The 1Stack AI Mission Control API provides OpenAI-compatible and Anthropic-compatible interfaces for LLM inference, embeddings, reranking, and audio processing. All endpoints use standard HTTP methods and return JSON (except audio/speech which returns binary audio).
Base URL: https://taas.cloudsigma.com/v1
Codex native base URL: https://taas.cloudsigma.com/v1/codex
1Stack AI Mission Control now also exposes a native Codex-compatible endpoint for clients such as Codex ACP and other Codex-native integrations. Use the standard /v1 OpenAI-compatible base for normal SDK traffic, and the native Codex base when a client expects Codex-specific request semantics. Anthropic SDK clients are served natively via POST /v1/messages on the same base URL.
Authenticate every request with a 1Stack AI Mission Control Bearer token in the Authorization header.
Tokens are prefixed with taas_ and are scoped per team or project.
Authorization: Bearer taas_xxxxxxxxxxxxxxxx
Generate API tokens from Settings → API Tokens in the 1Stack AI Mission Control console. Keep tokens secret — they grant full access to your account's quota.
If you want repeated requests to stay on the same routed session, you can send a stable session identifier. This improves continuation reliability and makes request-level debugging easier in the 1Stack AI Mission Control console.
| Input | Where to send it | Notes |
|---|---|---|
X-Session-Id | HTTP header | Highest precedence. Best option when you control headers. |
metadata.session_id | JSON body metadata | Stable explicit session id for OpenAI-compatible request bodies. |
metadata.sticky_key | JSON body metadata | Compatibility fallback, especially useful for Anthropic-style continuation flows. |
Precedence is X-Session-Id → metadata.session_id → metadata.sticky_key. If none are supplied, 1Stack AI Mission Control can still infer continuity heuristically, but explicit IDs are more predictable.
Application errors use code - human-readable explanation. For automation, read error.code; for a user-facing message, prefer error.human_message. The complete formatted string remains in error.message.
{"error":{"code":"billing_limit_reached","type":"permission_error","message":"billing_limit_reached - Out of funds. Contact your billing administrator.","human_message":"Out of funds. Contact your billing administrator."}}
A 403 billing_limit_reached means the effective payer needs attention. The payer may be a project, organisation billing user, delegated user, or yourself. A budget is a spending setting, not wallet funds. Changing a valid API key will not fix an empty payer wallet.
HTTP status, Retry-After, and structured retry metadata remain authoritative. Do not automate by matching English text. Handle unknown codes safely, and do not automatically repeat operations with side effects. Streams may begin with HTTP 200 and later emit an error event; treat partial output as incomplete.
Legacy portal responses retain their error or detail fields and add top-level code, message, and human_message. Provider-specific and numeric JSON-RPC codes keep their original identity. Split a formatted message at the first exact - separator only when its prefix is a valid code.
Machine-readable error code reference · Download reference JSON
These are statically declared platform codes. Dynamic provider codes may also occur. Use the actual response status for codes without one fixed HTTP status.
