Altimate AI EVALS API
The EVALS API from Altimate AI — 18 operation(s) for evals.
The EVALS API from Altimate AI — 18 operation(s) for evals.
openapi: 3.1.0
info:
title: Fast ACCOUNT_COSTS EVALS API
version: 0.1.0
tags:
- name: EVALS
paths:
/evals/test-cases:
get:
tags:
- EVALS
summary: List Test Cases
description: List eval test cases filtered by tenant.
operationId: list_test_cases_evals_test_cases_get
responses:
'200':
description: Successful Response
content:
application/json:
schema:
items:
$ref: '#/components/schemas/TestCaseResponse'
type: array
title: Response List Test Cases Evals Test Cases Get
security:
- HTTPBearer: []
- HTTPBearer: []
post:
tags:
- EVALS
summary: Create Test Case
description: Create a new eval test case.
operationId: create_test_case_evals_test_cases_post
requestBody:
content:
application/json:
schema:
$ref: '#/components/schemas/TestCaseCreate'
required: true
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/TestCaseResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
/evals/test-cases/{test_case_id}:
get:
tags:
- EVALS
summary: Get Test Case
description: Get a specific test case by ID.
operationId: get_test_case_evals_test_cases__test_case_id__get
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: test_case_id
in: path
required: true
schema:
type: string
title: Test Case Id
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/TestCaseResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
delete:
tags:
- EVALS
summary: Delete Test Case
description: Delete a test case.
operationId: delete_test_case_evals_test_cases__test_case_id__delete
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: test_case_id
in: path
required: true
schema:
type: string
title: Test Case Id
responses:
'200':
description: Successful Response
content:
application/json:
schema: {}
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/test-cases/{test_case_id}/update:
post:
tags:
- EVALS
summary: Update Test Case
description: Update an existing test case.
operationId: update_test_case_evals_test_cases__test_case_id__update_post
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: test_case_id
in: path
required: true
schema:
type: string
title: Test Case Id
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/TestCaseCreate'
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/TestCaseResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/run:
post:
tags:
- EVALS
summary: Run Eval
description: 'Run an evaluation for a test case.
This simulates the conversation by calling the /ask endpoint and evaluates responses.'
operationId: run_eval_evals_run_post
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: navigationSource
in: query
required: false
schema:
anyOf:
- type: string
- type: 'null'
title: Navigationsource
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/RunEvalRequest'
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/EvalResultResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/batch-run:
post:
tags:
- EVALS
summary: Batch Run Eval
description: 'Run evaluations for multiple test cases.
Allows selecting specific test cases and naming the run.'
operationId: batch_run_eval_evals_batch_run_post
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: navigationSource
in: query
required: false
schema:
anyOf:
- type: string
- type: 'null'
title: Navigationsource
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/BatchRunEvalRequest'
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/BatchRunResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/results:
get:
tags:
- EVALS
summary: List Results
description: List eval results with server-side pagination.
operationId: list_results_evals_results_get
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: test_case_id
in: query
required: false
schema:
anyOf:
- type: string
- type: 'null'
title: Test Case Id
- name: page
in: query
required: false
schema:
type: integer
minimum: 1
default: 1
title: Page
- name: page_size
in: query
required: false
schema:
type: integer
maximum: 5000
minimum: 1
default: 1000
title: Page Size
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/PaginatedEvalResultsResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/results/{result_id}:
get:
tags:
- EVALS
summary: Get Result
description: Get a specific eval result by ID.
operationId: get_result_evals_results__result_id__get
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: result_id
in: path
required: true
schema:
type: string
title: Result Id
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/EvalResultResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
delete:
tags:
- EVALS
summary: Delete Result
description: Delete an eval result.
operationId: delete_result_evals_results__result_id__delete
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: result_id
in: path
required: true
schema:
type: string
title: Result Id
responses:
'200':
description: Successful Response
content:
application/json:
schema: {}
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/criterions:
get:
tags:
- EVALS
summary: List Criterions
description: List criterions filtered by tenant.
operationId: list_criterions_evals_criterions_get
responses:
'200':
description: Successful Response
content:
application/json:
schema:
items:
$ref: '#/components/schemas/CriterionResponse'
type: array
title: Response List Criterions Evals Criterions Get
security:
- HTTPBearer: []
- HTTPBearer: []
post:
tags:
- EVALS
summary: Create Criterion
description: Create a new criterion.
