SYNCRETOS
Where knowledge
begins
Panel Operational
Deterministic grounding
0
Fact Units Parsed
0
Active Models
0%
Citation Grounding
Syncretos Panel · 3 Models
SESSION 0847 · UTC
Initializing deliberation pipeline…
QueryComplete
Parallel Drafts3 MODELS
Fusion AuditJUDGE v4
Ingestion Database
Grounding corpus for every deliberation. All citations trace to parsed fact units.
248
Documents
+12 this week
12,480
Fact Units
+340 today
99.2%
Index Health
All shards nominal
0.87
Min Reliability Threshold
Strict mode
Drop documents or click to ingest
PDF, DOCX, CSV, HTML · parsed into fact units within seconds
DocumentTypeUnitsStatusIngested
Processing QueueALL PARSED
Grounding Policy
Citation required per claimENFORCED
Unsourced assertion handlingQUARANTINE
Preprint weighting0.75×
Government source weighting1.15×
Contradiction escalationFUSION JUDGE
Model Panel
Select the models seated on your deliberation panel. Drafts run in parallel on every query.
Workspace Telemetry & Analytics
Comprehensive model benchmarking, consensus tracking, grounding audit, and cost analytics.
LAST 7 DAYS
JULY 2026
1,284
Total Syntheses
+18.2% vs prior week
98.7%
Avg Citation Grounding
+0.4 pts vs target
84.0%
Model Agreement Rate
-2.1 pts · high dissent
$0.042
Cost / Synthesis
-11% after routing
1.42s
p95 Panel Latency
-0.18s optimization
14,820
Fact Units Parsed
+1,240 this week
Syntheses per Day
7-DAY VOLUME · ALL DELIBERATION MODES
Draft Share Distribution
ENSEMBLE CONTRIBUTION BY MODEL
Claude Opus 538%
GPT-5.6 Sol34%
Gemini 3.6 Flash28%
Grounding Rate Trend
12-WEEK % CLAIMS VERIFIED BY INGESTION DB
Claims fully grounded96.4%
Partially grounded2.7%
Quarantined0.9%
Frontier Model Performance & Efficiency Leaderboard
Empirical Artificial Analysis & MMLU-Pro benchmarks across active panel models (July 2026 Corpus).
| Model & Provider | Share | Avg Latency | Reasoning Index | Citation Accuracy | Token Cost | Status |
|---|---|---|---|---|---|---|
Claude Opus 5 Anthropic |
38% | 1.42s | 97.8 / 100 | 98.9% | $0.0150 / 1k | Dominant |
GPT-5.6 Sol OpenAI |
34% | 1.18s | 96.9 / 100 | 98.2% | $0.0100 / 1k | High Speed |
Gemini 3.6 Flash Google |
28% | 0.48s | 94.8 / 100 | 97.1% | $0.0015 / 1k | Cost-Optimal |
Grok 4.5 xAI |
22% | 1.05s | 94.2 / 100 | 96.4% | $0.0060 / 1k | Analytical |
Claude Fable 5 Anthropic |
19% | 0.92s | 95.4 / 100 | 97.8% | $0.0080 / 1k | Creative |
Muse Spark 1.1 OpenAI |
15% | 0.38s | 92.1 / 100 | 95.2% | $0.0008 / 1k | Ultra Fast |
DeepSeek V4 DeepSeek |
12% | 1.28s | 93.6 / 100 | 96.7% | $0.0005 / 1k | Open Weights |
Domain Consensus Index
AGREEMENT vs DISSENT BREAKDOWN BY SUBJECT
Research Query Distribution
DELIBERATION VOLUME BY DOMAIN