| Code | Meaning | Suggested action |
|---|---|---|
account_status | Account status. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
alias_redirect_conflict | Alias redirect conflict. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
all_providers_exhausted | All providers exhausted. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
anthropic_terminal_finalization_failed | Anthropic terminal finalization failed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
async_completion_too_large | Async completion too large. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
atomic_canary_binding_missing | Atomic canary binding missing. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
atomic_canary_forbidden | Atomic canary forbidden. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
atomic_canary_identity_missing | Atomic canary identity missing. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
atomic_canary_lane_mismatch | Atomic canary lane mismatch. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
atomic_canary_manual_review | A dispatched atomic paid canary has an unconfirmed settlement. | Do not repeat the operation or release its hold. Give support the request ID for operator review and reconciliation. |
atomic_canary_refused | Atomic canary refused. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
atomic_canary_unknown | A dispatched atomic paid canary has an unknown supplier outcome. | Do not repeat the operation or release its hold. Give support the request ID to reconcile the original supplier outcome and settlement. |
auth_context_denied | Auth context denied. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
auth_issue | Auth issue. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
auth_unavailable | Auth unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
authentication | Authentication. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
authentication_required | Authentication required. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
autorouter_no_candidate | Autorouter no candidate. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
autorouter_unavailable | Autorouter unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
backend_error | Backend error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
backup_chain_exhausted | Backup chain exhausted. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
billing_limit_reached | The effective payer cannot fund or authorize this request. | Check the effective payer, available funds, spending limits and billing configuration. Fund the payer or contact its billing administrator; retry only after correction. |
billing_proof_unavailable | Billing proof unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
billing_unavailable | Billing unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
bridge_operation_failed | Bridge operation failed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
bridge_timeout | Bridge timeout. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
budget_alert_history_unavailable | Budget alert history unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
budget_exceeded | The API key spending or usage budget was reached. | Review the key budget and usage. Increase an authorized limit or wait for the applicable reset; adding wallet funds alone may not resolve a budget cap. |
byok_budget_unknown | Byok budget unknown. | Review the effective payer, balance, spending limits and payment configuration. Contact the billing administrator before retrying. |
byok_mode_unavailable | Byok mode unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
capability_denied | Capability denied. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
capability_unsupported | Capability unsupported. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
capacity_exceeded | Capacity exceeded. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
claim_expired | Claim expired. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
claim_mismatch | Claim mismatch. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
claude_cli_result_error | Claude cli result error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
claude_code_async_tool_result_forward_failed | Claude code async tool result forward failed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
claude_code_incomplete_stream | Claude code incomplete stream. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
claude_code_missing_endpoint | Claude code missing endpoint. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
claude_code_node_error | Claude code node error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
claude_code_tool_result_forward_failed | Claude code tool result forward failed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
client_secret_collision | Client secret collision. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
client_secret_limit | Client secret limit. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
client_secret_origin_mismatch | Client secret origin mismatch. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
client_secret_principal_mismatch | Client secret principal mismatch. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
client_secret_scope_mismatch | Client secret scope mismatch. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
codex_backend_error | Codex backend error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
codex_error | Codex error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
codex_org_context_required | Codex org context required. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
codex_org_lane_disabled | Codex org lane disabled. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
codex_token_error | Codex token error. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
collection_error | Collection error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
configuration_error | Configuration error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
conflict | Conflict. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
context_length | Context length. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
context_length_exceeded | Context length exceeded. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
context_length_exceeded_orchestration | Context length exceeded orchestration. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
corp_key_exception | Corp key exception. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
csrf_failed | Csrf failed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
csrf_rejected | The browser request failed the same-origin security check. | Reload the page on the correct portal domain, sign in again, and repeat the action. |
current_quota_unavailable | Current quota unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
delivery_lease_unknown | Delivery lease unknown. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
document_conversion_error | Document conversion error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
document_conversion_operation_unavailable | Document conversion operation unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
document_conversion_provider_policy_invalid | Document conversion provider policy invalid. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
document_conversion_settlement_unavailable | Document conversion settlement unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