operationId: create_criterion_evals_criterions_post
requestBody:
content:
application/json:
schema:
$ref: '#/components/schemas/CriterionCreate'
required: true
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/CriterionResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
/evals/criterions/{criterion_id}:
get:
tags:
- EVALS
summary: Get Criterion
description: Get a specific criterion by ID.
operationId: get_criterion_evals_criterions__criterion_id__get
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: criterion_id
in: path
required: true
schema:
type: string
title: Criterion Id
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/CriterionResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
delete:
tags:
- EVALS
summary: Delete Criterion
description: Delete a criterion.
operationId: delete_criterion_evals_criterions__criterion_id__delete
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: criterion_id
in: path
required: true
schema:
type: string
title: Criterion Id
responses:
'200':
description: Successful Response
content:
application/json:
schema: {}
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/criterions/{criterion_id}/update:
post:
tags:
- EVALS
summary: Update Criterion
description: Update an existing criterion.
operationId: update_criterion_evals_criterions__criterion_id__update_post
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: criterion_id
in: path
required: true
schema:
type: string
title: Criterion Id
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/CriterionCreate'
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/CriterionResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/agents:
get:
tags:
- EVALS
summary: Get Playground Agents
description: 'Get all production agents from Langfuse for evaluation playground.
Returns a dictionary of agents with their details including:
- name, description, version, all_versions, prompt, labels, tags
Also returns the list of available models for agent configuration.'
operationId: get_playground_agents_evals_agents_get
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/PlaygroundAgentsResponse'
security:
- HTTPBearer: []
- HTTPBearer: []
/evals/agents/{agent_name}:
get:
tags:
- EVALS
summary: Get Playground Agent Details
description: 'Get detailed information for a specific agent from Langfuse.
Returns complete agent details including:
- agent_name, description, model, version, all_versions
- prompt (if include_prompt=True)
- labels, tags, config'
operationId: get_playground_agent_details_evals_agents__agent_name__get
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: agent_name
in: path
required: true
schema:
type: string
title: Agent Name
- name: include_prompt
in: query
required: false
schema:
type: boolean
default: false
title: Include Prompt
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/PlaygroundAgentInfo'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/agents/{agent_name}/update:
post:
tags:
- EVALS
summary: Update Playground Agent
description: 'Update an agent''s configuration in Langfuse.
Updates the agent''s prompt and config (description, model, etc.).
Creates a new version of the agent in Langfuse.
Returns the updated agent details.'
operationId: update_playground_agent_evals_agents__agent_name__update_post
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: agent_name
in: path
required: true
schema:
type: string
title: Agent Name
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/UpdateAgentConfigRequest'
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/PlaygroundAgentInfo'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/agents/{agent_name}/versions:
get:
tags:
- EVALS
summary: Get Agent Versions
description: 'Get all versions of an agent from Langfuse.
Returns a list of all versions with their metadata (description, model, labels, tags).
Versions are sorted in descending order (newest first).'
operationId: get_agent_versions_evals_agents__agent_name__versions_get
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: agent_name
in: path
required: true
schema:
type: string
title: Agent Name
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/AgentVersionsResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/agents/{agent_name}/versions/{version}:
get:
tags:
- EVALS
summary: Get Agent Version
description: 'Get a specific version of an agent from Langfuse.
Returns the complete agent details for the specified version including the prompt.'
operationId: get_agent_version_evals_agents__agent_name__versions__version__get
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: agent_name
in: path
required: true
schema:
type: string
title: Agent Name
- name: version
in: path
required: true
schema:
type: integer
title: Version
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/PlaygroundAgentInfo'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/models:
get:
tags:
- EVALS
summary: Get Available Models
description: 'Get list of all supported AI models for evaluation playground.
Returns a list of models with their IDs, names, providers, and descriptions.
Models include both OpenAI (GPT) and Anthropic (Claude) models.'
operationId: get_available_models_evals_models_get
responses:
'200':
description: Successful Response
content:
application/json:
schema: {}
security:
- HTTPBearer: []
- HTTPBearer: []
/evals/evaluate:
post:
tags:
- EVALS
summary: Run Batch Evaluation
description: 'Run evaluations for multiple test cases with all selected agents and criteria.
