--- title: "Update Application" method: PATCH path: "/v1/applications/{application_id}" tags: ["applications"] --- # Update Application `PATCH /v1/applications/{application_id}` ## Request - Base URL: `https://api.branchly.io` (the document declares no server; this is the origin it was published from) - URL: `https://api.branchly.io/v1/applications/{application_id}` - Auth: HTTP bearer ## Path parameters - `application_id` string, uuid, required ## Request body - ApplicationAPIUpdate - `name` string, nullable - `embed_location` string[], nullable - `organization_name` string, nullable - `primary_organization_id` string, nullable - `environment` 'production' | 'development' | 'disabled' - `search_output_sensitivity` 0.2 | 0.15 | 0.1 - `primary_edge_weight` 'weight' | 'weight_manual' | 'weight_rolling_30' | 'weight_rolling_60' | 'weight_rolling_alltime' - `secondary_edge_weight` 'weight' | 'weight_manual' | 'weight_rolling_30' | 'weight_rolling_60' | 'weight_rolling_alltime' - `valid_locales` string[], nullable - `search_result_highlight` 'page_metadata_description' | 'chunk' - `dense_model` 'all-MiniLM-L12-v2' | 'text-embedding-ada-002' | 'multilingual-e5-small' | 'text-embedding-3-large' | 'embeddinggemma-300m' - `sparse_model` 'bm25' - `search_mode` 'dense' | 'sparse' | 'hybrid' - `search_rerank_settings` SearchRerankSettingsInput - `datetime_reranking` DatetimeRerankSettings - `active_interfaces` string[]: Which user interfaces (or endpoints) rerank search results by datetime. - `datetime_field` 'metadata.page_metadata.article.published_time' | 'metadata.page_metadata.article.modified_time': The metadata field to use for datetime-based reranking. - `scale_in_days` integer: The scale for the time decay function in days. - `missing_datetime_scale_multiplier` number: Penalty strength for chunks missing the datetime_field. A default datetime is set to 'current_time - (scale_in_days * this_multiplier)'. Higher values push missing dates further into the past, reducing their scores. - `mode` 'sum' | 'mult': The method to combine the datetime-based score with the original score. - `record_source_reranking` IndexRecordSourceRerankSettings - `active_interfaces` string[]: Which user interfaces (or endpoints) rerank search results by chunk source. - `title_boost` number: The boost to apply to chunks from sources with a title match. - `title_fields` IndexRecordContentSource[]: The metadata fields to use for title matching. - `text_boost` number: The boost to apply to chunks from sources with a text match. - `text_fields` IndexRecordContentSource[]: The metadata fields to use for text matching. - `custom_boosting` CustomBoostSettings - `active_interfaces` string[]: Which user interfaces (or endpoints) apply custom boosting. - `boost_field` 'metadata.custom_metadata.score_boost' | 'metadata.score_boost': The metadata field to use for custom boosting. - `mode` 'sum' | 'mult': The method to combine the custom boost with the original score. - `modifier` 'log1p' | 'sqrt', nullable: A function to apply to the boost value before combining it with the score. 