API Rate Limit
All API endpoints share a global rate limit based on your API key.
| Parameter | Value |
|---|
| Requests per window | 50 |
| Window duration | 60 seconds |
| Scope | Per API key |
| Header | Description |
|---|
X-RateLimit-Limit | Maximum requests allowed in the window |
X-RateLimit-Remaining | Requests remaining in the current window |
X-RateLimit-Reset | Seconds until the window resets |
Retry-After | Seconds to wait before retrying |
RateLimitError with a retry_after attribute you can use to sleep and retry.
from valiqor import RateLimitError
import time
try:
result = client.eval.evaluate(dataset=data, metrics=["coherence"])
except RateLimitError as e:
time.sleep(e.retry_after or 5)
result = client.eval.evaluate(dataset=data, metrics=["coherence"])
Monthly Quotas
API Call Quotas
Each organisation has a monthly API call quota that resets automatically.
| Tier | Monthly API Calls |
|---|
| Free | 100 |
| Pro | 1,000 |
| Enterprise | 10,000 |
Token Quotas
A separate token-based quota tracks LLM token consumption across all services.
| Tier | Monthly Token Quota |
|---|
| Default | 1,000,000 tokens |
QuotaExceededError (for API call quotas) or TokenQuotaExceededError (for token quotas).
from valiqor import QuotaExceededError, TokenQuotaExceededError
try:
result = client.failure_analysis.run(dataset=data)
except TokenQuotaExceededError as e:
print(f"Token quota exceeded: {e.current_tokens}/{e.token_limit}")
except QuotaExceededError as e:
print(f"API quota exceeded: {e.current_usage}/{e.quota_limit}")
Trial Limits
Users who have not verified their email operate in trial mode.
| Limit | Value |
|---|
| Rows per run | 25 (dataset is silently truncated) |
| Total trial runs | 3 |
| After 3 runs | Blocked until email is verified |
Run valiqor verify to verify your email and unlock 50 free runs on the free tier.
Dataset Size Limits
| Limit | Value |
|---|
| Maximum items per request | 1,000 |
evaluate(), audit(), and failure_analysis.run(). The SDK raises DatasetTooLargeError if the limit is exceeded.
from valiqor import DatasetTooLargeError
try:
result = client.eval.evaluate(dataset=large_dataset, metrics=["coherence"])
except DatasetTooLargeError as e:
print(f"Dataset has {e.dataset_size} items, max is {e.max_allowed}")
Playground Limits
The Failure Analysis playground (client.failure_analysis.playground()) has stricter limits because it is a lightweight, single-item endpoint.
| Limit | Value |
|---|
| Requests per minute | 2 |
| Analyses per day | 10 |
| Max input tokens | 4,000 |
| Max input field length | 2,000 characters |
| Max output field length | 5,000 characters |
| Max context field length | 5,000 characters |
| Max tool calls | 5 |
Other Limits
| Resource | Limit |
|---|
| API keys per user | 5 |
| Request timeout | 300 seconds (5 minutes) |
Checking Your Usage
SDK
from valiqor import ValiqorClient
client = ValiqorClient(api_key="vq_...")
usage = client.get_usage()
print(usage)
# Returns per-service API call counts, per-service token usage,
# quota limits, remaining quota, period start/end, and reset date.
API
curl -H "Authorization: Bearer vq_..." \
https://api.valiqor.com/v2/auth/usage
| Field | Description |
|---|
total_api_usage | Total API calls this period |
evaluation_usage | Eval API call count |
security_audit_usage | Security audit call count |
security_redteam_usage | Red team simulation call count |
tracing_usage | Trace upload call count |
scanner_usage | Scan upload call count |
failure_analysis_usage | FA API call count |
total_token_usage | Total tokens consumed this period |
evaluation_tokens | Tokens used by evaluations |
security_audit_tokens | Tokens used by security audits |
security_redteam_tokens | Tokens used by red team runs |
tracing_tokens | Tokens used by tracing |
scanner_tokens | Tokens used by scanner |
failure_analysis_tokens | Tokens used by failure analysis |
api_quota_limit | Your plan’s API call quota |
api_quota_remaining | API calls remaining this period |
token_quota_limit | Your plan’s token quota |
token_quota_remaining | Tokens remaining this period |
period_start | Start of the current billing period |
period_end | End of the current billing period |
reset_date | When quotas reset (monthly) |
org_id | Your organisation ID |
org_name | Your organisation name |
plan | Your current plan (free, pro, enterprise) |
CLI
Shows authentication status, current configuration, cloud connection, and module availability.
