Mistral-Small-3.2-24B-Instruct-2506
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
Evaluation results from judge-model scoring across diverse task categories. Scores reflect coherence, accuracy and instruction-following.
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
$0.0900
input / 1M
— stable
$0.2800
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
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
100.0%
n=1,469
Last 30 days
100.0%
n=5,476
Median response time
1,765ms
n=5,476
Based on 5,856 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)
5,476
OK responses (30d)
5,476
Total calls (7d)
1,469
OK responses (7d)
1,469
Tokonomix benchmark verdicts
Quality dips to 94.8 as latency recovers with 52% improvement
Mistral-Small-3.2-24B-Instruct-2506 shows a mixed performance shift in this benchmark window. The overall quality score decreased from 96.7 to 94.8, representing a modest decline of 1.9 points. However, the model achieved a substantial latency improvement, with p50 response time dropping from 9982ms to 4758ms—a 52% reduction that brings performance back to more competitive levels. Category performance reveals notable changes in capability distribution. Coding scores increased impressively from 94 to 99, while multilingual performance improved from 96 to 100, demonstrating excellence in these domains. Creative tasks scored 85 in the current window, though no previous creative score exists for direct comparison. Notably, reasoning scores were 100 in the previous window but are not reported in the current results, making it unclear whether this category was tested. The dramatic latency recovery suggests infrastructure or optimization improvements at the OVH GRA endpoint, reversing the degradation seen in the previous window. Users requiring fast response times will benefit from this enhancement. The slight overall quality decrease appears driven by shifts in category mix and performance rather than across-the-board degradation, with coding and multilingual capabilities reaching peak scores.
Quality
94.8
Latency p50
4,758 ms
Test runs
5