'log1p' computes log(1 + x) to handle zero values gracefully. - `session_personalization_settings` SessionPersonalizationSettings: Per-application session personalization config. Each element type has its own optional sub-config. - `suggested_questions` SuggestedQuestionsPersonalization - `enabled` boolean - `agent_settings` AgentSettings: Per-agent-type LLM configuration of an application. Agent types that are not configured (null) and agent types set to "default" use the platform default models. - `chat_routing` AgentLLMConfig: LLM configuration for a single agent type. The model is either a name from the platform model registry (app.api.common.llms.registry.MODEL_REGISTRY) or "default", which selects the platform default model for this agent type. The provider (Azure, OVHCloud, Google, ...) and all task-specific settings are derived from the registry, so nothing else is configurable here. - `model` union, required: Model name from the platform model registry, e.g. 'gpt-5-mini', or 'default' for the platform default model. - 'gpt-4.1-mini' | 'gpt-5-chat' | 'gpt-5-chat-2' | 'gpt-5-mini' | 'gpt-5.4-mini' | 'gpt-chat-latest' | 'gpt-5.6-terra' | 'gpt-5.6-luna' | 'gpt-6-luna' | 'gpt-6-sol' | 'mistral-large' | 'gpt-oss-120b' | 'Qwen3-32B' | 'Qwen3.5-9B' | 'Qwen3.5-397B-A17B' | 'Qwen3.6-27B' | 'Qwen3.8-27B' | 'gemini-2.5-flash-lite' | 'gemini-3.1-flash-lite' | 'gemini-3.5-flash-lite' | 'gemini-3.5-flash' | 'gemini-3.7-flash' | 'gemini-3.8-flash' | 'grok-4-1-fast-non-reasoning' | '@cf/google/gemma-4-26b-a4b-it' | '@cf/zai-org/glm-4.7-flash' | '@cf/moonshotai/kimi-k2.6' | 'Mistral-Small-4-119B-2603-Preview' | 'GLM-5.2-Preview' - 'default' - `chat_output` AgentLLMConfig: LLM configuration for a single agent type. The model is either a name from the platform model registry (app.api.common.llms.registry.MODEL_REGISTRY) or "default", which selects the platform default model for this agent type. The provider (Azure, OVHCloud, Google, ...) and all task-specific settings are derived from the registry, so nothing else is configurable here. - `model` union, required: Model name from the platform model registry, e.g. 'gpt-5-mini', or 'default' for the platform default model. - 'gpt-4.1-mini' | 'gpt-5-chat' | 'gpt-5-chat-2' | 'gpt-5-mini' | 'gpt-5.4-mini' | 'gpt-chat-latest' | 'gpt-5.6-terra' | 'gpt-5.6-luna' | 'gpt-6-luna' | 'gpt-6-sol' | 'mistral-large' | 'gpt-oss-120b' | 'Qwen3-32B' | 'Qwen3.5-9B' | 'Qwen3.5-397B-A17B' | 'Qwen3.6-27B' | 'Qwen3.8-27B' | 'gemini-2.5-flash-lite' | 'gemini-3.1-flash-lite' | 'gemini-3.5-flash-lite' | 'gemini-3.5-flash' | 'gemini-3.7-flash' | 'gemini-3.8-flash' | 'grok-4-1-fast-non-reasoning' | '@cf/google/gemma-4-26b-a4b-it' | '@cf/zai-org/glm-4.7-flash' | '@cf/moonshotai/kimi-k2.6' | 'Mistral-Small-4-119B-2603-Preview' | 'GLM-5.2-Preview' - 'default' - `search_answer` AgentLLMConfig: LLM configuration for a single agent type. The model is either a name from the platform model registry (app.api.common.llms.registry.MODEL_REGISTRY) or "default", which selects the platform default model for this agent type. The provider (Azure, OVHCloud, Google, ...) and all task-specific settings are derived from the registry, so nothing else is configurable here. - `model` union, required: Model name from the platform model registry, e.g. 