feat(agent/core): Add Anthropic Claude 3 support (#7085)

- feat(agent/core): Add `AnthropicProvider`
  - Add `ANTHROPIC_API_KEY` to .env.template and docs

  Notable differences in logic compared to `OpenAIProvider`:
  - Merges subsequent user messages in `AnthropicProvider._get_chat_completion_args`
  - Merges and extracts all system messages into `system` parameter in `AnthropicProvider._get_chat_completion_args`
  - Supports prefill; merges prefill content (if any) into generated response

- Prompt changes to improve compatibility with `AnthropicProvider`
  Anthropic has a slightly different API compared to OpenAI, and has much stricter input validation. E.g. Anthropic only supports a single `system` prompt, where OpenAI allows multiple `system` messages. Anthropic also forbids sequences of multiple `user` or `assistant` messages and requires that messages alternate between roles.
  - Move response format instruction from separate message into main system prompt
  - Fix clock message format
  - Add pre-fill to `OneShot` generated prompt

- refactor(agent/core): Tweak `model_providers.schema`
  - Simplify `ModelProviderUsage`
     - Remove attribute `total_tokens` as it is always equal to `prompt_tokens + completion_tokens`
     - Modify signature of `update_usage(..)`; no longer requires a full `ModelResponse` object as input
  - Improve `ModelProviderBudget`
     - Change type of attribute `usage` to `defaultdict[str, ModelProviderUsage]` -> allow per-model usage tracking
     - Modify signature of `update_usage_and_cost(..)`; no longer requires a full `ModelResponse` object as input
     - Allow `ModelProviderBudget` zero-argument instantiation
  - Fix type of `AssistantChatMessage.role` to match `ChatMessage.role` (str -> `ChatMessage.Role`)
  - Add shared attributes and constructor to `ModelProvider` base class
  - Add `max_output_tokens` parameter to `create_chat_completion` interface
  - Add pre-filling as a global feature
    - Add `prefill_response` field to `ChatPrompt` model
    - Add `prefill_response` parameter to `create_chat_completion` interface
  - Add `ChatModelProvider.get_available_models()` and remove `ApiManager`
  - Remove unused `OpenAIChatParser` typedef in openai.py
  - Remove redundant `budget` attribute definition on `OpenAISettings`
  - Remove unnecessary `usage` in `OpenAIProvider` > `default_settings` > `budget`

- feat(agent): Allow use of any available LLM provider through `MultiProvider`
  - Add `MultiProvider` (`model_providers.multi`)
  - Replace all references to / uses of `OpenAIProvider` with `MultiProvider`
  - Change type of `Config.smart_llm` and `Config.fast_llm` from `str` to `ModelName`

- feat(agent/core): Validate function call arguments in `create_chat_completion`
    - Add `validate_call` method to `CompletionModelFunction` in `model_providers.schema`
    - Add `validate_tool_calls` utility function in `model_providers.utils`
    - Add tool call validation step to `create_chat_completion` in `OpenAIProvider` and `AnthropicProvider`
    - Remove (now redundant) command argument validation logic in agent.py and models/command.py

- refactor(agent): Rename `get_openai_command_specs` to `function_specs_from_commands`

autogpts/autogpt/.env.template

@@ -2,8 +2,11 @@
 ### AutoGPT - GENERAL SETTINGS
 ################################################################################
 
-## OPENAI_API_KEY - OpenAI API Key (Example: my-openai-api-key)
+## OPENAI_API_KEY - OpenAI API Key (Example: sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx)
-OPENAI_API_KEY=your-openai-api-key
+# OPENAI_API_KEY=
+
+## ANTHROPIC_API_KEY - Anthropic API Key (Example: sk-ant-api03-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx)
+# ANTHROPIC_API_KEY=
 
 ## TELEMETRY_OPT_IN - Share telemetry on errors and other issues with the AutoGPT team, e.g. through Sentry.
 ##   This helps us to spot and solve problems earlier & faster. (Default: DISABLED)

autogpts/autogpt/agbenchmark_config/benchmarks.py

@@ -5,8 +5,7 @@ from pathlib import Path
 
 from autogpt.agent_manager.agent_manager import AgentManager
 from autogpt.agents.agent import Agent, AgentConfiguration, AgentSettings
-from autogpt.agents.prompt_strategies.one_shot import OneShotAgentPromptStrategy
+from autogpt.app.main import _configure_llm_provider, run_interaction_loop
-from autogpt.app.main import _configure_openai_provider, run_interaction_loop
 from autogpt.config import AIProfile, ConfigBuilder
 from autogpt.file_storage import FileStorageBackendName, get_storage
 from autogpt.logs.config import configure_logging
@@ -38,10 +37,6 @@ def bootstrap_agent(task: str, continuous_mode: bool) -> Agent:
         ai_goals=[task],
     )
 
-    agent_prompt_config = OneShotAgentPromptStrategy.default_configuration.copy(
-        deep=True
-    )
-    agent_prompt_config.use_functions_api = config.openai_functions
     agent_settings = AgentSettings(
         name=Agent.default_settings.name,
         agent_id=AgentManager.generate_id("AutoGPT-benchmark"),
@@ -53,7 +48,6 @@ def bootstrap_agent(task: str, continuous_mode: bool) -> Agent:
             allow_fs_access=not config.restrict_to_workspace,
             use_functions_api=config.openai_functions,
         ),
-        prompt_config=agent_prompt_config,
         history=Agent.default_settings.history.copy(deep=True),
     )
 