edge_error | The edge could not reach the application service. | Retry transient failures with bounded exponential backoff. If persistent, contact support. |
email_not_verified | Email not verified. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
email_taken | Email taken. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
empty_response | Empty response. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
empty_stream | Empty stream. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
error_catalog_unavailable | Error catalog unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
exception | Exception. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
expired_api_key | Expired api key. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
expired_client_secret | Expired client secret. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
fake_adapter_forbidden | Fake adapter forbidden. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
forbidden | Forbidden. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
free_token_grant_create_failed | Free token grant create failed. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
free_token_grant_update_failed | Free token grant update failed. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
from_address_domain_mismatch | From address domain mismatch. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
idempotency_conflict | Idempotency conflict. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
identity_cycle | Identity cycle. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
identity_execution_capacity | The selected execution identity has no available capacity. | Wait for current work to finish. Respect Retry-After and retryable/reallocate metadata; do not switch identities when reallocation is forbidden. |
internal_error | Internal error. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
invalid_api_key | Invalid api key. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
invalid_arguments | Invalid arguments. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_backend_response | Invalid backend response. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_client_secret | Invalid client secret. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_credentials | Invalid credentials. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
invalid_document_input | Invalid document input. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_domain | Invalid domain. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_email | Invalid email. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_event_type | Invalid event type. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_expires_after | Invalid expires after. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_from_address | Invalid from address. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_idempotency_key | Invalid idempotency key. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_image_count | Invalid image count. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_image_dimensions | Invalid image dimensions. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_input | Invalid input. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_json | Invalid json. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_lane_pin | Invalid lane pin. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_ocr_input | Invalid ocr input. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_provider | Invalid provider. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_provider_change_request | Invalid provider change request. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_provider_sdp | Invalid provider sdp. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_publication_request | Invalid publication request. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_realtime_mode | Invalid realtime mode. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_request | Invalid request. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_schema_version | Invalid schema version. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_sdp | Invalid sdp. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_session | Invalid session. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
invalid_terminal | Invalid terminal. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
invalid_token | Invalid token. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
invalid_upstream_call_id | Invalid upstream call id. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
invalid_workload_attribution | Invalid workload attribution. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
json_decode_error | Json decode error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
last_error_no_content | Last error no content. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
lease_expired_or_unknown | Lease expired or unknown. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
lease_mismatch | Lease mismatch. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
live_metric_divergence_unavailable | Live metric divergence unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
live_metric_health_unavailable | Live metric health unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
materializer_unavailable | Materializer unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
max_delivery_attempts_exceeded | Max delivery attempts exceeded. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
max_hops_exceeded | Max hops exceeded. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
method_not_allowed | Method not allowed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_api_key | Missing api key. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
missing_client_secret | Missing client secret. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_email | Missing email. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_events | Missing events. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_fields | Missing fields. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_model | Missing model. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_node | Missing node. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_prompt | Missing prompt. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_session_key | Missing session key. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
missing_sticky_key | Missing sticky key. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_terminal | Missing terminal. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_tool_call_id | Missing tool call id. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
missing_workspace | Missing workspace. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
model_claude_code_unauthorized | Model claude code unauthorized. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
model_does_not_support_tools | Model does not support tools. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
model_not_accessible | Model not accessible. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
model_not_allowed | Model not allowed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
model_not_available | Model not available. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
model_not_configured | Model not configured. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
model_not_found | Model not found. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
model_unavailable | Model unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
no_corporate_keys | No corporate keys. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
