Hard31.147
VideoMMMU8285.4
VITA-Bench33.645.6
τ²-Bench Telecom (AA run)93.693
τ³-Bench13.617.9

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, alibaba-official), otherwise the lowest tracked offer.