@@ -66,7 +60,7 @@ def bootstrap_agent(task: str, continuous_mode: bool) -> Agent:
 
     agent = Agent(
         settings=agent_settings,
-        llm_provider=_configure_openai_provider(config),
+        llm_provider=_configure_llm_provider(config),
         file_storage=file_storage,
         legacy_config=config,
     )

autogpts/autogpt/autogpt/agents/agent.py

@@ -19,7 +19,6 @@ from autogpt.components.event_history import EventHistoryComponent
 from autogpt.core.configuration import Configurable
 from autogpt.core.prompting import ChatPrompt
 from autogpt.core.resource.model_providers import (
-    AssistantChatMessage,
     AssistantFunctionCall,
     ChatMessage,
     ChatModelProvider,
@@ -27,7 +26,7 @@ from autogpt.core.resource.model_providers import (
 )
 from autogpt.core.runner.client_lib.logging.helpers import dump_prompt
 from autogpt.file_storage.base import FileStorage
-from autogpt.llm.providers.openai import get_openai_command_specs
+from autogpt.llm.providers.openai import function_specs_from_commands
 from autogpt.logs.log_cycle import (
     CURRENT_CONTEXT_FILE_NAME,
     NEXT_ACTION_FILE_NAME,
@@ -46,7 +45,6 @@ from autogpt.utils.exceptions import (
     AgentException,
     AgentTerminated,
     CommandExecutionError,
-    InvalidArgumentError,
     UnknownCommandError,
 )
 
@@ -104,7 +102,11 @@ class Agent(BaseAgent, Configurable[AgentSettings]):
         self.ai_profile = settings.ai_profile
         self.directives = settings.directives
         prompt_config = OneShotAgentPromptStrategy.default_configuration.copy(deep=True)
-        prompt_config.use_functions_api = settings.config.use_functions_api
+        prompt_config.use_functions_api = (
+            settings.config.use_functions_api
+            # Anthropic currently doesn't support tools + prefilling :(
+            and self.llm.provider_name != "anthropic"
+        )
         self.prompt_strategy = OneShotAgentPromptStrategy(prompt_config, logger)
         self.commands: list[Command] = []
 
@@ -172,7 +174,7 @@ class Agent(BaseAgent, Configurable[AgentSettings]):
             task=self.state.task,
             ai_profile=self.state.ai_profile,
             ai_directives=directives,
-            commands=get_openai_command_specs(self.commands),
+            commands=function_specs_from_commands(self.commands),
             include_os_info=self.legacy_config.execute_local_commands,
         )
 
@@ -202,12 +204,9 @@ class Agent(BaseAgent, Configurable[AgentSettings]):
         ] = await self.llm_provider.create_chat_completion(
             prompt.messages,
             model_name=self.llm.name,
-            completion_parser=self.parse_and_validate_response,
+            completion_parser=self.prompt_strategy.parse_response_content,
-            functions=(
+            functions=prompt.functions,
-                get_openai_command_specs(self.commands)
+            prefill_response=prompt.prefill_response,
-                if self.config.use_functions_api
-                else []
-            ),
         )
         result = response.parsed_result
 
@@ -223,28 +222,6 @@ class Agent(BaseAgent, Configurable[AgentSettings]):
 
         return result
 
-    def parse_and_validate_response(
-        self, llm_response: AssistantChatMessage
-    ) -> OneShotAgentActionProposal:
-        parsed_response = self.prompt_strategy.parse_response_content(llm_response)
-
-        # Validate command arguments
-        command_name = parsed_response.use_tool.name
-        command = self._get_command(command_name)
-        if arg_errors := command.validate_args(parsed_response.use_tool.arguments)[1]:
-            fmt_errors = [
-                f"{'.'.join(str(p) for p in f.path)}: {f.message}"
-                if f.path
-                else f.message
-                for f in arg_errors
-            ]
-            raise InvalidArgumentError(
-                f"The set of arguments supplied for {command_name} is invalid:\n"
-                + "\n".join(fmt_errors)
-            )
-
-        return parsed_response
-
     async def execute(
         self,
         proposal: OneShotAgentActionProposal,

autogpts/autogpt/autogpt/agents/base.py

@@ -39,11 +39,12 @@ from autogpt.core.configuration import (
     SystemSettings,
     UserConfigurable,
 )
-from autogpt.core.resource.model_providers import AssistantFunctionCall
+from autogpt.core.resource.model_providers import (
-from autogpt.core.resource.model_providers.openai import (
+    CHAT_MODELS,
-    OPEN_AI_CHAT_MODELS,
+    AssistantFunctionCall,
-    OpenAIModelName,
+    ModelName,
 )
+from autogpt.core.resource.model_providers.openai import OpenAIModelName
 from autogpt.models.utils import ModelWithSummary
 from autogpt.prompts.prompt import DEFAULT_TRIGGERING_PROMPT
 
@@ -56,8 +57,8 @@ P = ParamSpec("P")
 class BaseAgentConfiguration(SystemConfiguration):
     allow_fs_access: bool = UserConfigurable(default=False)
 
-    fast_llm: OpenAIModelName = UserConfigurable(default=OpenAIModelName.GPT3_16k)
+    fast_llm: ModelName = UserConfigurable(default=OpenAIModelName.GPT3_16k)
-    smart_llm: OpenAIModelName = UserConfigurable(default=OpenAIModelName.GPT4)
+    smart_llm: ModelName = UserConfigurable(default=OpenAIModelName.GPT4)
     use_functions_api: bool = UserConfigurable(default=False)
 
     default_cycle_instruction: str = DEFAULT_TRIGGERING_PROMPT
@@ -174,7 +175,7 @@ class BaseAgent(Configurable[BaseAgentSettings], metaclass=AgentMeta):
         llm_name = (
             self.config.smart_llm if self.config.big_brain else self.config.fast_llm
         )
-        return OPEN_AI_CHAT_MODELS[llm_name]
+        return CHAT_MODELS[llm_name]
 