'gpt-5-mini', or 'default' for the platform default model. - 'gpt-4.1-mini' | 'gpt-5-chat' | 'gpt-5-chat-2' | 'gpt-5-mini' | 'gpt-5.4-mini' | 'gpt-chat-latest' | 'gpt-5.6-terra' | 'gpt-5.6-luna' | 'gpt-6-luna' | 'gpt-6-sol' | 'mistral-large' | 'gpt-oss-120b' | 'Qwen3-32B' | 'Qwen3.5-9B' | 'Qwen3.5-397B-A17B' | 'Qwen3.6-27B' | 'Qwen3.8-27B' | 'gemini-2.5-flash-lite' | 'gemini-3.1-flash-lite' | 'gemini-3.5-flash-lite' | 'gemini-3.5-flash' | 'gemini-3.7-flash' | 'gemini-3.8-flash' | 'grok-4-1-fast-non-reasoning' | '@cf/google/gemma-4-26b-a4b-it' | '@cf/zai-org/glm-4.7-flash' | '@cf/moonshotai/kimi-k2.6' | 'Mistral-Small-4-119B-2603-Preview' | 'GLM-5.2-Preview' - 'default' - `form_answer` AgentLLMConfig: LLM configuration for a single agent type. The model is either a name from the platform model registry (app.api.common.llms.registry.MODEL_REGISTRY) or "default", which selects the platform default model for this agent type. The provider (Azure, OVHCloud, Google, ...) and all task-specific settings are derived from the registry, so nothing else is configurable here. - `model` union, required: Model name from the platform model registry, e.g. 'gpt-5-mini', or 'default' for the platform default model. - 'gpt-4.1-mini' | 'gpt-5-chat' | 'gpt-5-chat-2' | 'gpt-5-mini' | 'gpt-5.4-mini' | 'gpt-chat-latest' | 'gpt-5.6-terra' | 'gpt-5.6-luna' | 'gpt-6-luna' | 'gpt-6-sol' | 'mistral-large' | 'gpt-oss-120b' | 'Qwen3-32B' | 'Qwen3.5-9B' | 'Qwen3.5-397B-A17B' | 'Qwen3.6-27B' | 'Qwen3.8-27B' | 'gemini-2.5-flash-lite' | 'gemini-3.1-flash-lite' | 'gemini-3.5-flash-lite' | 'gemini-3.5-flash' | 'gemini-3.7-flash' | 'gemini-3.8-flash' | 'grok-4-1-fast-non-reasoning' | '@cf/google/gemma-4-26b-a4b-it' | '@cf/zai-org/glm-4.7-flash' | '@cf/moonshotai/kimi-k2.6' | 'Mistral-Small-4-119B-2603-Preview' | 'GLM-5.2-Preview' - 'default' - `form_routing` AgentLLMConfig: LLM configuration for a single agent type. The model is either a name from the platform model registry (app.api.common.llms.registry.MODEL_REGISTRY) or "default", which selects the platform default model for this agent type. The provider (Azure, OVHCloud, Google, ...) and all task-specific settings are derived from the registry, so nothing else is configurable here. - `model` union, required: Model name from the platform model registry, e.g. 'gpt-5-mini', or 'default' for the platform default model. - 'gpt-4.1-mini' | 'gpt-5-chat' | 'gpt-5-chat-2' | 'gpt-5-mini' | 'gpt-5.4-mini' | 'gpt-chat-latest' | 'gpt-5.6-terra' | 'gpt-5.6-luna' | 'gpt-6-luna' | 'gpt-6-sol' | 'mistral-large' | 'gpt-oss-120b' | 'Qwen3-32B' | 'Qwen3.5-9B' | 'Qwen3.5-397B-A17B' | 'Qwen3.6-27B' | 'Qwen3.8-27B' | 'gemini-2.5-flash-lite' | 'gemini-3.1-flash-lite' | 'gemini-3.5-flash-lite' | 'gemini-3.5-flash' | 'gemini-3.7-flash' | 'gemini-3.8-flash' | 'grok-4-1-fast-non-reasoning' | '@cf/google/gemma-4-26b-a4b-it' | '@cf/zai-org/glm-4.7-flash' | '@cf/moonshotai/kimi-k2.6' | 'Mistral-Small-4-119B-2603-Preview' | 'GLM-5.2-Preview' - 'default' - `enable_spellchecker` boolean, nullable - `prompt_persona` string, nullable - `output_instructions` string, nullable - `chat_strategy` 'standard_rag' | 'agent' - `classification_mode` 'active' | 'inactive' | 'active_no_suggestions' - `reply_in_user_language` boolean, nullable - `use_browser_locale` boolean, nullable - `follow_up_actions` boolean, nullable - `web_mcp_enabled` boolean, nullable - `routing_context_nodes` string[], nullable - `generation_context_nodes` string[], nullable - `deletion_policy_days` integer, nullable ## Response `200` Successful Response - ApplicationAPI - `id` string, uuid, required - `name` string, required - `embed_location` string[] - `organization_name` string, nullable - `primary_organization_id` string, nullable, required - `environment` 'production' | 'development' | 'disabled', required - `search_output_sensitivity` 0.2 | 0.15 | 0.1 - `primary_edge_weight` 'weight' | 'weight_manual' | 'weight_rolling_30' | 'weight_rolling_60' | 'weight_rolling_alltime' - `secondary_edge_weight` 'weight' | 'weight_manual' | 'weight_rolling_30' | 'weight_rolling_60' | 'weight_rolling_alltime' - `valid_locales` string[] - `search_result_highlight` 'page_metadata_description' | 'chunk' - `dense_model` 'all-MiniLM-L12-v2' | 'text-embedding-ada-002' | 'multilingual-e5-small' | 'text-embedding-3-large' | 'embeddinggemma-300m' - `sparse_model` 'bm25' - `search_mode` 'dense' | 'sparse' | 'hybrid' - `search_rerank_settings` SearchRerankSettingsOutput - `datetime_reranking` DatetimeRerankSettings - `active_interfaces` string[]: Which user interfaces (or endpoints) rerank search results by datetime. - `datetime_field` 'metadata.page_metadata.article.published_time' | 'metadata.page_metadata.article.modified_time': The metadata field to use for datetime-based reranking. - `scale_in_days` integer: The scale for the time decay function in days. - `missing_datetime_scale_multiplier` number: Penalty strength for chunks missing the datetime_field. A default datetime is set to 'current_time - (scale_in_days * this_multiplier)'. Higher values push missing dates further into the past, reducing their scores. - `mode` 'sum' | 'mult': The method to combine the datetime-based score with the original score. - `record_source_reranking` IndexRecordSourceRerankSettings - `active_interfaces` string[]: Which user interfaces (or endpoints) rerank search results by chunk source. - `title_boost` number: The boost to apply to chunks from sources with a title match. - `title_fields` IndexRecordContentSource[]: The metadata fields to use for title matching. - `text_boost` number: The boost to apply to chunks from sources with a text match. - `text_fields` IndexRecordContentSource[]: The metadata fields to use for text matching. - `custom_boosting` CustomBoostSettings - `active_interfaces` string[]: Which user interfaces (or endpoints) apply custom boosting. - `boost_field` 'metadata.custom_metadata.score_boost' | 'metadata.score_boost': The metadata field to use for custom boosting. - `mode` 'sum' | 'mult': The method to combine the custom boost with the original score. - `modifier` 'log1p' | 'sqrt', nullable: A function to apply to the boost value before combining it with the score. 'log1p' computes log(1 + x) to handle zero values gracefully. - `session_personalization_settings` SessionPersonalizationSettings: Per-application session personalization config. Each element type has its own optional sub-config. - `suggested_questions` SuggestedQuestionsPersonalization - `enabled` boolean - `agent_settings` AgentSettings: Per-agent-type LLM configuration of an application. Agent types that are not configured (null) and agent types set to "default" use the platform default models. - `chat_routing` AgentLLMConfig: LLM configuration for a single agent type. The model is either a name from the platform model registry (app.api.common.llms.registry.MODEL_REGISTRY) or "default", which selects the platform default model for this agent type. The provider (Azure, OVHCloud, Google, ...) and all task-specific settings are derived from the registry, so nothing else is configurable here. - `model` union, required: Model name from the platform model registry, e.g. 