Retirement date
The provider will retire this model on November 4, 2026. Until then it works as usual.
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
What you pay per million tokens when you use this model on Tokonomix, plus an estimate for a typical conversation.
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
How often this model answers when we call it — measured across real API requests and live tests over the last 30 days. This is separate from quality: these numbers only tell you whether the model responds, not how good the answer is.
Last 7 days
—
Last 30 days
100.0%
n=30
Median response time
1,852ms
n=30
Based on 410 measurements over the last 30 days.
Technical details
Only live API calls and live-test requests count — internal probes and benchmark runs are excluded.
Calls with a custom API key (BYOK) are excluded: those failures are key-specific, not a sign of model downtime.
Failed calls are NOT included in quality scores — quality is measured on successful responses only. Availability and quality are independent signals.
Median response time (p50) across successful calls with a recorded duration. Outliers (very slow or very fast calls) pull the median less than the average.
Total calls (30d)
30
OK responses (30d)
30
Total calls (7d)
0
OK responses (7d)
0
Tokonomix benchmark verdicts
Major quality leap to 93 with factual scores up 37 points
Mistral-Small-3.2-24B-Instruct-2506 demonstrates a substantial quality improvement, jumping 13.8 points to reach an overall score of 93 out of 100. The most dramatic change is in factual accuracy, which surged from 55 to 92, resolving the critical weakness identified in the previous benchmark period. Reasoning capabilities now join the measured categories at a strong 92, indicating enhanced logical processing. Coding performance remains excellent, climbing slightly from 92 to 96, while multilingual support holds steady at 92, maintaining its previously strong position. Latency has increased moderately from 6438 ms to 7183 ms at the median, representing roughly an 11 percent slowdown. This tradeoff appears to correlate with the significant quality gains across categories. The model now presents a well-rounded performance profile without the factual quality gap that previously limited its reliability. Users requiring high-accuracy factual responses will find this iteration substantially more dependable, though those with strict latency requirements should account for the modest speed decrease. With consistent performance across five test runs in both windows, these improvements appear stable and reproducible.
Quality
93.0
Latency p50
7,183 ms
Test runs
5