     @property
     def send_token_limit(self) -> int:

autogpts/autogpt/autogpt/agents/prompt_strategies/one_shot.py

@@ -122,7 +122,7 @@ class OneShotAgentPromptStrategy(PromptStrategy):
         1. System prompt
         3. `cycle_instruction`
         """
-        system_prompt = self.build_system_prompt(
+        system_prompt, response_prefill = self.build_system_prompt(
             ai_profile=ai_profile,
             ai_directives=ai_directives,
             commands=commands,
@@ -131,24 +131,34 @@ class OneShotAgentPromptStrategy(PromptStrategy):
 
         final_instruction_msg = ChatMessage.user(self.config.choose_action_instruction)
 
-        prompt = ChatPrompt(
+        return ChatPrompt(
             messages=[
                 ChatMessage.system(system_prompt),
                 ChatMessage.user(f'"""{task}"""'),
                 *messages,
                 final_instruction_msg,
             ],
+            prefill_response=response_prefill,
+            functions=commands if self.config.use_functions_api else [],
         )
 
-        return prompt
-
     def build_system_prompt(
         self,
         ai_profile: AIProfile,
         ai_directives: AIDirectives,
         commands: list[CompletionModelFunction],
         include_os_info: bool,
-    ) -> str:
+    ) -> tuple[str, str]:
+        """
+        Builds the system prompt.
+
+        Returns:
+            str: The system prompt body
+            str: The desired start for the LLM's response; used to steer the output
+        """
+        response_fmt_instruction, response_prefill = self.response_format_instruction(
+            self.config.use_functions_api
+        )
         system_prompt_parts = (
             self._generate_intro_prompt(ai_profile)
             + (self._generate_os_info() if include_os_info else [])
@@ -169,16 +179,16 @@ class OneShotAgentPromptStrategy(PromptStrategy):
                 " in the next message. Your job is to complete the task while following"
                 " your directives as given above, and terminate when your task is done."
             ]
-            + [
+            + ["## RESPONSE FORMAT\n" + response_fmt_instruction]
-                "## RESPONSE FORMAT\n"
-                + self.response_format_instruction(self.config.use_functions_api)
-            ]
         )
 
         # Join non-empty parts together into paragraph format
-        return "\n\n".join(filter(None, system_prompt_parts)).strip("\n")
+        return (
+            "\n\n".join(filter(None, system_prompt_parts)).strip("\n"),
+            response_prefill,
+        )
 
-    def response_format_instruction(self, use_functions_api: bool) -> str:
+    def response_format_instruction(self, use_functions_api: bool) -> tuple[str, str]:
         response_schema = self.response_schema.copy(deep=True)
         if (
             use_functions_api
@@ -193,11 +203,15 @@ class OneShotAgentPromptStrategy(PromptStrategy):
             "\n",
             response_schema.to_typescript_object_interface(_RESPONSE_INTERFACE_NAME),
         )
+        response_prefill = f'{{\n    "{list(response_schema.properties.keys())[0]}":'
 
         return (
+            (
                 f"YOU MUST ALWAYS RESPOND WITH A JSON OBJECT OF THE FOLLOWING TYPE:\n"
                 f"{response_format}"
                 + ("\n\nYOU MUST ALSO INVOKE A TOOL!" if use_functions_api else "")
+            ),
+            response_prefill,
         )
 
     def _generate_intro_prompt(self, ai_profile: AIProfile) -> list[str]:

autogpts/autogpt/autogpt/app/agent_protocol_server.py

@@ -34,7 +34,6 @@ from autogpt.agent_manager import AgentManager
 from autogpt.app.utils import is_port_free
 from autogpt.config import Config
 from autogpt.core.resource.model_providers import ChatModelProvider, ModelProviderBudget
-from autogpt.core.resource.model_providers.openai import OpenAIProvider
 from autogpt.file_storage import FileStorage
 from autogpt.models.action_history import ActionErrorResult, ActionSuccessResult
 from autogpt.utils.exceptions import AgentFinished
@@ -464,20 +463,18 @@ class AgentProtocolServer:
         if task.additional_input and (user_id := task.additional_input.get("user_id")):
             _extra_request_headers["AutoGPT-UserID"] = user_id
 
-        task_llm_provider = None
-        if isinstance(self.llm_provider, OpenAIProvider):
         settings = self.llm_provider._settings.copy()
         settings.budget = task_llm_budget
-            settings.configuration = task_llm_provider_config  # type: ignore
+        settings.configuration = task_llm_provider_config
-            task_llm_provider = OpenAIProvider(
+        task_llm_provider = self.llm_provider.__class__(
             settings=settings,
-                logger=logger.getChild(f"Task-{task.task_id}_OpenAIProvider"),
+            logger=logger.getChild(
+                f"Task-{task.task_id}_{self.llm_provider.__class__.__name__}"
+            ),
         )
+        self._task_budgets[task.task_id] = task_llm_provider._budget  # type: ignore
 
