Meta-Llama-3_3-70B-Instruct
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
$0.6700
input / 1M
— stable
$0.6700
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=8
Last 30 days
100.0%
n=8
Median response time
7,284ms
n=8
Based on 76 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)
8
OK responses (30d)
8
Total calls (7d)
8
OK responses (7d)
8
Tokonomix benchmark verdicts
Meta-Llama-3_3-70B-Instruct maintains 97.0 quality with stable performance
Meta-Llama-3_3-70B-Instruct continues to deliver consistent performance in its second benchmark window, maintaining its overall quality score of 97.0 out of 100. The model shows no measurable changes in quality metrics, demonstrating reliability across evaluation cycles. Latency remains at the p50 mark of 10556 milliseconds, indicating stable response times for this 70B parameter model. The multilingual category score holds steady at 97, confirming the model's continued strength in handling multiple languages effectively. With only one test run in the current window matching the previous baseline, the consistency suggests predictable behavior for production deployments. Users can expect the same high-quality outputs and performance characteristics observed in the initial benchmark period. The lack of variation between windows indicates a mature, stable offering suitable for applications requiring dependable language model performance. OVH AI Endpoints in the GRA region continues to provide reliable hosting for this model without performance degradation.
Quality
—
Latency p50
—
Test runs
0