A valid request URL is required to generate request examples{
"name": "anthropic",
"network_config": {
"base_url": "<string>",
"extra_headers": {},
"default_request_timeout_in_seconds": 123,
"max_retries": 123,
"retry_backoff_initial": 123,
"retry_backoff_max": 123,
"insecure_skip_verify": true,
"ca_cert_pem": "<string>",
"allow_private_network": false
},
"concurrency_and_buffer_size": {
"concurrency": 123,
"buffer_size": 123
},
"proxy_config": {
"type": "none",
"url": "<string>",
"username": "<string>",
"password": "<string>",
"ca_cert_pem": "<string>"
},
"send_back_raw_request": true,
"send_back_raw_response": true,
"store_raw_request_response": true,
"custom_provider_config": {
"is_key_less": true,
"does_not_send_done_marker": true,
"base_provider_type": "anthropic",
"allowed_requests": {
"list_models": true,
"text_completion": true,
"text_completion_stream": true,
"chat_completion": true,
"chat_completion_stream": true,
"responses": true,
"responses_stream": true,
"count_tokens": true,
"embedding": true,
"decisions": true,
"speech": true,
"speech_stream": true,
"transcription": true,
"transcription_stream": true,
"image_generation": true,
"image_generation_stream": true,
"batch_create": true,
"batch_list": true,
"batch_retrieve": true,
"batch_cancel": true,
"batch_results": true,
"file_upload": true,
"file_list": true,
"file_retrieve": true,
"file_delete": true,
"file_content": true
},
"request_path_overrides": {}
},
"prompt_cache": {
"auto_inject": true,
"ttl": "1h",
"cache_control_injection_points": [
{
"location": "message",
"role": "system",
"index": 123
}
]
},
"injected_tools": [
{
"name": "<string>",
"mcp_client_name": "<string>",
"tool_name": "<string>"
}
],
"provider_status": "active",
"status": "<string>",
"description": "<string>",
"config_hash": "<string>"
}{
"event_id": "<string>",
"type": "<string>",
"is_bifrost_error": true,
"status_code": 123,
"error": {
"type": "<string>",
"code": "<string>",
"message": "<string>",
"param": "<string>",
"event_id": "<string>"
},
"extra_fields": {
"provider": "anthropic",
"model_requested": "<string>",
"request_type": "<string>",
"error_type": "<string>",
"retry_after_ms": 150500
}
}{
"event_id": "<string>",
"type": "<string>",
"is_bifrost_error": true,
"status_code": 123,
"error": {
"type": "<string>",
"code": "<string>",
"message": "<string>",
"param": "<string>",
"event_id": "<string>"
},
"extra_fields": {
"provider": "anthropic",
"model_requested": "<string>",
"request_type": "<string>",
"error_type": "<string>",
"retry_after_ms": 150500
}
}{
"event_id": "<string>",
"type": "<string>",
"is_bifrost_error": true,
"status_code": 123,
"error": {
"type": "<string>",
"code": "<string>",
"message": "<string>",
"param": "<string>",
"event_id": "<string>"
},
"extra_fields": {
"provider": "anthropic",
"model_requested": "<string>",
"request_type": "<string>",
"error_type": "<string>",
"retry_after_ms": 150500
}
}Add a new provider
Adds a new provider with the specified configuration. Setting network_config.base_url or network_config.allow_private_network requires a genuinely authenticated admin session - it is refused with 403 if dashboard authentication is disabled or unconfigured, even though other management endpoints remain reachable in that state. allow_private_network is guarded even without a base URL, because it also governs which addresses key-level URLs (Ollama/SGL/VLLM) may reach. The same applies to settings that let a third party read credentials in flight: proxy_config.url, proxy_config.ca_cert_pem, network_config.ca_cert_pem, network_config.insecure_skip_verify, and any absolute (scheme and host) custom_provider_config.request_path_overrides value.