no_domain | No domain. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
no_providers_available | No providers available. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
no_reseller | No reseller. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
non_json_provider_response | Non json provider response. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
not_found | Not found. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
not_implemented | Not implemented. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
observer_unavailable | Observer unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
openrouter_http_error | Openrouter http error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
operation_in_progress | Operation in progress. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
operation_too_large | Operation too large. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
operation_unknown | Operation unknown. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
origin_context_unavailable | Origin context unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
origin_not_allowed | Origin not allowed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
origin_required | Origin required. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
payload_not_durable | Payload not durable. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
payment_required | Payment required. | Review the effective payer, balance, spending limits and payment configuration. Contact the billing administrator before retrying. |
permission_denied | Permission denied. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
pin_ambiguous_lane | Pin ambiguous lane. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
pin_forbidden | Pin forbidden. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
pin_no_lane | Pin no lane. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
pod_draining | Pod draining. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
preferred_node_unavailable | Preferred node unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
pricing_provenance_unavailable | Pricing provenance unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
promo_grant_create_failed | Promo grant create failed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
promo_grant_update_failed | Promo grant update failed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
provider_change_not_found | Provider change not found. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
provider_change_status_conflict | Provider change status conflict. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
provider_check_failed | Provider check failed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
provider_error | Provider error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
provider_missing_location | Provider missing location. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
provider_realtime_error | Provider realtime error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
provider_redirect_forbidden | Provider redirect forbidden. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
provider_response_too_large | Provider response too large. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
provider_route_mismatch | Provider route mismatch. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
provider_timeout | Provider timeout. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
provider_unavailable | Provider unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
proxy_event_ingest_disabled | Proxy event ingest disabled. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
publication_plan_conflict | Publication plan conflict. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
publication_plan_not_found | Publication plan not found. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
rate_limit_exceeded | Rate limit exceeded. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
rate_limit_state_unavailable | Rate limit state unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
raw_secret_field_rejected | Raw secret field rejected. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
realtime_call_recovery_required | A possibly created realtime call requires recovery. | Do not create a replacement call. Give support the request ID to recover or reconcile the original supplier call and its billing state. |
realtime_capacity_unavailable | Realtime capacity unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
realtime_keyring_malformed | Realtime keyring malformed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
realtime_keyring_unavailable | Realtime keyring unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
realtime_limit_exceeded | Realtime limit exceeded. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
realtime_service_unavailable | Realtime service unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
realtime_state_unavailable | Realtime state unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
realtime_transport_required | Realtime transport required. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
relay_error | Relay error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
replacement_required | Replacement required. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
request_exception | Request exception. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
request_rejected | Request rejected. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
request_timeout | Request timeout. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
request_too_large | Request too large. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
requester_bridge_async_completion_drain_failed | Requester bridge async completion drain failed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
requester_bridge_disabled | Requester bridge disabled. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
requester_bridge_required_unavailable | The required requester bridge is unavailable. | Have the service administrator restore the required bridge or choose a supported workflow. Do not repeatedly retry an unsupported route. |
responses_provider_unavailable | Responses provider unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
responses_terminal_finalization_failed | Responses terminal finalization failed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
responses_terminal_not_completed | Responses terminal not completed. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
result_too_large | Result too large. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
retryable | Retryable. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
secondary_only_evidence_cannot_stage | Secondary only evidence cannot stage. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
server_error | Server error. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
service_unavailable | Service unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
session_collision | Session collision. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
session_limit | Session limit. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
session_too_large | Session too large. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
settlement_pending | Document conversion succeeded and its durable zero-cost settlement is pending. | Keep the original request body and Idempotency-Key when checking or recovering the result. Do not use a new key to repeat conversion; contact support if settlement remains pending. |
