diff --git a/docs/openapi_generator/generate.py b/docs/openapi_generator/generate.py
index a6fec5ca4..6195001a2 100644
--- a/docs/openapi_generator/generate.py
+++ b/docs/openapi_generator/generate.py
@@ -43,6 +43,9 @@ from llama_stack.apis.post_training import * # noqa: F403
from llama_stack.apis.reward_scoring import * # noqa: F403
from llama_stack.apis.synthetic_data_generation import * # noqa: F403
from llama_stack.apis.safety import * # noqa: F403
+from llama_stack.apis.models import * # noqa: F403
+from llama_stack.apis.memory_banks import * # noqa: F403
+from llama_stack.apis.shields import * # noqa: F403
class LlamaStack(
@@ -57,6 +60,9 @@ class LlamaStack(
PostTraining,
Memory,
Evaluations,
+ Models,
+ MemoryBanks,
+ Shields,
):
pass
diff --git a/docs/resources/llama-stack-spec.html b/docs/resources/llama-stack-spec.html
index 95b08d6ca..9932b67cc 100644
--- a/docs/resources/llama-stack-spec.html
+++ b/docs/resources/llama-stack-spec.html
@@ -21,7 +21,7 @@
"info": {
"title": "[DRAFT] Llama Stack Specification",
"version": "0.0.1",
- "description": "This is the specification of the llama stack that provides\n a set of endpoints and their corresponding interfaces that are tailored to\n best leverage Llama Models. The specification is still in draft and subject to change.\n Generated at 2024-09-23 01:08:55.758597"
+ "description": "This is the specification of the llama stack that provides\n a set of endpoints and their corresponding interfaces that are tailored to\n best leverage Llama Models. The specification is still in draft and subject to change.\n Generated at 2024-09-23 09:14:46.697401"
},
"servers": [
{
@@ -1202,7 +1202,7 @@
}
},
"tags": [
- "Memory"
+ "MemoryBanks"
],
"parameters": [
{
@@ -1225,6 +1225,96 @@
]
}
},
+ "/models/get": {
+ "get": {
+ "responses": {
+ "200": {
+ "description": "OK",
+ "content": {
+ "application/json": {
+ "schema": {
+ "oneOf": [
+ {
+ "$ref": "#/components/schemas/ModelServingSpec"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ }
+ }
+ }
+ }
+ },
+ "tags": [
+ "Models"
+ ],
+ "parameters": [
+ {
+ "name": "core_model_id",
+ "in": "query",
+ "required": true,
+ "schema": {
+ "type": "string"
+ }
+ },
+ {
+ "name": "X-LlamaStack-ProviderData",
+ "in": "header",
+ "description": "JSON-encoded provider data which will be made available to the adapter servicing the API",
+ "required": false,
+ "schema": {
+ "type": "string"
+ }
+ }
+ ]
+ }
+ },
+ "/shields/get": {
+ "get": {
+ "responses": {
+ "200": {
+ "description": "OK",
+ "content": {
+ "application/json": {
+ "schema": {
+ "oneOf": [
+ {
+ "$ref": "#/components/schemas/ShieldSpec"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ }
+ }
+ }
+ }
+ },
+ "tags": [
+ "Shields"
+ ],
+ "parameters": [
+ {
+ "name": "shield_type",
+ "in": "query",
+ "required": true,
+ "schema": {
+ "type": "string"
+ }
+ },
+ {
+ "name": "X-LlamaStack-ProviderData",
+ "in": "header",
+ "description": "JSON-encoded provider data which will be made available to the adapter servicing the API",
+ "required": false,
+ "schema": {
+ "type": "string"
+ }
+ }
+ ]
+ }
+ },
"/telemetry/get_trace": {
"get": {
"responses": {
@@ -1455,7 +1545,67 @@
}
},
"tags": [
- "Memory"
+ "MemoryBanks"
+ ],
+ "parameters": [
+ {
+ "name": "X-LlamaStack-ProviderData",