Each test case is evaluated with ALL selected agents available for use.
This is the main evaluation endpoint that accepts:
- List of agents with full configuration (all agents are available for each test case)
- List of test case IDs (reference to pool)
- List of criteria with complete data
Returns a job ID and status information.'
operationId: run_batch_evaluation_evals_evaluate_post
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: navigationSource
in: query
required: false
schema:
anyOf:
- type: string
- type: 'null'
title: Navigationsource
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/BatchEvaluationRequest'
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/EvaluationJobResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
/evals/jobs/{job_id}/status:
get:
tags:
- EVALS
summary: Get Job Status
description: 'Get the current status of an evaluation job.
Returns:
- job_id: The job identifier
- status: Current job status (submitted, running, completed, failed)
- total: Total number of test cases
- completed: Number of completed test cases
- successful: Number of successful test cases
- failed: Number of failed test cases
- test_statuses: List of individual test case statuses
- agents_count: Number of agents used
- criteria_count: Number of criteria used'
operationId: get_job_status_evals_jobs__job_id__status_get
deprecated: true
security:
- HTTPBearer: []
- HTTPBearer: []
parameters:
- name: job_id
in: path
required: true
schema:
type: string
title: Job Id
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/EvaluationJobStatus'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
components:
schemas:
PlaygroundAgentInfo:
properties:
agent_name:
type: string
title: Agent Name
description:
type: string
title: Description
model:
type: string
title: Model
version:
type: integer
title: Version
all_versions:
items:
type: integer
type: array
title: All Versions
prompt:
type: string
title: Prompt
labels:
items:
type: string
type: array
title: Labels
tags:
items:
type: string
type: array
title: Tags
config:
anyOf:
- additionalProperties: true
type: object
- type: 'null'
title: Config
type: object
required:
- agent_name
- description
- model
- version
- all_versions
- prompt
- labels
- tags
title: PlaygroundAgentInfo
description: Response for GET /evals/agents/{agent_name} endpoint
TurnResult:
properties:
turn_number:
type: integer
title: Turn Number
user_message:
type: string
title: User Message
expected_ai_response:
type: string
title: Expected Ai Response
actual_ai_response:
type: string
title: Actual Ai Response
evaluation:
additionalProperties: true
type: object
title: Evaluation
intermediate_messages:
anyOf:
- type: string
- type: 'null'
title: Intermediate Messages
type: object
required:
- turn_number
- user_message
- expected_ai_response
- actual_ai_response
- evaluation
title: TurnResult
CustomAgent:
properties:
name:
type: string
title: Name
description:
type: string
title: Description
prompt:
type: string
title: Prompt
type: object
required:
- name
- description
- prompt
title: CustomAgent
TestCaseCreate:
properties:
name:
type: string
title: Name
description:
anyOf:
- type: string
- type: 'null'
title: Description
turns:
items:
$ref: '#/components/schemas/TestTurn'
type: array
title: Turns
criteria:
anyOf:
- additionalProperties:
type: string
type: object
- type: 'null'
title: Criteria
tags:
anyOf:
- items:
type: string
type: array
- type: 'null'
title: Tags
tenant:
type: string
title: Tenant
default: All
team_id:
anyOf:
- type: string
- type: 'null'
title: Team Id
custom_agents:
anyOf:
- items:
$ref: '#/components/schemas/CustomAgent'
type: array
- type: 'null'
title: Custom Agents
user_view:
anyOf:
- additionalProperties: true
type: object
- type: 'null'
title: User View
type: object
required:
- name
- turns
title: TestCaseCreate
TestTurn:
properties:
user_message:
type: string
title: User Message
expected_ai_response:
type: string
title: Expected Ai Response
type: object
required:
- user_message
- expected_ai_response
title: TestTurn
CriterionCreate:
properties:
name:
type: string
title: Name
description:
type: string
title: Description
tenant:
type: string
title: Tenant
default: All
type: object
required:
- name
- description
title: CriterionCreate
BatchRunEvalRequest:
properties:
test_case_ids:
items:
type: string
type: array
title: Test Case Ids
run_name:
anyOf:
- type: string
- type: 'null'
title: Run Name
session_id:
anyOf:
- type: string
- type: 'null'
title: Session Id
type: object
required:
- test_case_ids
title: BatchRunEvalRequest
BatchEvaluationRequest:
properties:
agents:
items:
$ref: '#/components/schemas/app__schemas__evals__AgentConfig'
type: array
title: Agents
test_case_ids:
items:
type: string
type: array
title: Test Case Ids
criteria:
items:
$ref: '#/components/schemas/CriterionResponse'
type: array
title: Criteria
global_context:
$ref: '#/components/schemas/GlobalContext'
agent_context_overrides:
items:
$ref: '#/components/schemas/AgentContextOverride'
type: array
title: Agent Context Overrides
run_name:
anyOf:
- type: string
- type: 'null'
title: Run Name
session_id:
anyOf:
- type: string
- type: 'null'
title: Session Id
type: object
required:
- agents
- test_case_ids
- criteria
title: BatchEvaluationRequest
description: 'Request for running evaluations across agents, test cases, and criteria
This request contains agent and criteria data, with test case IDs for reference.