sovereign_route_unavailable | Sovereign route unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
sovereign_route_validation_unavailable | Sovereign route validation unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
sse_bridge_capacity_exhausted | Sse bridge capacity exhausted. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
sso_only | Sso only. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
stale_reconciliation_action | Stale reconciliation action. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
stream_drop | Stream drop. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
stream_dropped | Stream dropped. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
stream_error | The response stream could not be completed. | Treat output as incomplete. Reconcile side effects before retrying non-idempotent work. |
stream_exception | Stream exception. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
stream_interrupted | The response stream stopped before completion. | Treat output as incomplete. Reconcile side effects before retrying non-idempotent work. |
streaming_settlement_buffer_exceeded | The private response buffer exceeded its settlement limit. | Have support reconcile the original operation before retrying; then reduce the response size or correct the authorized buffer configuration. |
streaming_settlement_cancelled | Exact streaming settlement was cancelled. | Cancellation does not prove the supplier did no work. Do not repeat the operation until support confirms its execution and billing status. |
streaming_settlement_failed | Exact streaming settlement did not complete. | Treat the operation as unconfirmed. Do not repeat it until support reconciles supplier execution and billing using the request ID. |
streaming_settlement_required | Streaming settlement required. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
strict_preferred_runtime_node_unavailable | Strict preferred runtime node unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
supplier_limit | Supplier limit. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
supplier_price_unresolved | The required supplier pricing authority could not be confirmed. | Have an administrator restore the exact supplier pricing configuration and check whether this request already executed before retrying. This code may occur before or after supplier execution. |
supplier_settlement_failed | Usage settlement could not be confirmed after possible supplier execution. | Do not repeat the operation. Give support the request ID to reconcile supplier execution, usage and any wallet debit before retrying. |
supplier_settlement_unavailable | Usage settlement is unavailable. | Do not automatically repeat the request. Have support confirm the original operation and billing status before authorizing a retry. |
timeout | Timeout. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
too_many_events | Too many events. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
tos_not_accepted | Tos not accepted. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
transcription_provider_unavailable | Transcription provider unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
unauthenticated | Unauthenticated. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
unauthorized | Unauthorized. | Check the named authentication or credential condition. Sign in again or have the account administrator correct it; do not loop retries with the same rejected credentials. |
unresolved_image_dimensions | Unresolved image dimensions. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
unsafe_realtime_upstream | Unsafe realtime upstream. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
unsupported_auto_mode | Unsupported auto mode. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
unsupported_autorouter_algorithm | Unsupported autorouter algorithm. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
unsupported_document_type | Unsupported document type. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
unsupported_image_model | Unsupported image model. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
unsupported_media_type | Unsupported media type. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
unsupported_ocr_adapter | Unsupported ocr adapter. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
unsupported_parameter | Unsupported parameter. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
unsupported_realtime_event | Unsupported realtime event. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
unsupported_realtime_model | Unsupported realtime model. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
upstream_closed_without_session | Upstream closed without session. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
upstream_error | Upstream error. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
upstream_failed | Upstream failed. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
upstream_invalid_response | Upstream invalid response. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
upstream_timeout | Upstream timeout. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
upstream_unavailable | Upstream unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
user_not_found | User not found. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
validation_error | Validation error. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
weak_password | Weak password. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
webchat_exchange_unavailable | Webchat exchange unavailable. | Respect Retry-After and any structured retryable metadata. Use bounded exponential backoff only when retry is appropriate; contact support if persistent. |
webchat_key_scope | Webchat key scope. | Check the explanation and required account, model, project or organisation policy. Correct the configuration or request the necessary access before retrying. |
webrtc_broker_not_enabled | Webrtc broker not enabled. | Read human_message and any param/detail fields, correct the request or resource state, and retry only after the condition changes. |
Returns the list of all models available to your account. Each model object includes its identifier, type, and owner. Use the id field as the model parameter in inference requests.
| Field | Type | Description |
|---|---|---|
id | string | Model identifier to use in API calls (e.g. claude-sonnet-4) |
object | string | Always "model" |
owned_by | string | Provider name (e.g. anthropic, openai) |
# List available models curl https://taas.cloudsigma.com/v1/models \ -H "Authorization: Bearer taas_xxxxxxxxxxxxxxxx"
import requests response = requests.get( "https://taas.cloudsigma.com/v1/models", headers={"Authorization": "Bearer taas_xxxxxxxxxxxxxxxx"} ) models = response.json() for m in models: print(m["id"]) # e.g. "claude-sonnet-4"
const response = await fetch( "https://taas.cloudsigma.com/v1/models", { headers: { "Authorization": "Bearer taas_xxxxxxxxxxxxxxxx" } } ); const models = await response.json(); models.forEach(m => console.log(m.id));
[
{
"id": "claude-sonnet-4",
"object": "model",
"owned_by": "anthropic"
},
{
"id": "gpt-4o",
"object": "model",
"owned_by": "openai"
},
{
"id": "minimax-m2",
"object": "model",
"owned_by": "minimax"
}
]
Generate a chat completion response from a model. Supports streaming via Server-Sent Events. Compatible with OpenAI's chat/completions API — existing OpenAI clients work by changing the base URL and API key.
