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.
Quality scores
How this model compares to the rest of the field on each prompt category, from a pairwise fit over the same prompts. The raw judge score sits underneath each number.
Win rate per category: how often this model beats a field-average model on a prompt from that category. 50% is average, not a failing grade. It is not a percentage of correct answers.
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
$0.0700
input / 1M
— stable
$0.2600
output / 1M
— stable
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.
Capabilities
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
Quality drops 6.5 points to 84.7 despite 37% latency improvement
Qwen3-Coder-30B-A3B-Instruct shows a concerning performance decline in this benchmark window, with overall quality falling from 91.1 to 84.7 points. The model demonstrates uneven category performance, maintaining exceptional multilingual capabilities at 100 and strong coding performance at 94, but experiencing a sharp drop in creative tasks to just 60 points. This represents a significant departure from the previous window where creative scoring reached 92. The absence of factual and reasoning category scores in the current window makes direct comparison challenging. On the positive side, latency improved substantially by 37%, dropping from 3176ms to 2009ms at the median, which should provide a noticeably faster user experience. The model appears optimized for code-related and multilingual tasks, but the creative performance regression and overall quality decline suggest potential issues with recent changes or configuration. Users requiring strong creative writing capabilities may want to exercise caution, while those focused on coding and multilingual support will still find solid performance with improved response times.
Quality
84.7
Latency p50
2,009 ms
Test runs
5