Files one score event for the caller's org and answers 201 with it.

POST /v1/eval/scores

Eval → · All API references → · Six flows, four surfaces →


Address	`https://api.hanzo.ai/v1/eval/scores`
Method	`POST`
Operation	`post_eval_scores`
Auth	`Authorization: Bearer $HANZO_API_KEY`

This is how human review and out-of-band graders land beside the automatic ones: name the score, give it a value (or a stringValue for a categorical label), and attach it to a trace, a run, a dataset example, or any combination.

Scores are validated fail-closed. A value must be FINITE — NaN and Inf are 400 — and if the org has declared a rubric for this name, that rubric decides the type and the value must satisfy it: inside the numeric bounds, or one of the allowed categories. A caller cannot override the declared type by sending a different dataType.

A score is TELEMETRY, not metadata, so it needs the datastore: a deployment with none wired answers 503 rather than accepting a score it cannot persist. Requires a validated principal; 403 without one, and the org is stamped from the validated claim rather than read off the body.

Request

9 fields, body application/json (required).

Field	In	Type	Required	Description
`comment`	body	string	—	Comment is the grader's reasoning, truncated at 2000 characters.
`dataType`	body	string	—	DataType is NUMERIC, CATEGORICAL or BOOLEAN.
`datasetItemId`	body	string	—	ItemID attaches the score to one graded example.
`datasetName`	body	string	—	Dataset attaches the score to one dataset.
`name`	body	string	yes	Name is the score name.
`runName`	body	string	—	RunName attaches the score to one run.
`stringValue`	body	string	—	StringValue is the label of a CATEGORICAL score, which must be one the rubric allows.
`traceId`	body	string	—	TraceID attaches the score to one model call.
`value`	body	number	—	Value is the numeric score, which must be finite: NaN and Inf are refused.

Response

Status	Body	Meaning
`201`	scoreView	created

201 body — 9 fields.

Field	In	Type	Always	Description
`comment`	body	string	—	Comment is the grader's reasoning, truncated at 2000 characters.
`dataType`	body	string	—	DataType is NUMERIC, CATEGORICAL or BOOLEAN.
`id`	body	string	—	ID is the score event's handle.
`name`	body	string	—	Name is the score name, which a rubric of the same name governs.
`runName`	body	string	—	RunName is the run this score was recorded under, when it came from one.
`stringValue`	body	string	—	StringValue is the label of a CATEGORICAL score.
`timestamp`	body	string	—	Timestamp is when the score was recorded.
`traceId`	body	string	—	TraceID is the model call this score grades, when it grades one.
`value`	body	number	—	Value is the numeric score; for BOOLEAN it is 0 or 1.

Failure carries the platform error shape — see Errors.

Examples

hanzo evals scores create --name <name>

import { Configuration, EvalApi } from 'hanzoai';

const api = new EvalApi(new Configuration({ accessToken: process.env.HANZO_API_KEY }));
const { data } = await api.postEvalScores({ name: "<name>" });

from hanzoai.cloud import ApiClient, Configuration
from hanzoai.cloud.api import EvalApi

client = ApiClient(Configuration(access_token=os.environ["HANZO_API_KEY"]))
result = EvalApi(client).post_eval_scores(name="<name>")

cfg := cloud.NewConfiguration()
cfg.AddDefaultHeader("Authorization", "Bearer "+os.Getenv("HANZO_API_KEY"))
client := cloud.NewAPIClient(cfg)

resp, _, err := client.EvalAPI.PostEvalScores(context.Background()).Execute()
if err != nil {
	return err
}

use hanzo_cloud::apis::{configuration::Configuration, eval_api};

let mut cfg = Configuration::new();
cfg.bearer_access_token = std::env::var("HANZO_API_KEY").ok();

let result = eval_api::post_eval_scores(&cfg, Default::default()).await?;

import ai.hanzo.cloud.ApiClient;
import ai.hanzo.cloud.api.EvalApi;

ApiClient client = new ApiClient();
client.setRequestInterceptor(b -> b.header("Authorization", "Bearer " + System.getenv("HANZO_API_KEY")));

var result = new EvalApi(client).postEvalScores();

The method above is the one at the current release of the document. [email protected] (npm) and [email protected] (PyPI) were generated from an earlier release, where this operation carried a different id, so it spells the method differently — regenerating the clients is what makes the two agree. SDKs →

curl -X POST https://api.hanzo.ai/v1/eval/scores \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
       "name": "<name>"
     }'

The door reaches eval through the evals tool, which names its 16 operations with its own verbs — this one among them, under a name only the door declares. describe explains any of them:

curl -X POST https://api.hanzo.ai/v1/mcp \
  -H "Content-Type: application/json" \
  -d '{
       "jsonrpc": "2.0",
       "id": 1,
       "method": "tools/call",
       "params": {
         "name": "describe",
         "arguments": {
           "op": "list_eval_datasets"
         }
       }
     }'

Eval API · All Hanzo APIs · Interactive reference

Files one score event for the caller's org and answers 201 with it.

Request

Response

Examples

On this page