-        if task_llm_provider and task_llm_provider._budget:
+        return task_llm_provider
-            self._task_budgets[task.task_id] = task_llm_provider._budget
-
-        return task_llm_provider or self.llm_provider
 
 
 def task_agent_id(task_id: str | int) -> str:

autogpts/autogpt/autogpt/app/configurator.py

@@ -10,7 +10,7 @@ from colorama import Back, Fore, Style
 
 from autogpt.config import Config
 from autogpt.config.config import GPT_3_MODEL, GPT_4_MODEL
-from autogpt.core.resource.model_providers.openai import OpenAIModelName, OpenAIProvider
+from autogpt.core.resource.model_providers import ModelName, MultiProvider
 from autogpt.logs.helpers import request_user_double_check
 from autogpt.memory.vector import get_supported_memory_backends
 from autogpt.utils import utils
@@ -150,11 +150,11 @@ async def apply_overrides_to_config(
 
 
 async def check_model(
-    model_name: OpenAIModelName, model_type: Literal["smart_llm", "fast_llm"]
+    model_name: ModelName, model_type: Literal["smart_llm", "fast_llm"]
-) -> OpenAIModelName:
+) -> ModelName:
     """Check if model is available for use. If not, return gpt-3.5-turbo."""
-    openai = OpenAIProvider()
+    multi_provider = MultiProvider()
-    models = await openai.get_available_models()
+    models = await multi_provider.get_available_models()
 
     if any(model_name == m.name for m in models):
         return model_name

autogpts/autogpt/autogpt/app/main.py

@@ -35,7 +35,7 @@ from autogpt.config import (
     ConfigBuilder,
     assert_config_has_openai_api_key,
 )
-from autogpt.core.resource.model_providers.openai import OpenAIProvider
+from autogpt.core.resource.model_providers import MultiProvider
 from autogpt.core.runner.client_lib.utils import coroutine
 from autogpt.file_storage import FileStorageBackendName, get_storage
 from autogpt.logs.config import configure_logging
@@ -123,7 +123,7 @@ async def run_auto_gpt(
         skip_news=skip_news,
     )
 
-    llm_provider = _configure_openai_provider(config)
+    llm_provider = _configure_llm_provider(config)
 
     logger = logging.getLogger(__name__)
 
@@ -399,7 +399,7 @@ async def run_auto_gpt_server(
         allow_downloads=allow_downloads,
     )
 
-    llm_provider = _configure_openai_provider(config)
+    llm_provider = _configure_llm_provider(config)
 
     # Set up & start server
     database = AgentDB(
@@ -421,24 +421,12 @@ async def run_auto_gpt_server(
     )
 
 
-def _configure_openai_provider(config: Config) -> OpenAIProvider:
+def _configure_llm_provider(config: Config) -> MultiProvider:
-    """Create a configured OpenAIProvider object.
+    multi_provider = MultiProvider()
-
+    for model in [config.smart_llm, config.fast_llm]:
-    Args:
+        # Ensure model providers for configured LLMs are available
-        config: The program's configuration.
+        multi_provider.get_model_provider(model)
-
+    return multi_provider
-    Returns:
-        A configured OpenAIProvider object.
-    """
-    if config.openai_credentials is None:
-        raise RuntimeError("OpenAI key is not configured")
-
-    openai_settings = OpenAIProvider.default_settings.copy(deep=True)
-    openai_settings.credentials = config.openai_credentials
-    return OpenAIProvider(
-        settings=openai_settings,
-        logger=logging.getLogger("OpenAIProvider"),
-    )
 
 
 def _get_cycle_budget(continuous_mode: bool, continuous_limit: int) -> int | float:

autogpts/autogpt/autogpt/commands/system.py

@@ -31,7 +31,9 @@ class SystemComponent(DirectiveProvider, MessageProvider, CommandProvider):
 
     def get_messages(self) -> Iterator[ChatMessage]:
         # Clock
-        yield ChatMessage.system(f"The current time and date is {time.strftime('%c')}")
+        yield ChatMessage.system(
+            f"## Clock\nThe current time and date is {time.strftime('%c')}"
+        )
 
     def get_commands(self) -> Iterator[Command]:
         yield self.finish

autogpts/autogpt/autogpt/config/config.py

@@ -17,8 +17,8 @@ from autogpt.core.configuration.schema import (
     SystemSettings,
     UserConfigurable,
 )
+from autogpt.core.resource.model_providers import CHAT_MODELS, ModelName
 from autogpt.core.resource.model_providers.openai import (
-    OPEN_AI_CHAT_MODELS,
     OpenAICredentials,
     OpenAIModelName,
 )
@@ -74,11 +74,11 @@ class Config(SystemSettings, arbitrary_types_allowed=True):
     )
 
     # Model configuration
-    fast_llm: OpenAIModelName = UserConfigurable(
+    fast_llm: ModelName = UserConfigurable(
         default=OpenAIModelName.GPT3,
         from_env="FAST_LLM",
     )
-    smart_llm: OpenAIModelName = UserConfigurable(
+    smart_llm: ModelName = UserConfigurable(
         default=OpenAIModelName.GPT4_TURBO,
         from_env="SMART_LLM",
     )
@@ -206,8 +206,8 @@ class Config(SystemSettings, arbitrary_types_allowed=True):
     def validate_openai_functions(cls, v: bool, values: dict[str, Any]):
         if v:
             smart_llm = values["smart_llm"]
-            assert OPEN_AI_CHAT_MODELS[smart_llm].has_function_call_api, (
+            assert CHAT_MODELS[smart_llm].has_function_call_api, (
-                f"Model {smart_llm} does not support OpenAI Functions. "
+                f"Model {smart_llm} does not support tool calling. "
                 "Please disable OPENAI_FUNCTIONS or choose a suitable model."
             )
         return v