| Field | Type | Description |
|---|---|---|
model required | string | Model ID from /v1/models (e.g. claude-sonnet-4) |
messages required | array | Array of {role, content} objects. Roles: system, user, assistant |
metadata.session_id | string | Optional explicit session id for continuity and easier per-request tracing. |
metadata.sticky_key | string | Optional sticky routing key. Used after X-Session-Id and metadata.session_id. |
stream | boolean | Stream tokens via SSE (default: false) |
temperature | number | Sampling temperature 0–2 (default: 1.0). Higher = more creative |
max_tokens | integer | Maximum tokens to generate |
top_p | number | Nucleus sampling probability 0–1 (default: 1.0) |
| Field | Type | Description |
|---|---|---|
choices[].message.role | string | Always "assistant" |
choices[].message.content | string | Generated text |
usage.prompt_tokens | integer | Tokens in the input messages |
usage.completion_tokens | integer | Tokens generated |
usage.total_tokens | integer | Sum of prompt + completion tokens |
# Chat completion with explicit session continuity curl -X POST https://taas.cloudsigma.com/v1/chat/completions \ -H "Authorization: Bearer taas_xxxxxxxxxxxxxxxx" \ -H "Content-Type: application/json" \ -H "X-Session-Id: user-42-chat-a" \ -d '{ "model": "claude-sonnet-4", "messages": [ {"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Explain quantum entanglement."} ], "metadata": { "session_id": "user-42-chat-a" }, "temperature": 0.7, "max_tokens": 512 }'
import requests response = requests.post( "https://taas.cloudsigma.com/v1/chat/completions", headers={"Authorization": "Bearer taas_xxxxxxxxxxxxxxxx"}, json={ "model": "claude-sonnet-4", "messages": [ {"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Explain quantum entanglement."} ], "temperature": 0.7, "max_tokens": 512 } ) result = response.json() print(result["choices"][0]["message"]["content"])
const response = await fetch( "https://taas.cloudsigma.com/v1/chat/completions", { method: "POST", headers: { "Authorization": "Bearer taas_xxxxxxxxxxxxxxxx", "Content-Type": "application/json", }, body: JSON.stringify({ model: "claude-sonnet-4", messages: [ { role: "system", content: "You are a helpful assistant." }, { role: "user", content: "Explain quantum entanglement." } ], temperature: 0.7, max_tokens: 512, }), } ); const data = await response.json(); console.log(data.choices[0].message.content);
{
"id": "chatcmpl-abc123",
"object": "chat.completion",
"model": "claude-sonnet-4",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "Quantum entanglement is a phenomenon..."
},
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 28,
"completion_tokens": 142,
"total_tokens": 170
}
}
Anthropic Messages API-compatible endpoint, running fully in parallel with /v1/chat/completions. Point the Anthropic SDK's base_url at 1Stack AI Mission Control and authenticate with your 1Stack AI Mission Control API key (sent as x-api-key or a Bearer token — both are accepted). Supports system prompts, tools with tool_use/tool_result blocks, images, and streaming with the full Anthropic SSE event sequence. Works with every model in the catalogue, including auto: requests to OpenAI-backed models are converted transparently in both directions, while Anthropic-backed models pass through natively. POST /v1/messages/count_tokens is also available.
curl -X POST https://taas.cloudsigma.com/v1/messages \
-H "x-api-key: $TAAS_API_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "content-type: application/json" \
-d '{"model": "auto", "max_tokens": 1024, "messages": [{"role": "user", "content": "Hello"}]}'
# Python (Anthropic SDK) — change only base_url and api_key
from anthropic import Anthropic
client = Anthropic(base_url="https://taas.cloudsigma.com", api_key=TAAS_API_KEY)
msg = client.messages.create(model="auto", max_tokens=1024,
messages=[{"role": "user", "content": "Hello"}])
Native Codex-compatible passthrough endpoint for Codex-specific clients such as Codex ACP. Requests sent here are routed to Codex OAuth tokens only and are forwarded without the Chat Completions → Codex transformation used by the OpenAI-compatible proxy path.
Important: this endpoint expects Codex-native request semantics.
In particular, input must be a list and stream must be set to true.
| Field | Type | Description |
|---|---|---|
model required | string | Codex-capable model id (for example gpt-5.4) |
instructions | string | Optional system/instructions text |
input required | array | Native Codex input message array |
stream required | boolean | Must be true for native Codex requests |
store | boolean | Passed through as provided |
Recommended for harnesses and agent runtimes that already speak Codex-native protocol. For ordinary OpenAI SDK traffic, keep using /v1/chat/completions and the standard OpenAI-compatible base URL.
curl -X POST https://taas.cloudsigma.com/v1/codex/responses \
-H "Authorization: Bearer taas_xxxxxxxxxxxxxxxx" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-5.4",
"store": false,
"stream": true,
"instructions": "Reply with exactly OK",
"input": [
{"role": "user", "content": "Reply with exactly OK"}
]
}'
Generate vector embeddings for one or more text inputs. Use embeddings for semantic search, similarity comparison, clustering, and retrieval-augmented generation (RAG).
