Live Platform Data
Drux Activity
Real-world AI model performance from every query run on Drux. Speed, reliability, and consensus measured on actual user questions — not controlled benchmarks.
2,072 searches · 6,268 model calls · live
22.6Mtokens processed in the last 30 days
Platform Summary
614 Hub articles published, read 3464 times in total. 2072 total searches completed. 6268 individual model calls recorded. Average consensus score: 65%. Model completion rate: 77%.
Top Models by Speed
- inception/mercury-2: avg 5.6s, 98% reliability, 208 queries
- microsoft/phi-4: avg 8.3s, 98% reliability, 206 queries
- mistralai/mistral-small-2603: avg 8.5s, 76% reliability, 205 queries
- anthropic/claude-sonnet-4.6: avg 8.5s, 100% reliability, 3 queries
- google/gemini-3.5-flash: avg 9.3s, 100% reliability, 3 queries
Most Agreed Questions
- Is olive oil healthier than vegetable oil? — consensus score 90%
- Can you negotiate salary after receiving a job offer? — consensus score 90%
- How does Google’s People Also Ask feature work? — consensus score 90%
- What’s healthier: coffee or tea? — consensus score 90%
- Is it true that sugar causes hyperactivity in children? — consensus score 90%
Platform Pulse
Total Searches
2,072
This Week
91
last 7 days
Today
4
Model Calls
6,268
across all searches
Avg Consensus Score
65%
across all completed searches
Model Completion Rate
77%
calls that returned successfully
Model Leaderboard — on Drux
Last 1,000 searches · click column to sort
| Model | Avg Speed▲ |
|---|---|
1Mercury 2FREE | 5.6s |
2Phi-4FREE | 8.3s |
3Mistral Small 4FREE | 8.5s |
4Claude Sonnet 4.6PAID | 8.5s |
5Gemini 3.5 FlashPAID | 9.3s |
6GPT-5.5PAID | 10.0s |
7GLM 5.2FREE | 13.2s |
8Solar Pro 3FREE | 13.5s |
9Hermes 3 70BFREE | 14.1s |
10Command RFREE | 15.1s |
11Gemma 4 31BFREE | 18.5s |
12GPT OSS 120BFREE | 19.0s |
13Llama 4 MaverickFREE | 19.6s |
14DeepSeek V3.2FREE | 21.3s |
15Seed 1.6FREE | 30.3s |
16Qwen3.5 397BPAID | 31.7s |
⚡ Fastest on Drux
Mercury 2
avg 5.6s · 98% reliable
✓ Most Reliable on Drux
GPT-5.5
100% completion · 3 calls
Consensus Digest — Recent Public Searches
Updated every 5 min · Data from last 1,000 searches
About This Data
How is this different from other AI benchmarks?
Most AI benchmarks are run under controlled laboratory conditions on standardised test sets. Drux Activity measures performance on real user questions — diverse, unpredictable, and representative of actual use. Speed and reliability numbers here reflect what users actually experience.
What is a Consensus Win?
When multiple models answer the same question and their responses agree closely (consensus score ≥ 7/10), all models that responded are credited with a Consensus Win. A high Consensus Win rate means a model consistently lands on the same answer as its peers — a signal of reliability beyond just accuracy.
Why do free models have more queries?
Drux randomly selects models for each search. Free-tier models are included in more searches because they are available to all users. Paid models are selected when users opt into premium tiers. Query count reflects availability and tier distribution, not quality — sort by Speed or Reliability for a fairer comparison.
How often is this updated?
Every figure is computed live when the page loads. Search, model-call, consensus and completion totals cover every completed search; the per-model leaderboard uses the most recent 1,000 searches, so it reflects how models perform now rather than months ago.