autogpts/autogpt/autogpt/core/prompting/schema.py

@@ -24,6 +24,7 @@ class LanguageModelClassification(str, enum.Enum):
 class ChatPrompt(BaseModel):
     messages: list[ChatMessage]
     functions: list[CompletionModelFunction] = Field(default_factory=list)
+    prefill_response: str = ""
 
     def raw(self) -> list[ChatMessageDict]:
         return [m.dict() for m in self.messages]

autogpts/autogpt/autogpt/core/resource/model_providers/__init__.py

@@ -1,3 +1,4 @@
+from .multi import CHAT_MODELS, ModelName, MultiProvider
 from .openai import (
     OPEN_AI_CHAT_MODELS,
     OPEN_AI_EMBEDDING_MODELS,
@@ -42,11 +43,13 @@ __all__ = [
     "ChatModelProvider",
     "ChatModelResponse",
     "CompletionModelFunction",
+    "CHAT_MODELS",
     "Embedding",
     "EmbeddingModelInfo",
     "EmbeddingModelProvider",
     "EmbeddingModelResponse",
     "ModelInfo",
+    "ModelName",
     "ModelProvider",
     "ModelProviderBudget",
     "ModelProviderCredentials",
@@ -56,6 +59,7 @@ __all__ = [
     "ModelProviderUsage",
     "ModelResponse",
     "ModelTokenizer",
+    "MultiProvider",
     "OPEN_AI_MODELS",
     "OPEN_AI_CHAT_MODELS",
     "OPEN_AI_EMBEDDING_MODELS",

