GET/api/v1/eval/arena/leaderboard

Bradley-Terry + bootstrap-CI + style-controlled Elo leaderboard (T0-C round-6)

Returns the persisted arena ratings for the caller's org: rating ± 95% bootstrap CI + style-controlled rating per model. Populated nightly by `recompute-arena-ratings-job` from `pairwise_comparisons`. Each row exposes (model, rating, ciLow, ciHigh, styleControlledRating, battles, btScore).

Authentication

Send Authorization: Bearer YOUR_API_KEY on every request. Generate API keys at /dashboard/api-keys.

Parameters

orgId in query
string
cluster in query

Topic-cluster filter; omit for the global leaderboard.

string
minBattles in query
integer
limit in query
integer

Response

200 example

{
  "success": true
}

All status codes

200Leaderboard.
401(no description)
403(no description)
429(no description)

Code samples

cURL

curl -X GET \
  https://evalguard.ai/api/v1/eval/arena/leaderboard \
  -H "Authorization: Bearer $EVALGUARD_API_KEY" \

TypeScript

import { EvalGuard } from "@evalguard/sdk";

const client = new EvalGuard({ apiKey: process.env.EVALGUARD_API_KEY });

const response = await client.request({
  method: "GET",
  path: "/api/v1/eval/arena/leaderboard",
});
console.log(response);

Python

from evalguard import EvalGuard
import os

client = EvalGuard(api_key=os.environ["EVALGUARD_API_KEY"])

response = client.request(method="GET", path="/api/v1/eval/arena/leaderboard")
print(response)

Go

package main

import (
	"context"
	"fmt"
	"os"

	"github.com/evalguard/evalguard-go"
)

func main() {
	client := evalguard.NewClient(os.Getenv("EVALGUARD_API_KEY"))
	resp, err := client.Request(context.Background(), "GET", "/api/v1/eval/arena/leaderboard", nil)
	if err != nil { panic(err) }
	fmt.Println(resp)
}

Errors

401403429

Other Evals endpoints