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
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
gpt-5.6-terra quality jumps 23 points with 29% faster response times
OpenAI's gpt-5.6-terra shows substantial improvement across both performance dimensions in this benchmark window. Overall quality increased from 69.0 to 92.0, representing a 23-point gain that places the model in strong competitive standing. Response latency improved from 3477ms to 2454ms at the median, a 29% reduction that should noticeably impact user experience. The quality improvements appear concentrated in specific areas. While coding performance decreased slightly from a perfect 100 to 92, the model now shows balanced performance with creative writing scoring 92 and reasoning at 92. The previous window's concerning factual accuracy score of 11 is absent from current results, though it's unclear whether this represents true improvement or simply wasn't tested this window. With only 3 test runs compared to 4 previously, confidence in these measurements is moderate. Users should expect meaningfully faster responses and more consistent quality across task types. The convergence of coding, creative, and reasoning scores at 92 suggests OpenAI may have rebalanced the model's capabilities rather than optimizing for narrow benchmarks. Production deployments should monitor factual accuracy given the previous window's weakness in this area.
Quality
92.0
Latency p50
2,454 ms
Test runs
3
gpt-5.6-terra
by OpenAI
- Context window
- 1.05M tokens
- Input price
- $2.60 / 1M
- Output price
- $15.60 / 1M
- Tier
- Tier B — Production
- Modality
- Text + vision
- API type
- REST · streaming
- Benchmark runs
- 240
More from OpenAI