autogpts/autogpt/autogpt/core/resource/model_providers/anthropic.py

@@ -0,0 +1,495 @@
+from __future__ import annotations
+
+import enum
+import logging
+from typing import TYPE_CHECKING, Callable, Optional, ParamSpec, TypeVar
+
+import sentry_sdk
+import tenacity
+import tiktoken
+from anthropic import APIConnectionError, APIStatusError
+from pydantic import SecretStr
+
+from autogpt.core.configuration import Configurable, UserConfigurable
+from autogpt.core.resource.model_providers.schema import (
+    AssistantChatMessage,
+    AssistantFunctionCall,
+    AssistantToolCall,
+    ChatMessage,
+    ChatModelInfo,
+    ChatModelProvider,
+    ChatModelResponse,
+    CompletionModelFunction,
+    ModelProviderBudget,
+    ModelProviderConfiguration,
+    ModelProviderCredentials,
+    ModelProviderName,
+    ModelProviderSettings,
+    ModelTokenizer,
+    ToolResultMessage,
+)
+
+from .utils import validate_tool_calls
+
+if TYPE_CHECKING:
+    from anthropic.types.beta.tools import MessageCreateParams
+    from anthropic.types.beta.tools import ToolsBetaMessage as Message
+    from anthropic.types.beta.tools import ToolsBetaMessageParam as MessageParam
+
+_T = TypeVar("_T")
+_P = ParamSpec("_P")
+
+
+class AnthropicModelName(str, enum.Enum):
+    CLAUDE3_OPUS_v1 = "claude-3-opus-20240229"
+    CLAUDE3_SONNET_v1 = "claude-3-sonnet-20240229"
+    CLAUDE3_HAIKU_v1 = "claude-3-haiku-20240307"
+
+
+ANTHROPIC_CHAT_MODELS = {
+    info.name: info
+    for info in [
+        ChatModelInfo(
+            name=AnthropicModelName.CLAUDE3_OPUS_v1,
+            provider_name=ModelProviderName.ANTHROPIC,
+            prompt_token_cost=15 / 1e6,
+            completion_token_cost=75 / 1e6,
+            max_tokens=200000,
+            has_function_call_api=True,
+        ),
+        ChatModelInfo(
+            name=AnthropicModelName.CLAUDE3_SONNET_v1,
+            provider_name=ModelProviderName.ANTHROPIC,
+            prompt_token_cost=3 / 1e6,
+            completion_token_cost=15 / 1e6,
+            max_tokens=200000,
+            has_function_call_api=True,
+        ),
+        ChatModelInfo(
+            name=AnthropicModelName.CLAUDE3_HAIKU_v1,
+            provider_name=ModelProviderName.ANTHROPIC,
+            prompt_token_cost=0.25 / 1e6,
+            completion_token_cost=1.25 / 1e6,
+            max_tokens=200000,
+            has_function_call_api=True,
+        ),
+    ]
+}
+
+
+class AnthropicConfiguration(ModelProviderConfiguration):
+    fix_failed_parse_tries: int = UserConfigurable(3)
+
+
+class AnthropicCredentials(ModelProviderCredentials):
+    """Credentials for Anthropic."""
+
+    api_key: SecretStr = UserConfigurable(from_env="ANTHROPIC_API_KEY")
+    api_base: Optional[SecretStr] = UserConfigurable(
+        default=None, from_env="ANTHROPIC_API_BASE_URL"
+    )
+
+    def get_api_access_kwargs(self) -> dict[str, str]:
+        return {
+            k: (v.get_secret_value() if type(v) is SecretStr else v)
+            for k, v in {
+                "api_key": self.api_key,
+                "base_url": self.api_base,
+            }.items()
+            if v is not None
+        }
+
+
+class AnthropicSettings(ModelProviderSettings):
+    configuration: AnthropicConfiguration
+    credentials: Optional[AnthropicCredentials]
+    budget: ModelProviderBudget
+
+
+class AnthropicProvider(Configurable[AnthropicSettings], ChatModelProvider):
+    default_settings = AnthropicSettings(
+        name="anthropic_provider",
+        description="Provides access to Anthropic's API.",
+        configuration=AnthropicConfiguration(
+            retries_per_request=7,
+        ),
+        credentials=None,
+        budget=ModelProviderBudget(),
+    )
+
+    _settings: AnthropicSettings
+    _configuration: AnthropicConfiguration
+    _credentials: AnthropicCredentials
+    _budget: ModelProviderBudget
+
+    def __init__(
+        self,
+        settings: Optional[AnthropicSettings] = None,
+        logger: Optional[logging.Logger] = None,
+    ):
+        if not settings:
+            settings = self.default_settings.copy(deep=True)
+        if not settings.credentials:
+            settings.credentials = AnthropicCredentials.from_env()
+
+        super(AnthropicProvider, self).__init__(settings=settings, logger=logger)
+
+        from anthropic import AsyncAnthropic
+
+        self._client = AsyncAnthropic(**self._credentials.get_api_access_kwargs())
+
+    async def get_available_models(self) -> list[ChatModelInfo]:
+        return list(ANTHROPIC_CHAT_MODELS.values())
+
+    def get_token_limit(self, model_name: str) -> int:
+        """Get the token limit for a given model."""
+        return ANTHROPIC_CHAT_MODELS[model_name].max_tokens
+
+    @classmethod
+    def get_tokenizer(cls, model_name: AnthropicModelName) -> ModelTokenizer:
+        # HACK: No official tokenizer is available for Claude 3
+        return tiktoken.encoding_for_model(model_name)
+
+    @classmethod
+    def count_tokens(cls, text: str, model_name: AnthropicModelName) -> int:
+        return 0  # HACK: No official tokenizer is available for Claude 3
+
+    @classmethod
+    def count_message_tokens(
+        cls,
+        messages: ChatMessage | list[ChatMessage],
+        model_name: AnthropicModelName,
+    ) -> int:
+        return 0  # HACK: No official tokenizer is available for Claude 3
+
+    async def create_chat_completion(
+        self,
+        model_prompt: list[ChatMessage],
+        model_name: AnthropicModelName,
+        completion_parser: Callable[[AssistantChatMessage], _T] = lambda _: None,
+        functions: Optional[list[CompletionModelFunction]] = None,
+        max_output_tokens: Optional[int] = None,
+        prefill_response: str = "",
+        **kwargs,
+    ) -> ChatModelResponse[_T]:
+        """Create a completion using the Anthropic API."""
+        anthropic_messages, completion_kwargs = self._get_chat_completion_args(
+            prompt_messages=model_prompt,
+            model=model_name,
+            functions=functions,
+            max_output_tokens=max_output_tokens,
+            **kwargs,
+        )
+
+        total_cost = 0.0
+        attempts = 0
+        while True:
+            completion_kwargs["messages"] = anthropic_messages.copy()
+            if prefill_response:
+                completion_kwargs["messages"].append(
+                    {"role": "assistant", "content": prefill_response}
+                )
+
+            (
+                _assistant_msg,
+                cost,
+                t_input,
+                t_output,
+            ) = await self._create_chat_completion(completion_kwargs)
+            total_cost += cost
+            self._logger.debug(
+                f"Completion usage: {t_input} input, {t_output} output "
+                f"- ${round(cost, 5)}"
+            )
+
+            # Merge prefill into generated response
+            if prefill_response:
+                first_text_block = next(
+                    b for b in _assistant_msg.content if b.type == "text"
+                )
+                first_text_block.text = prefill_response + first_text_block.text
+
+            assistant_msg = AssistantChatMessage(
+                content="\n\n".join(
+                    b.text for b in _assistant_msg.content if b.type == "text"
+                ),
+                tool_calls=self._parse_assistant_tool_calls(_assistant_msg),
+            )
+
+            # If parsing the response fails, append the error to the prompt, and let the
+            # LLM fix its mistake(s).
+            attempts += 1
+            tool_call_errors = []
+            try:
+                # Validate tool calls
+                if assistant_msg.tool_calls and functions:
+                    tool_call_errors = validate_tool_calls(
+                        assistant_msg.tool_calls, functions
+                    )
+                    if tool_call_errors:
+                        raise ValueError(
+                            "Invalid tool use(s):\n"
+                            + "\n".join(str(e) for e in tool_call_errors)
+                        )
+
+                parsed_result = completion_parser(assistant_msg)
+                break
+            except Exception as e:
+                self._logger.debug(
+                    f"Parsing failed on response: '''{_assistant_msg}'''"
+                )
+                self._logger.warning(f"Parsing attempt #{attempts} failed: {e}")
+                sentry_sdk.capture_exception(
+                    error=e,
+                    extras={"assistant_msg": _assistant_msg, "i_attempt": attempts},
+                )
+                if attempts < self._configuration.fix_failed_parse_tries:
+                    anthropic_messages.append(
+                        _assistant_msg.dict(include={"role", "content"})
+                    )
+                    anthropic_messages.append(
+                        {
+                            "role": "user",
+                            "content": [
+                                *(
+                                    # tool_result is required if last assistant message
+                                    # had tool_use block(s)
+                                    {
+                                        "type": "tool_result",
+                                        "tool_use_id": tc.id,
+                                        "is_error": True,
+                                        "content": [
+                                            {
+                                                "type": "text",
+                                                "text": "Not executed because parsing "
+                                                "of your last message failed"