'gpt-5-mini', or 'default' for the platform default model. - 'gpt-4.1-mini' | 'gpt-5-chat' | 'gpt-5-chat-2' | 'gpt-5-mini' | 'gpt-5.4-mini' | 'gpt-chat-latest' | 'gpt-5.6-terra' | 'gpt-5.6-luna' | 'gpt-6-luna' | 'gpt-6-sol' | 'mistral-large' | 'gpt-oss-120b' | 'Qwen3-32B' | 'Qwen3.5-9B' | 'Qwen3.5-397B-A17B' | 'Qwen3.6-27B' | 'Qwen3.8-27B' | 'gemini-2.5-flash-lite' | 'gemini-3.1-flash-lite' | 'gemini-3.5-flash-lite' | 'gemini-3.5-flash' | 'gemini-3.7-flash' | 'gemini-3.8-flash' | 'grok-4-1-fast-non-reasoning' | '@cf/google/gemma-4-26b-a4b-it' | '@cf/zai-org/glm-4.7-flash' | '@cf/moonshotai/kimi-k2.6' | 'Mistral-Small-4-119B-2603-Preview' | 'GLM-5.2-Preview' - 'default' - `chat_output` AgentLLMConfig: LLM configuration for a single agent type. The model is either a name from the platform model registry (app.api.common.llms.registry.MODEL_REGISTRY) or "default", which selects the platform default model for this agent type. The provider (Azure, OVHCloud, Google, ...) and all task-specific settings are derived from the registry, so nothing else is configurable here. - `model` union, required: Model name from the platform model registry, e.g. 'gpt-5-mini', or 'default' for the platform default model. - 'gpt-4.1-mini' | 'gpt-5-chat' | 'gpt-5-chat-2' | 'gpt-5-mini' | 'gpt-5.4-mini' | 'gpt-chat-latest' | 'gpt-5.6-terra' | 'gpt-5.6-luna' | 'gpt-6-luna' | 'gpt-6-sol' | 'mistral-large' | 'gpt-oss-120b' | 'Qwen3-32B' | 'Qwen3.5-9B' | 'Qwen3.5-397B-A17B' | 'Qwen3.6-27B' | 'Qwen3.8-27B' | 'gemini-2.5-flash-lite' | 'gemini-3.1-flash-lite' | 'gemini-3.5-flash-lite' | 'gemini-3.5-flash' | 'gemini-3.7-flash' | 'gemini-3.8-flash' | 'grok-4-1-fast-non-reasoning' | '@cf/google/gemma-4-26b-a4b-it' | '@cf/zai-org/glm-4.7-flash' | '@cf/moonshotai/kimi-k2.6' | 'Mistral-Small-4-119B-2603-Preview' | 'GLM-5.2-Preview' - 'default' - `search_answer` AgentLLMConfig: LLM configuration for a single agent type. The model is either a name from the platform model registry (app.api.common.llms.registry.MODEL_REGISTRY) or "default", which selects the platform default model for this agent type. The provider (Azure, OVHCloud, Google, ...) and all task-specific settings are derived from the registry, so nothing else is configurable here. - `model` union, required: Model name from the platform model registry, e.g. 'gpt-5-mini', or 'default' for the platform default model. - 'gpt-4.1-mini' | 'gpt-5-chat' | 'gpt-5-chat-2' | 'gpt-5-mini' | 'gpt-5.4-mini' | 'gpt-chat-latest' | 'gpt-5.6-terra' | 'gpt-5.6-luna' | 'gpt-6-luna' | 'gpt-6-sol' | 'mistral-large' | 'gpt-oss-120b' | 'Qwen3-32B' | 'Qwen3.5-9B' | 'Qwen3.5-397B-A17B' | 'Qwen3.6-27B' | 'Qwen3.8-27B' | 'gemini-2.5-flash-lite' | 'gemini-3.1-flash-lite' | 'gemini-3.5-flash-lite' | 'gemini-3.5-flash' | 'gemini-3.7-flash' | 'gemini-3.8-flash' | 'grok-4-1-fast-non-reasoning' | '@cf/google/gemma-4-26b-a4b-it' | '@cf/zai-org/glm-4.7-flash' | '@cf/moonshotai/kimi-k2.6' | 