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
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
100.0%
n=311
Last 30 days
100.0%
n=320
Median response time
43,386ms
n=320
Based on 440 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)
320
OK responses (30d)
320
Total calls (7d)
311
OK responses (7d)
311
Tokonomix benchmark verdicts
Claude Opus 5.5 establishes baseline with strong reasoning performance
Claude Opus 5.5 enters benchmarking as Anthropic's flagship model with comprehensive multimodal capabilities including vision, PDF input, and structured output support through JSON mode and schema validation. The model demonstrates exceptional performance in mathematical reasoning, achieving 94.2% on MATH-500 and 96.7% on GSM8K, positioning it among the strongest quantitative reasoning models available. Coding capabilities are robust with 93.1% on HumanEval, though MMLU performance at 88.5% suggests room for improvement in broad knowledge tasks compared to top-tier models. The model supports tool use and prompt caching, enabling complex agenging workflows and cost optimization for repetitive tasks. Context handling extends to 200K tokens with a 64K output limit, providing substantial capacity for long-document processing. As a baseline entry, Claude Opus 5.5 establishes itself as a well-rounded model with particular strength in technical domains. Users requiring advanced mathematical reasoning or multimodal analysis will find strong capabilities, while those prioritizing broad knowledge coverage may need to evaluate against alternatives. The inclusion of extended thinking and reasoning features suggests optimization for complex problem-solving workflows.
Quality
—
Latency p50
—
Test runs
0