+                                                if not tool_call_errors
+                                                else str(e)
+                                                if (
+                                                    e := next(
+                                                        (
+                                                            tce
+                                                            for tce in tool_call_errors
+                                                            if tce.name
+                                                            == tc.function.name
+                                                        ),
+                                                        None,
+                                                    )
+                                                )
+                                                else "Not executed because validation "
+                                                "of tool input failed",
+                                            }
+                                        ],
+                                    }
+                                    for tc in assistant_msg.tool_calls or []
+                                ),
+                                {
+                                    "type": "text",
+                                    "text": (
+                                        "ERROR PARSING YOUR RESPONSE:\n\n"
+                                        f"{e.__class__.__name__}: {e}"
+                                    ),
+                                },
+                            ],
+                        }
+                    )
+                else:
+                    raise
+
+        if attempts > 1:
+            self._logger.debug(
+                f"Total cost for {attempts} attempts: ${round(total_cost, 5)}"
+            )
+
+        return ChatModelResponse(
+            response=assistant_msg,
+            parsed_result=parsed_result,
+            model_info=ANTHROPIC_CHAT_MODELS[model_name],
+            prompt_tokens_used=t_input,
+            completion_tokens_used=t_output,
+        )
+
+    def _get_chat_completion_args(
+        self,
+        prompt_messages: list[ChatMessage],
+        model: AnthropicModelName,
+        functions: Optional[list[CompletionModelFunction]] = None,
+        max_output_tokens: Optional[int] = None,
+        **kwargs,
+    ) -> tuple[list[MessageParam], MessageCreateParams]:
+        """Prepare arguments for message completion API call.
+
+        Args:
+            prompt_messages: List of ChatMessages.
+            model: The model to use.
+            functions: Optional list of functions available to the LLM.
+            kwargs: Additional keyword arguments.
+
+        Returns:
+            list[MessageParam]: Prompt messages for the Anthropic call
+            dict[str, Any]: Any other kwargs for the Anthropic call
+        """
+        kwargs["model"] = model
+
+        if functions:
+            kwargs["tools"] = [
+                {
+                    "name": f.name,
+                    "description": f.description,
+                    "input_schema": {
+                        "type": "object",
+                        "properties": {
+                            name: param.to_dict()
+                            for name, param in f.parameters.items()
+                        },
+                        "required": [
+                            name
+                            for name, param in f.parameters.items()
+                            if param.required
+                        ],
+                    },
+                }
+                for f in functions
+            ]
+
+        kwargs["max_tokens"] = max_output_tokens or 4096
+
+        if extra_headers := self._configuration.extra_request_headers:
+            kwargs["extra_headers"] = kwargs.get("extra_headers", {})
+            kwargs["extra_headers"].update(extra_headers.copy())
+
+        system_messages = [
+            m for m in prompt_messages if m.role == ChatMessage.Role.SYSTEM
+        ]
+        if (_n := len(system_messages)) > 1:
+            self._logger.warning(
+                f"Prompt has {_n} system messages; Anthropic supports only 1. "
+                "They will be merged, and removed from the rest of the prompt."
+            )
+        kwargs["system"] = "\n\n".join(sm.content for sm in system_messages)
+
+        messages: list[MessageParam] = []
+        for message in prompt_messages:
+            if message.role == ChatMessage.Role.SYSTEM:
+                continue
+            elif message.role == ChatMessage.Role.USER:
+                # Merge subsequent user messages
+                if messages and (prev_msg := messages[-1])["role"] == "user":
+                    if isinstance(prev_msg["content"], str):
+                        prev_msg["content"] += f"\n\n{message.content}"
+                    else:
+                        assert isinstance(prev_msg["content"], list)
+                        prev_msg["content"].append(
+                            {"type": "text", "text": message.content}
+                        )
+                else:
+                    messages.append({"role": "user", "content": message.content})
+                # TODO: add support for image blocks
+            elif message.role == ChatMessage.Role.ASSISTANT:
+                if isinstance(message, AssistantChatMessage) and message.tool_calls:
+                    messages.append(
+                        {
+                            "role": "assistant",
+                            "content": [
+                                *(
+                                    [{"type": "text", "text": message.content}]
+                                    if message.content
+                                    else []
+                                ),
+                                *(
+                                    {
+                                        "type": "tool_use",
+                                        "id": tc.id,
+                                        "name": tc.function.name,
+                                        "input": tc.function.arguments,
+                                    }
+                                    for tc in message.tool_calls
+                                ),
+                            ],
+                        }
+                    )
+                elif message.content:
+                    messages.append(
+                        {
+                            "role": "assistant",
+                            "content": message.content,
+                        }
+                    )
+            elif isinstance(message, ToolResultMessage):
+                messages.append(
+                    {
+                        "role": "user",
+                        "content": [
+                            {
+                                "type": "tool_result",
+                                "tool_use_id": message.tool_call_id,
+                                "content": [{"type": "text", "text": message.content}],
+                                "is_error": message.is_error,
+                            }
+                        ],
+                    }
+                )
+
+        return messages, kwargs  # type: ignore
+
+    async def _create_chat_completion(
+        self, completion_kwargs: MessageCreateParams
+    ) -> tuple[Message, float, int, int]:
+        """
+        Create a chat completion using the Anthropic API with retry handling.
+
+        Params:
+            completion_kwargs: Keyword arguments for an Anthropic Messages API call
+
+        Returns:
+            Message: The message completion object
+            float: The cost ($) of this completion
+            int: Number of input tokens used
+            int: Number of output tokens used
+        """
+
+        @self._retry_api_request
+        async def _create_chat_completion_with_retry(
+            completion_kwargs: MessageCreateParams,
+        ) -> Message:
+            return await self._client.beta.tools.messages.create(
+                **completion_kwargs  # type: ignore
+            )
+
+        response = await _create_chat_completion_with_retry(completion_kwargs)
+
+        cost = self._budget.update_usage_and_cost(
+            model_info=ANTHROPIC_CHAT_MODELS[completion_kwargs["model"]],
+            input_tokens_used=response.usage.input_tokens,
+            output_tokens_used=response.usage.output_tokens,
+        )
+        return response, cost, response.usage.input_tokens, response.usage.output_tokens
+
+    def _parse_assistant_tool_calls(
+        self, assistant_message: Message
+    ) -> list[AssistantToolCall]:
+        return [
+            AssistantToolCall(
+                id=c.id,
+                type="function",
+                function=AssistantFunctionCall(name=c.name, arguments=c.input),
+            )
+            for c in assistant_message.content
+            if c.type == "tool_use"
+        ]
+
+    def _retry_api_request(self, func: Callable[_P, _T]) -> Callable[_P, _T]:
+        return tenacity.retry(
+            retry=(
+                tenacity.retry_if_exception_type(APIConnectionError)
+                | tenacity.retry_if_exception(
+                    lambda e: isinstance(e, APIStatusError) and e.status_code >= 500
+                )
+            ),
+            wait=tenacity.wait_exponential(),
+            stop=tenacity.stop_after_attempt(self._configuration.retries_per_request),
+            after=tenacity.after_log(self._logger, logging.DEBUG),
+        )(func)
+
+    def __repr__(self):
+        return "AnthropicProvider()"

