Is the traces behind your evaluations — one per model call an evaluation made,…

Is the traces behind your evaluations — one per model call an evaluation made, carrying its input, output, model and timing — narrowed by any of…

GET /v1/eval/traces

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


Address	`https://api.hanzo.ai/v1/eval/traces`
Method	`GET`
Operation	`get_eval_traces`
Auth	`Authorization: Bearer $HANZO_API_KEY`

Is the traces behind your evaluations — one per model call an evaluation made, carrying its input, output, model and timing — narrowed by any of sessionId, runName and datasetName.

Scoped by org AND by project: the project is the caller's server-minted scope, not a parameter, so it cannot be widened by asking. Requires a validated principal; 403 without one. Traces live in the datastore, so a deployment with none wired answers 503 rather than an empty page.

Request

4 fields.

Field	In	Type	Required	Description
`sessionId`	query	string	—	SessionID narrows to one session, which for an evaluation is one run.
`runName`	query	string	—	RunName narrows to the calls one run made.
`datasetName`	query	string	—	Dataset narrows to the calls made against one dataset.
`limit`	query	integer	—

Response

Status	Body	Meaning
`200`	traceList	ok

200 body — 16 fields.

Field	In	Type	Always	Description
`data`	body	traceView[]	—	Data is the caller org's traces matching the filters, bounded by limit.
`data[].apiKeyHash`	body	string	—	APIKeyHash is the non-reversible credential ref (never a plaintext key), so a trace correlates to the key that drove it without the store holding a secret.
`data[].datasetItemId`	body	string	—	ItemID is the example the call answered.
`data[].datasetName`	body	string	—	Dataset is the set the graded example came from.
`data[].endTime`	body	string	—	EndTime is when it returned.
`data[].id`	body	string	—	ID is the trace's handle, the value a score points at.
`data[].input`	body	object	—	Input is what the model was given.
`data[].latencyMs`	body	number	—	LatencyMs is EndTime-StartTime in milliseconds, nil when the trace carries no timing (so the console renders "—", never a fabricated 0).
`data[].model`	body	string	—	Model is the model that answered.
`data[].name`	body	string	—	Name is the trace's label, "eval:<run>" for a call a run made.
`data[].output`	body	string	—	Output is what it answered.
`data[].projectId`	body	string	—	ProjectID is the sub-scope within the org the call was made under.
`data[].runName`	body	string	—	RunName is the run the call belongs to.
`data[].sessionId`	body	string	—	SessionID groups the calls of one run.
`data[].startTime`	body	string	—	StartTime is when the call began.
`data[].timestamp`	body	string	—	Timestamp is the trace's own clock, equal to StartTime for a timed call.

Failure carries the platform error shape — see Errors.

Examples

hanzo evals traces

import { Configuration, EvalApi } from 'hanzoai';

const api = new EvalApi(new Configuration({ accessToken: process.env.HANZO_API_KEY }));
const { data } = await api.getEvalTraces();

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).get_eval_traces()

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

resp, _, err := client.EvalAPI.GetEvalTraces(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::get_eval_traces(&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).getEvalTraces();

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 https://api.hanzo.ai/v1/eval/traces \
  -H "Authorization: Bearer $HANZO_API_KEY"

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

Is the traces behind your evaluations — one per model call an evaluation made,…

Request

Response

Examples

On this page