Supports using OpenAI models as independent raters for multi-model evaluation and statistical consensus measurement alongside other AI providers.
One command. Find out if your AI agrees with itself.
ConKurrence is a statistically validated consensus measurement toolkit for AI evaluation pipelines. It uses multiple AI models as independent raters, measures inter-rater reliability with Fleiss' kappa and bootstrap confidence intervals, and routes contested items to human experts.
npm install -g conkurrenceRelated MCP server: CI-1T Prediction Stability Engine
Use ConKurrence as an MCP server in Claude Desktop or any MCP-compatible client:
npx conkurrence mcpAdd to your claude_desktop_config.json:
{
"mcpServers": {
"conkurrence": {
"command": "npx",
"args": ["-y", "conkurrence", "mcp"]
}
}
}/plugin marketplace add AlligatorC0der/conkurrenceMulti-model evaluation — Run your schema against Bedrock, OpenAI, and Gemini models simultaneously
Statistical rigor — Fleiss' kappa with bootstrap confidence intervals, Kendall's W for validity
Self-consistency mode — No API keys needed; uses the host model via MCP Sampling
Schema suggestion — AI-powered schema design from your data
Trend tracking — Compare runs over time, detect agreement degradation
Cost estimation — Know the cost before running
Tool | Description |
| Execute an evaluation across multiple AI raters |
| Generate a detailed markdown report |
| Side-by-side comparison of two runs |
| Track agreement over multiple runs |
| AI-powered schema suggestion from your data |
| Validate a schema before running |
| Estimate cost and token usage |
Homepage: conkurrence.com
Terms of Service: app.conkurrence.com/terms
Privacy Policy: app.conkurrence.com/privacy
BUSL-1.1 — Business Source License 1.1
This server cannot be deployed
MCP-native AI evaluation: rubric audits, eval suites, and proof reports for AI/LLM output.
Measure what ChatGPT, Claude, Gemini and 4 more AI engines say about any business. No auth.
Multi-LLM council: 25+ frontier models in parallel, consensus scoring, verdict-first code review.
Prompt evals over MCP: run a prompt on your dataset, score each output 1-5 with an LLM judge.
