Speed analysis
Latency measured across all benchmark runs. P50 (median) and P95 (95th percentile) give a realistic picture of response speed under normal and peak load.
Pricing history
Direct provider rates per million tokens, plus a typical-conversation cost estimate.
Pricing over time
Input & output per 1M tokens · step-line = price changes
$0.4000
input / 1M
▲ +25% since first
$1.60
output / 1M
▲ +25% since first
Tokens per second
Throughput in tokens per second, derived from measured P50 latency. Higher is better; fluctuations track provider-side load.
Estimated from P50 latency × 200 output tokens — the absolute number depends on this assumption; the trend is what matters.
Availability
Availability
No measurements yet
We haven't recorded enough API calls to show availability stats for this model. Data appears once the model starts receiving live traffic.
Tokonomix benchmark verdicts
Qwen3.7 Plus maintains strong coding capabilities with consistent performance
Qwen3.7 Plus continues to demonstrate its primary strength in coding tasks, maintaining stable performance across benchmark windows. The model shows no significant changes in core capabilities, with its established pattern of strong software development support remaining intact. Users can expect consistent behavior for programming-related workflows, making it a reliable choice for technical applications. The model's reasoning capabilities remain at their previous level, neither advancing nor regressing in complex analytical tasks. For teams already integrated with Qwen3.7 Plus, this stability means no workflow disruptions or retraining requirements. The pricing update represents the primary change in this window, though the model's technical performance profile remains unchanged. Organizations evaluating this model should consider it for code generation, technical documentation, and software development assistance where it has proven effectiveness. However, those requiring advanced reasoning for complex problem-solving may still find limitations compared to frontier models. The consistency across windows suggests a mature, production-ready offering that prioritizes reliability over experimental feature additions.
Quality
—
Latency p50
—
Test runs
0