+ "in": "header",
+ "description": "JSON-encoded provider data which will be made available to the adapter servicing the API",
+ "required": false,
+ "schema": {
+ "type": "string"
+ }
+ }
+ ]
+ }
+ },
+ "/models/list": {
+ "get": {
+ "responses": {
+ "200": {
+ "description": "OK",
+ "content": {
+ "application/jsonl": {
+ "schema": {
+ "$ref": "#/components/schemas/ModelServingSpec"
+ }
+ }
+ }
+ }
+ },
+ "tags": [
+ "Models"
+ ],
+ "parameters": [
+ {
+ "name": "X-LlamaStack-ProviderData",
+ "in": "header",
+ "description": "JSON-encoded provider data which will be made available to the adapter servicing the API",
+ "required": false,
+ "schema": {
+ "type": "string"
+ }
+ }
+ ]
+ }
+ },
+ "/shields/list": {
+ "get": {
+ "responses": {
+ "200": {
+ "description": "OK",
+ "content": {
+ "application/jsonl": {
+ "schema": {
+ "$ref": "#/components/schemas/ShieldSpec"
+ }
+ }
+ }
+ }
+ },
+ "tags": [
+ "Shields"
],
"parameters": [
{
@@ -4321,6 +4471,111 @@
"job_uuid"
]
},
+ "Model": {
+ "description": "The model family and SKU of the model along with other parameters corresponding to the model."
+ },
+ "ModelServingSpec": {
+ "type": "object",
+ "properties": {
+ "llama_model": {
+ "$ref": "#/components/schemas/Model"
+ },
+ "provider_config": {
+ "type": "object",
+ "properties": {
+ "provider_id": {
+ "type": "string"
+ },
+ "config": {
+ "type": "object",
+ "additionalProperties": {
+ "oneOf": [
+ {
+ "type": "null"
+ },
+ {
+ "type": "boolean"
+ },
+ {
+ "type": "number"
+ },
+ {
+ "type": "string"
+ },
+ {
+ "type": "array"
+ },
+ {
+ "type": "object"
+ }
+ ]
+ }
+ }
+ },
+ "additionalProperties": false,
+ "required": [
+ "provider_id",
+ "config"
+ ]
+ }
+ },
+ "additionalProperties": false,
+ "required": [
+ "llama_model",
+ "provider_config"
+ ]
+ },
+ "ShieldSpec": {
+ "type": "object",
+ "properties": {
+ "shield_type": {
+ "type": "string"
+ },
+ "provider_config": {
+ "type": "object",
+ "properties": {
+ "provider_id": {
+ "type": "string"
+ },
+ "config": {
+ "type": "object",
+ "additionalProperties": {
+ "oneOf": [
+ {
+ "type": "null"
+ },
+ {
+ "type": "boolean"
+ },
+ {
+ "type": "number"
+ },
+ {
+ "type": "string"
+ },
+ {
+ "type": "array"
+ },
+ {
+ "type": "object"
+ }
+ ]
+ }
+ }
+ },
+ "additionalProperties": false,
+ "required": [
+ "provider_id",
+ "config"
+ ]
+ }
+ },
+ "additionalProperties": false,
+ "required": [
+ "shield_type",
+ "provider_config"
+ ]
+ },
"Trace": {
"type": "object",
"properties": {
@@ -5579,39 +5834,48 @@
}
],
"tags": [
- {
- "name": "PostTraining"
- },
- {
- "name": "Safety"
- },
- {
- "name": "SyntheticDataGeneration"
- },
- {
- "name": "Datasets"
- },
- {
- "name": "Telemetry"
- },
{
"name": "Evaluations"
},
{
- "name": "RewardScoring"
- },
- {
- "name": "Agents"
+ "name": "MemoryBanks"
},
{
"name": "Memory"
},
{
- "name": "BatchInference"
+ "name": "Datasets"
},
{
"name": "Inference"
},
+ {
+ "name": "BatchInference"
+ },
+ {
+ "name": "PostTraining"
+ },
+ {
+ "name": "SyntheticDataGeneration"
+ },
+ {
+ "name": "Models"
+ },
+ {
+ "name": "RewardScoring"
+ },
+ {
+ "name": "Telemetry"
+ },
+ {
+ "name": "Agents"
+ },
+ {
+ "name": "Safety"
+ },
+ {
+ "name": "Shields"
+ },
{
"name": "BuiltinTool",
"description": ""
@@ -5952,6 +6216,18 @@
"name": "EvaluationJobStatusResponse",
