PLPramaana

Alibaba

Qwen 3.7 Plus

Release date: 6/1/2026

An efficient multilingual, open-weight model with strong coding and tool-use results.

Pramaana IndexAccuracy: StandardLatency: StrongCost: Leading
Accuracy / Pramaana Index
62.52%± 0.65
Latency / Pramaana Index
284.80s
Cost / Test / Pramaana Index
$0.32
Context window
256k
Max output tokens
64k
Input modality
Text · Image
Hyperparameter settings

temperature: 0.2

reasoning: extended

fallbacks: disabled

Benchmark performance for Qwen 3.7 Plus

BenchmarksAccuracyRanking
Pramaana Index62.52% ± 0.6510 / 35
Pramaana Multimodal Index60.51% ± 0.8210 / 24
Legal Research Fidelity56.92% ± 0.9910 / 18
Financial Modeling Agent54.70% ± 1.1610 / 22
Clinical Documentation Safety56.22% ± 1.3310 / 19
MATH-50073.42% ± 1.509 / 112
GPQA Diamond76.40% ± 1.6710 / 121
MMLU-Pro71.82% ± 0.4810 / 119
Adaptive Tutoring Dialogue58.12% ± 0.6510 / 16
SWE-bench Verified55.60% ± 0.827 / 69
LiveCodeBench72.51% ± 0.998 / 127
Agentic Reliability48.71% ± 1.1610 / 12
Strategic Games Arena56.80% ± 1.3310 / 20
Benefits Navigator60.10% ± 1.5010 / 16

License: Apache 2.0 · Evaluation settings, model endpoints, and benchmark versions are frozen with each published run.

Read our methodology