| Field | Type | Description |
|---|---|---|
model required | string | Embedding model ID (e.g. bge-m3) |
input required | string or array | Text string or array of strings to embed. Max 2048 tokens per string. |
| Field | Type | Description |
|---|---|---|
data[].embedding | array | Dense float vector representing the input text |
data[].index | integer | Index of the input string this embedding corresponds to |
usage.prompt_tokens | integer | Total tokens processed |
usage.total_tokens | integer | Same as prompt_tokens for embeddings |
# Generate embeddings curl -X POST https://taas.cloudsigma.com/v1/embeddings \ -H "Authorization: Bearer taas_xxxxxxxxxxxxxxxx" \ -H "Content-Type: application/json" \ -d '{ "model": "bge-m3", "input": ["Hello world", "How are you?"] }'
import requests response = requests.post( "https://taas.cloudsigma.com/v1/embeddings", headers={"Authorization": "Bearer taas_xxxxxxxxxxxxxxxx"}, json={ "model": "bge-m3", "input": ["Hello world", "How are you?"] } ) data = response.json() vector = data["data"][0]["embedding"] print(f"Dimensions: {len(vector)}")
const response = await fetch( "https://taas.cloudsigma.com/v1/embeddings", { method: "POST", headers: { "Authorization": "Bearer taas_xxxxxxxxxxxxxxxx", "Content-Type": "application/json", }, body: JSON.stringify({ model: "bge-m3", input: ["Hello world", "How are you?"], }), } ); const data = await response.json(); console.log(`Dims: ${data.data[0].embedding.length}`);
{
"object": "list",
"model": "bge-m3",
"data": [
{
"object": "embedding",
"index": 0,
"embedding": [0.0234, -0.0871, 0.1203, ...]
},
{
"object": "embedding",
"index": 1,
"embedding": [0.0512, -0.0344, 0.0987, ...]
}
],
"usage": {
"prompt_tokens": 7,
"total_tokens": 7
}
}
Rerank a list of documents by relevance to a query. Ideal for improving retrieval quality in RAG pipelines — pass candidate documents from a vector search and get them sorted by true semantic relevance.
| Field | Type | Description |
|---|---|---|
model required | string | Reranker model ID (e.g. bge-reranker-v2-m3) |
query required | string | The search query to rank documents against |
documents required | array | Array of document strings to score and rank |
| Field | Type | Description |
|---|---|---|
results[].index | integer | Original index of the document in the input array |
results[].relevance_score | number | Relevance score (higher = more relevant). Results are sorted descending. |
# Rerank documents curl -X POST https://taas.cloudsigma.com/v1/rerank \ -H "Authorization: Bearer taas_xxxxxxxxxxxxxxxx" \ -H "Content-Type: application/json" \ -d '{ "model": "bge-reranker-v2-m3", "query": "What is quantum computing?", "documents": [ "Quantum computing uses qubits to process information.", "Classical computers use binary bits.", "The weather today is sunny." ] }'
import requests response = requests.post( "https://taas.cloudsigma.com/v1/rerank", headers={"Authorization": "Bearer taas_xxxxxxxxxxxxxxxx"}, json={ "model": "bge-reranker-v2-m3", "query": "What is quantum computing?", "documents": [ "Quantum computing uses qubits to process information.", "Classical computers use binary bits.", "The weather today is sunny." ] } ) results = response.json()["results"] for r in results: print(f"idx={r['index']} score={r['relevance_score']:.4f}")
const response = await fetch( "https://taas.cloudsigma.com/v1/rerank", { method: "POST", headers: { "Authorization": "Bearer taas_xxxxxxxxxxxxxxxx", "Content-Type": "application/json", }, body: JSON.stringify({ model: "bge-reranker-v2-m3", query: "What is quantum computing?", documents: [ "Quantum computing uses qubits to process information.", "Classical computers use binary bits.", "The weather today is sunny.", ], }), } ); const { results } = await response.json();
{
"model": "bge-reranker-v2-m3",
"results": [
{
"index": 0,
"relevance_score": 0.9823
},
{
"index": 1,
"relevance_score": 0.4512
},
{
"index": 2,
"relevance_score": 0.0031
}
]
}
Transcribe audio to text using Whisper or another speech-to-text model. Send the audio file as multipart/form-data. Supports WAV, MP3, M4A, FLAC, OGG, and WebM formats.