"description": ""
},
+ {
+ "name": "Model",
+ "description": "The model family and SKU of the model along with other parameters corresponding to the model.\n\n"
+ },
+ {
+ "name": "ModelServingSpec",
+ "description": ""
+ },
+ {
+ "name": "ShieldSpec",
+ "description": ""
+ },
{
"name": "Trace",
"description": ""
@@ -6115,9 +6391,12 @@
"Evaluations",
"Inference",
"Memory",
+ "MemoryBanks",
+ "Models",
"PostTraining",
"RewardScoring",
"Safety",
+ "Shields",
"SyntheticDataGeneration",
"Telemetry"
]
@@ -6191,6 +6470,8 @@
"MemoryRetrievalStep",
"MemoryToolDefinition",
"MetricEvent",
+ "Model",
+ "ModelServingSpec",
"OptimizerConfig",
"PhotogenToolDefinition",
"PostTrainingJob",
@@ -6217,6 +6498,7 @@
"SearchToolDefinition",
"Session",
"ShieldCallStep",
+ "ShieldSpec",
"SpanEndPayload",
"SpanStartPayload",
"SpanStatus",
diff --git a/docs/resources/llama-stack-spec.yaml b/docs/resources/llama-stack-spec.yaml
index d08a2a2c1..5a0b5393d 100644
--- a/docs/resources/llama-stack-spec.yaml
+++ b/docs/resources/llama-stack-spec.yaml
@@ -1230,6 +1230,37 @@ components:
- value
- unit
type: object
+ Model:
+ description: The model family and SKU of the model along with other parameters
+ corresponding to the model.
+ ModelServingSpec:
+ additionalProperties: false
+ properties:
+ llama_model:
+ $ref: '#/components/schemas/Model'
+ provider_config:
+ additionalProperties: false
+ properties:
+ config:
+ additionalProperties:
+ oneOf:
+ - type: 'null'
+ - type: boolean
+ - type: number
+ - type: string
+ - type: array
+ - type: object
+ type: object
+ provider_id:
+ type: string
+ required:
+ - provider_id
+ - config
+ type: object
+ required:
+ - llama_model
+ - provider_config
+ type: object
OptimizerConfig:
additionalProperties: false
properties:
@@ -1742,6 +1773,34 @@ components:
- step_id
- step_type
type: object
+ ShieldSpec:
+ additionalProperties: false
+ properties:
+ provider_config:
+ additionalProperties: false
+ properties:
+ config:
+ additionalProperties:
+ oneOf:
+ - type: 'null'
+ - type: boolean
+ - type: number
+ - type: string
+ - type: array
+ - type: object
+ type: object
+ provider_id:
+ type: string
+ required:
+ - provider_id
+ - config
+ type: object
+ shield_type:
+ type: string
+ required:
+ - shield_type
+ - provider_config
+ type: object
SpanEndPayload:
additionalProperties: false
properties:
@@ -2349,7 +2408,7 @@ info:
description: "This is the specification of the llama stack that provides\n \
\ a set of endpoints and their corresponding interfaces that are tailored\
\ to\n best leverage Llama Models. The specification is still in\
- \ draft and subject to change.\n Generated at 2024-09-23 01:08:55.758597"
+ \ draft and subject to change.\n Generated at 2024-09-23 09:14:46.697401"
title: '[DRAFT] Llama Stack Specification'
version: 0.0.1
jsonSchemaDialect: https://json-schema.org/draft/2020-12/schema
@@ -3078,7 +3137,7 @@ paths:
- type: 'null'
description: OK
tags:
- - Memory
+ - MemoryBanks
/memory/insert:
post:
parameters:
@@ -3118,7 +3177,7 @@ paths:
$ref: '#/components/schemas/MemoryBank'
description: OK
tags:
- - Memory
+ - MemoryBanks
/memory/query:
post:
parameters:
@@ -3165,6 +3224,51 @@ paths:
description: OK
tags:
- Memory
+ /models/get:
+ get:
+ parameters:
+ - in: query
+ name: core_model_id