A valid request URL is required to generate request examples{
"name": "anthropic",
"network_config": {
"base_url": "<string>",
"extra_headers": {},
"default_request_timeout_in_seconds": 123,
"max_retries": 123,
"retry_backoff_initial": 123,
"retry_backoff_max": 123,
"insecure_skip_verify": true,
"ca_cert_pem": "<string>",
"allow_private_network": false
},
"concurrency_and_buffer_size": {
"concurrency": 123,
"buffer_size": 123
},
"proxy_config": {
"type": "none",
"url": "<string>",
"username": "<string>",
"password": "<string>",
"ca_cert_pem": "<string>"
},
"send_back_raw_request": true,
"send_back_raw_response": true,
"store_raw_request_response": true,
"custom_provider_config": {
"is_key_less": true,
"does_not_send_done_marker": true,
"base_provider_type": "anthropic",
"allowed_requests": {
"list_models": true,
"text_completion": true,
"text_completion_stream": true,
"chat_completion": true,
"chat_completion_stream": true,
"responses": true,
"responses_stream": true,
"count_tokens": true,
"embedding": true,
"decisions": true,
"speech": true,
"speech_stream": true,
"transcription": true,
"transcription_stream": true,
"image_generation": true,
"image_generation_stream": true,
"batch_create": true,
"batch_list": true,
"batch_retrieve": true,
"batch_cancel": true,
"batch_results": true,
"file_upload": true,
"file_list": true,
"file_retrieve": true,
"file_delete": true,
"file_content": true
},
"request_path_overrides": {}
},
"prompt_cache": {
"auto_inject": true,
"ttl": "1h",
"cache_control_injection_points": [
{
"location": "message",
"role": "system",
"index": 123
}
]
},
"injected_tools": [
{
"name": "<string>",
"mcp_client_name": "<string>",
"tool_name": "<string>"
}
],
"provider_status": "active",
"status": "<string>",
"description": "<string>",
"config_hash": "<string>"
}{
"event_id": "<string>",
"type": "<string>",
"is_bifrost_error": true,
"status_code": 123,
"error": {
"type": "<string>",
"code": "<string>",
"message": "<string>",
"param": "<string>",
"event_id": "<string>"
},
"extra_fields": {
"provider": "anthropic",
"model_requested": "<string>",
"request_type": "<string>",
"error_type": "<string>",
"retry_after_ms": 150500
}
}{
"event_id": "<string>",
"type": "<string>",
"is_bifrost_error": true,
"status_code": 123,
"error": {
"type": "<string>",
"code": "<string>",
"message": "<string>",
"param": "<string>",
"event_id": "<string>"
},
"extra_fields": {
"provider": "anthropic",
"model_requested": "<string>",
"request_type": "<string>",
"error_type": "<string>",
"retry_after_ms": 150500
}
}{
"event_id": "<string>",
"type": "<string>",
"is_bifrost_error": true,
"status_code": 123,
"error": {
"type": "<string>",
"code": "<string>",
"message": "<string>",
"param": "<string>",
"event_id": "<string>"
},
"extra_fields": {
"provider": "anthropic",
"model_requested": "<string>",
"request_type": "<string>",
"error_type": "<string>",
"retry_after_ms": 150500
}
}| Permission | Granted via |
|---|---|
ModelProvider:Create | RBAC role or API key scope |
Authorizations
Management API authentication for /api/* endpoints. Use the Authorization header
with Bearer <token>, where <token> is one of:
- a Bifrost management API key,
- a dashboard session token issued by
POST /api/session/login, - base64 of
<admin-username>:<admin-password>(legacy equivalent ofBasicAuth).
Virtual keys (sk-bf-*) and the x-api-key header are not accepted on management APIs -
the sole exception is GET /api/governance/virtual-keys/quota, which is virtual-key-only.
Authentication alone is not sufficient in Bifrost Enterprise: each operation page shows a
Required Permissions table (Resource:Operation, for example Dashboard:View) above
its Authorizations section, and the caller's RBAC role or management API key scopes must
include what it lists, otherwise the request is rejected with 403 Forbidden.
A local admin — authenticated with the admin password, or any caller on a deployment with dashboard auth disabled — bypasses these checks and can call every management endpoint.
OSS setup lock. On Bifrost OSS, while dashboard auth is not active (no admin account,
or auth disabled), every management endpoint except the public ones (/health,
/api/version, /api/session/is-auth-enabled, /api/session/login, ...) requires the
operator's setup token in the X-Bifrost-Setup-Token header, in place of Authorization.
The token is set with setup_token in config.json or the BIFROST_SETUP_TOKEN
environment variable. A missing header returns 401, a wrong token 403. The header
stops working once dashboard auth is enabled. The dashboard instead trades the token once
for an HttpOnly bifrost_setup_session cookie via POST /api/session/setup.
See Required permissions for how
permissions are derived and which endpoints are exempt.
Body
Add provider request. Keys are managed separately via /api/providers/{provider}/keys.
AI model provider identifier
anthropic, azure, bedrock, bedrock_mantle, cerebras, cohere, deepseek, gemini, groq, mistral, ollama, opencode-go, opencode-zen, openai, parasail, perplexity, sgl, vertex, openrouter, elevenlabs, huggingface, nebius, xai, replicate, vllm, runway, runware, fireworks, sarvam, wafer, databricks, cloudflare, typesafe Network configuration for provider connections
Show child attributes
Show child attributes
Concurrency settings
Show child attributes
Show child attributes
Proxy configuration
Show child attributes
Show child attributes
Custom provider configuration
Show child attributes
Show child attributes
Synthesize prompt-cache breakpoints for requests that carry none. Off by default. Requests that already carry their own cache markers are never modified.
Show child attributes
Show child attributes
MCP tools injected into every chat and responses request served by this provider and executed server side when the model calls them. The client never sees the injected tool call, only the final answer. Names must be unique within a provider.
Show child attributes
Show child attributes
Response
Provider added successfully
Provider configuration response
AI model provider identifier
anthropic, azure, bedrock, bedrock_mantle, cerebras, cohere, deepseek, gemini, groq, mistral, ollama, opencode-go, opencode-zen, openai, parasail, perplexity, sgl, vertex, openrouter, elevenlabs, huggingface, nebius, xai, replicate, vllm, runway, runware, fireworks, sarvam, wafer, databricks, cloudflare, typesafe Network configuration for provider connections
Show child attributes
Show child attributes
Concurrency settings
Show child attributes
Show child attributes
Proxy configuration
Show child attributes
Show child attributes
Custom provider configuration
Show child attributes
Show child attributes
Synthesize prompt-cache breakpoints for requests that carry none. Off by default. Requests that already carry their own cache markers are never modified.
Show child attributes
Show child attributes
MCP tools injected into every chat and responses request served by this provider and executed server side when the model calls them. The client never sees the injected tool call, only the final answer. Names must be unique within a provider.
Show child attributes
Show child attributes
Status of the provider
active, error, deleted Operational status (e.g., list_models_failed)
Error/status description
Hash of config.json version, used for change detection
Was this page helpful?