Get Eval Run

Get a single eval run with full per-case results.

GET /api/evals/runs/{run_id}

Get Eval Run

cURL

curl --request GET \
  --url https://api.example.com/api/evals/runs/{run_id}

200

{
  "avg_context_recall": 0.5,
  "avg_factual_correctness": 0.5,
  "avg_faithfulness": 0.5,
  "dataset_version": "<string>",
  "results": [
    {
      "context_recall": 0.5,
      "factual_correctness": 0.5,
      "faithfulness": 0.5,
      "latency_ms": 1,
      "test_case_id": "<string>",
      "generated_response": "",
      "reranker_available": true,
      "response_relevancy": 0.5,
      "retrieved_source_ids": [
        "<string>"
      ]
    }
  ],
  "run_id": "<string>",
  "total_cases": 1,
  "avg_response_relevancy": 0.5,
  "commit_sha": "<string>",
  "embedding_model": "<string>",
  "gate_passed": true,
  "llm_model": "<string>",
  "reranker_available_pct": 0.5,
  "reranker_model": "<string>",
  "timestamp": "2023-11-07T05:31:56Z"
}

Path Parameters

run_id · string · required

Response

200

application/json

Successful Response

Summary of a full evaluation run.

avg_context_recall · number · required

Required range: 0 <= x <= 1

avg_factual_correctness · number · required

Required range: 0 <= x <= 1

avg_faithfulness · number · required

Required range: 0 <= x <= 1

dataset_version · string · required

results · EvalResult · object[] · required

Show child attributes

run_id · string · required

total_cases · integer · required

Required range: x >= 0

avg_response_relevancy · number | null

Required range: 0 <= x <= 1

commit_sha · string | null

embedding_model · string | null

gate_passed · boolean | null

llm_model · string | null

reranker_available_pct · number | null

Required range: 0 <= x <= 1

reranker_model · string | null

timestamp · string<date-time>