+ required: true
+ schema:
+ type: string
+ - description: JSON-encoded provider data which will be made available to the
+ adapter servicing the API
+ in: header
+ name: X-LlamaStack-ProviderData
+ required: false
+ schema:
+ type: string
+ responses:
+ '200':
+ content:
+ application/json:
+ schema:
+ oneOf:
+ - $ref: '#/components/schemas/ModelServingSpec'
+ - type: 'null'
+ description: OK
+ tags:
+ - Models
+ /models/list:
+ get:
+ parameters:
+ - description: JSON-encoded provider data which will be made available to the
+ adapter servicing the API
+ in: header
+ name: X-LlamaStack-ProviderData
+ required: false
+ schema:
+ type: string
+ responses:
+ '200':
+ content:
+ application/jsonl:
+ schema:
+ $ref: '#/components/schemas/ModelServingSpec'
+ description: OK
+ tags:
+ - Models
/post_training/job/artifacts:
get:
parameters:
@@ -3377,6 +3481,51 @@ paths:
description: OK
tags:
- Safety
+ /shields/get:
+ get:
+ parameters:
+ - in: query
+ name: shield_type
+ required: true
+ schema:
+ type: string
+ - description: JSON-encoded provider data which will be made available to the
+ adapter servicing the API
+ in: header
+ name: X-LlamaStack-ProviderData
+ required: false
+ schema:
+ type: string
+ responses:
+ '200':
+ content:
+ application/json:
+ schema:
+ oneOf:
+ - $ref: '#/components/schemas/ShieldSpec'
+ - type: 'null'
+ description: OK
+ tags:
+ - Shields
+ /shields/list:
+ get:
+ parameters:
+ - description: JSON-encoded provider data which will be made available to the
+ adapter servicing the API
+ in: header
+ name: X-LlamaStack-ProviderData
+ required: false
+ schema:
+ type: string
+ responses:
+ '200':
+ content:
+ application/jsonl:
+ schema:
+ $ref: '#/components/schemas/ShieldSpec'
+ description: OK
+ tags:
+ - Shields
/synthetic_data_generation/generate:
post:
parameters:
@@ -3452,17 +3601,20 @@ security:
servers:
- url: http://any-hosted-llama-stack.com
tags:
-- name: PostTraining
-- name: Safety
-- name: SyntheticDataGeneration
-- name: Datasets
-- name: Telemetry
- name: Evaluations
-- name: RewardScoring
-- name: Agents
+- name: MemoryBanks
- name: Memory
-- name: BatchInference
+- name: Datasets
- name: Inference
+- name: BatchInference
+- name: PostTraining
+- name: SyntheticDataGeneration
+- name: Models
+- name: RewardScoring
+- name: Telemetry
+- name: Agents
+- name: Safety
+- name: Shields
- description:
name: BuiltinTool
- description:
name: EvaluationJobStatusResponse
+- description: 'The model family and SKU of the model along with other parameters
+ corresponding to the model.
+
+
+ '
+ name: Model
+- description:
+ name: ModelServingSpec
+- description:
+ name: ShieldSpec
- description:
name: Trace
- description: 'Checkpoint created during training runs
@@ -3862,9 +4025,12 @@ x-tagGroups:
- Evaluations
- Inference
- Memory
+ - MemoryBanks
+ - Models
- PostTraining
- RewardScoring
- Safety
+ - Shields
- SyntheticDataGeneration
- Telemetry
- name: Types
@@ -3935,6 +4101,8 @@ x-tagGroups:
- MemoryRetrievalStep
- MemoryToolDefinition
- MetricEvent
+ - Model
+ - ModelServingSpec
- OptimizerConfig
- PhotogenToolDefinition
- PostTrainingJob
@@ -3961,6 +4129,7 @@ x-tagGroups:
- SearchToolDefinition
- Session
- ShieldCallStep
+ - ShieldSpec
- SpanEndPayload
- SpanStartPayload
- SpanStatus