- Agents: Full configuration (both local and backend agents)
- Test Cases: IDs only (stored in pool, prevents duplication)
- Criteria: Full data (varies per run, stored in results for audit trail)
- Context: DataMate and Knowledge Base IDs (global and per-agent overrides)'
HTTPValidationError:
properties:
detail:
items:
$ref: '#/components/schemas/ValidationError'
type: array
title: Detail
type: object
title: HTTPValidationError
RunEvalRequest:
properties:
test_case_id:
type: string
title: Test Case Id
session_id:
anyOf:
- type: string
- type: 'null'
title: Session Id
type: object
required:
- test_case_id
title: RunEvalRequest
app__schemas__evals__AgentConfig:
properties:
id:
type: string
title: Id
name:
type: string
title: Name
description:
type: string
title: Description
prompt:
type: string
title: Prompt
model:
type: string
title: Model
is_local:
type: boolean
title: Is Local
default: false
type: object
required:
- id
- name
- description
- prompt
- model
title: AgentConfig
description: Agent configuration for evaluation - can be local or backend agent
TestCaseStatus:
properties:
test_case_id:
type: string
title: Test Case Id
test_case_name:
type: string
title: Test Case Name
status:
type: string
title: Status
score:
anyOf:
- type: number
- type: 'null'
title: Score
result_id:
anyOf:
- type: string
- type: 'null'
title: Result Id
type: object
required:
- test_case_id
- test_case_name
- status
title: TestCaseStatus
description: Status of a single test case in an evaluation job
PaginatedEvalResultsResponse:
properties:
results:
items:
$ref: '#/components/schemas/EvalResultResponse'
type: array
title: Results
total:
type: integer
title: Total
page:
type: integer
title: Page
page_size:
type: integer
title: Page Size
total_pages:
type: integer
title: Total Pages
has_next:
type: boolean
title: Has Next
has_prev:
type: boolean
title: Has Prev
type: object
required:
- results
- total
- page
- page_size
- total_pages
- has_next
- has_prev
title: PaginatedEvalResultsResponse
description: Paginated response for eval results.
CriterionResponse:
properties:
id:
type: string
title: Id
name:
type: string
title: Name
description:
type: string
title: Description
created_at:
type: string
title: Created At
updated_at:
type: string
title: Updated At
created_by:
anyOf:
- type: string
- type: 'null'
title: Created By
updated_by:
anyOf:
- type: string
- type: 'null'
title: Updated By
tenant:
type: string
title: Tenant
default: All
type: object
required:
- id
- name
- description
- created_at
- updated_at
title: CriterionResponse
PlaygroundAgentsResponse:
properties:
agents:
additionalProperties:
$ref: '#/components/schemas/PlaygroundAgentDetails'
type: object
title: Agents
available_models:
items:
type: string
type: array
title: Available Models
type: object
required:
- agents
- available_models
title: PlaygroundAgentsResponse
description: Response for GET /evals/agents endpoint
AgentVersionsResponse:
properties:
agent_name:
type: string
title: Agent Name
versions:
items:
$ref: '#/components/schemas/AgentVersionInfo'
type: array
title: Versions
total_versions:
type: integer
# --- truncated at 32 KB (41 KB total) ---
# Full source: https://raw.githubusercontent.com/api-evangelist/altimate-ai/refs/heads/main/openapi/altimate-ai-evals-api-openapi.yml