'Mistral-Small-4-119B-2603-Preview' | 'GLM-5.2-Preview' - 'default' - `form_answer` AgentLLMConfig: LLM configuration for a single agent type. The model is either a name from the platform model registry (app.api.common.llms.registry.MODEL_REGISTRY) or "default", which selects the platform default model for this agent type. The provider (Azure, OVHCloud, Google, ...) and all task-specific settings are derived from the registry, so nothing else is configurable here. - `model` union, required: Model name from the platform model registry, e.g. 'gpt-5-mini', or 'default' for the platform default model. - 'gpt-4.1-mini' | 'gpt-5-chat' | 'gpt-5-chat-2' | 'gpt-5-mini' | 'gpt-5.4-mini' | 'gpt-chat-latest' | 'gpt-5.6-terra' | 'gpt-5.6-luna' | 'gpt-6-luna' | 'gpt-6-sol' | 'mistral-large' | 'gpt-oss-120b' | 'Qwen3-32B' | 'Qwen3.5-9B' | 'Qwen3.5-397B-A17B' | 'Qwen3.6-27B' | 'Qwen3.8-27B' | 'gemini-2.5-flash-lite' | 'gemini-3.1-flash-lite' | 'gemini-3.5-flash-lite' | 'gemini-3.5-flash' | 'gemini-3.7-flash' | 'gemini-3.8-flash' | 'grok-4-1-fast-non-reasoning' | '@cf/google/gemma-4-26b-a4b-it' | '@cf/zai-org/glm-4.7-flash' | '@cf/moonshotai/kimi-k2.6' | 'Mistral-Small-4-119B-2603-Preview' | 'GLM-5.2-Preview' - 'default' - `form_routing` AgentLLMConfig: LLM configuration for a single agent type. The model is either a name from the platform model registry (app.api.common.llms.registry.MODEL_REGISTRY) or "default", which selects the platform default model for this agent type. The provider (Azure, OVHCloud, Google, ...) and all task-specific settings are derived from the registry, so nothing else is configurable here. - `model` union, required: Model name from the platform model registry, e.g. 'gpt-5-mini', or 'default' for the platform default model. - 'gpt-4.1-mini' | 'gpt-5-chat' | 'gpt-5-chat-2' | 'gpt-5-mini' | 'gpt-5.4-mini' | 'gpt-chat-latest' | 'gpt-5.6-terra' | 'gpt-5.6-luna' | 'gpt-6-luna' | 'gpt-6-sol' | 'mistral-large' | 'gpt-oss-120b' | 'Qwen3-32B' | 'Qwen3.5-9B' | 'Qwen3.5-397B-A17B' | 'Qwen3.6-27B' | 'Qwen3.8-27B' | 'gemini-2.5-flash-lite' | 'gemini-3.1-flash-lite' | 'gemini-3.5-flash-lite' | 'gemini-3.5-flash' | 'gemini-3.7-flash' | 'gemini-3.8-flash' | 'grok-4-1-fast-non-reasoning' | '@cf/google/gemma-4-26b-a4b-it' | '@cf/zai-org/glm-4.7-flash' | '@cf/moonshotai/kimi-k2.6' | 'Mistral-Small-4-119B-2603-Preview' | 'GLM-5.2-Preview' - 'default' - `enable_spellchecker` boolean, nullable - `prompt_persona` string - `output_instructions` string - `chat_strategy` 'standard_rag' | 'agent' - `classification_mode` 'active' | 'inactive' | 'active_no_suggestions' - `reply_in_user_language` boolean - `use_browser_locale` boolean - `follow_up_actions` boolean, nullable - `web_mcp_enabled` boolean - `routing_context_nodes` string[] - `generation_context_nodes` string[] - `deletion_policy_days` integer: Days after which data from request tables is deleted - `created_at` string, date-time, required - `updated_at` string, date-time, required - `updated_by` string, nullable - `created_by` string, nullable ## Other responses - `422`: Validation Error --- [API](https://skmtc.dev/branchly/apis/branchly-io-api.md) · [All operations](https://skmtc.dev/branchly/apis/branchly-io-api/llms.txt) · [OpenAPI document](https://skmtc.dev/branchly/apis/branchly-io-api/revisions/9ffb024f8ec8?raw)