| Field | Type | Description |
|---|---|---|
file required | file | Audio file to transcribe. Max 25 MB. |
model | string | Model to use (default: whisper) |
language | string | ISO-639-1 language code (e.g. en, de). Auto-detected if omitted. |
| Field | Type | Description |
|---|---|---|
text | string | The transcribed text from the audio file |
# Transcribe audio file curl -X POST https://taas.cloudsigma.com/v1/audio/transcriptions \ -H "Authorization: Bearer taas_xxxxxxxxxxxxxxxx" \ -F "[email protected]" \ -F "model=whisper"
import requests with open("recording.wav", "rb") as f: response = requests.post( "https://taas.cloudsigma.com/v1/audio/transcriptions", headers={"Authorization": "Bearer taas_xxxxxxxxxxxxxxxx"}, files={"file": ("recording.wav", f, "audio/wav")}, data={"model": "whisper"} ) result = response.json() print(result["text"])
const formData = new FormData(); formData.append("file", audioBlob, "recording.wav"); formData.append("model", "whisper"); const response = await fetch( "https://taas.cloudsigma.com/v1/audio/transcriptions", { method: "POST", headers: { "Authorization": "Bearer taas_xxxxxxxxxxxxxxxx" }, body: formData, } ); const { text } = await response.json(); console.log(text);
{
"text": "Hello, this is a test recording for the transcription API."
}
Convert text to natural-sounding audio. The response is a binary audio stream (WAV or MP3). Use the Kokoro model for high-quality multilingual speech synthesis.
| Field | Type | Description |
|---|---|---|
model required | string | TTS model ID (e.g. kokoro) |
input required | string | Text to synthesize into speech. Max 4096 characters. |
voice | string | Voice identifier (model-specific). Omit for model default. |
response_format | string | Audio format: mp3 or wav (default: mp3) |
speed | number | Speaking speed multiplier 0.25–4.0 (default: 1.0) |
Returns raw audio binary data with Content-Type: audio/mpeg (MP3) or audio/wav. Save the response body directly to a file.
# Text to speech — save to file curl -X POST https://taas.cloudsigma.com/v1/audio/speech \ -H "Authorization: Bearer taas_xxxxxxxxxxxxxxxx" \ -H "Content-Type: application/json" \ -d '{ "model": "kokoro", "input": "Welcome to AI Mission Control.", "voice": "af_sarah" }' \ --output speech.mp3
import requests response = requests.post( "https://taas.cloudsigma.com/v1/audio/speech", headers={"Authorization": "Bearer taas_xxxxxxxxxxxxxxxx"}, json={ "model": "kokoro", "input": "Welcome to AI Mission Control.", "voice": "af_sarah" } ) with open("speech.mp3", "wb") as f: f.write(response.content) print("Saved speech.mp3")
const response = await fetch( "https://taas.cloudsigma.com/v1/audio/speech", { method: "POST", headers: { "Authorization": "Bearer taas_xxxxxxxxxxxxxxxx", "Content-Type": "application/json", }, body: JSON.stringify({ model: "kokoro", input: "Welcome to AI Mission Control.", voice: "af_sarah", }), } ); const audioBuffer = await response.arrayBuffer(); // Play or save the audio binary
# Binary audio data (MP3 or WAV)
# Content-Type: audio/mpeg
# Save response body directly to speech.mp3
Returns the current health status of the 1Stack AI Mission Control API gateway. Use this endpoint to verify connectivity and confirm the service is operational before sending inference requests. No authentication required.
| Field | Type | Description |
|---|---|---|
status | string | Always "ok" when the service is healthy |
# Check API health
curl https://taas.cloudsigma.com/health
import requests response = requests.get("https://taas.cloudsigma.com/health") print(response.json()) # {"status": "ok"}
const response = await fetch( "https://taas.cloudsigma.com/health" ); const data = await response.json(); console.log(data.status); // "ok"
{
"status": "ok"
}