autogpts/autogpt/autogpt/core/resource/model_providers/multi.py

@@ -0,0 +1,162 @@
+from __future__ import annotations
+
+import logging
+from typing import Callable, Iterator, Optional, TypeVar
+
+from pydantic import ValidationError
+
+from autogpt.core.configuration import Configurable
+
+from .anthropic import ANTHROPIC_CHAT_MODELS, AnthropicModelName, AnthropicProvider
+from .openai import OPEN_AI_CHAT_MODELS, OpenAIModelName, OpenAIProvider
+from .schema import (
+    AssistantChatMessage,
+    ChatMessage,
+    ChatModelInfo,
+    ChatModelProvider,
+    ChatModelResponse,
+    CompletionModelFunction,
+    ModelProviderBudget,
+    ModelProviderConfiguration,
+    ModelProviderName,
+    ModelProviderSettings,
+    ModelTokenizer,
+)
+
+_T = TypeVar("_T")
+
+ModelName = AnthropicModelName | OpenAIModelName
+
+CHAT_MODELS = {**ANTHROPIC_CHAT_MODELS, **OPEN_AI_CHAT_MODELS}
+
+
+class MultiProvider(Configurable[ModelProviderSettings], ChatModelProvider):
+    default_settings = ModelProviderSettings(
+        name="multi_provider",
+        description=(
+            "Provides access to all of the available models, regardless of provider."
+        ),
+        configuration=ModelProviderConfiguration(
+            retries_per_request=7,
+        ),
+        budget=ModelProviderBudget(),
+    )
+
+    _budget: ModelProviderBudget
+
+    _provider_instances: dict[ModelProviderName, ChatModelProvider]
+
+    def __init__(
+        self,
+        settings: Optional[ModelProviderSettings] = None,
+        logger: Optional[logging.Logger] = None,
+    ):
+        super(MultiProvider, self).__init__(settings=settings, logger=logger)
+        self._budget = self._settings.budget or ModelProviderBudget()
+
+        self._provider_instances = {}
+
+    async def get_available_models(self) -> list[ChatModelInfo]:
+        models = []
+        for provider in self.get_available_providers():
+            models.extend(await provider.get_available_models())
+        return models
+
+    def get_token_limit(self, model_name: ModelName) -> int:
+        """Get the token limit for a given model."""
+        return self.get_model_provider(model_name).get_token_limit(model_name)
+
+    @classmethod
+    def get_tokenizer(cls, model_name: ModelName) -> ModelTokenizer:
+        return cls._get_model_provider_class(model_name).get_tokenizer(model_name)
+
+    @classmethod
+    def count_tokens(cls, text: str, model_name: ModelName) -> int:
+        return cls._get_model_provider_class(model_name).count_tokens(
+            text=text, model_name=model_name
+        )
+
+    @classmethod
+    def count_message_tokens(
+        cls, messages: ChatMessage | list[ChatMessage], model_name: ModelName
+    ) -> int:
+        return cls._get_model_provider_class(model_name).count_message_tokens(
+            messages=messages, model_name=model_name
+        )
+
+    async def create_chat_completion(
+        self,
+        model_prompt: list[ChatMessage],
+        model_name: ModelName,
+        completion_parser: Callable[[AssistantChatMessage], _T] = lambda _: None,
+        functions: Optional[list[CompletionModelFunction]] = None,
+        max_output_tokens: Optional[int] = None,
+        prefill_response: str = "",
+        **kwargs,
+    ) -> ChatModelResponse[_T]:
+        """Create a completion using the Anthropic API."""
+        return await self.get_model_provider(model_name).create_chat_completion(
+            model_prompt=model_prompt,
+            model_name=model_name,
+            completion_parser=completion_parser,
+            functions=functions,
+            max_output_tokens=max_output_tokens,
+            prefill_response=prefill_response,
+            **kwargs,
+        )
+
+    def get_model_provider(self, model: ModelName) -> ChatModelProvider:
+        model_info = CHAT_MODELS[model]
+        return self._get_provider(model_info.provider_name)
+
+    def get_available_providers(self) -> Iterator[ChatModelProvider]:
+        for provider_name in ModelProviderName:
+            try:
+                yield self._get_provider(provider_name)
+            except Exception:
+                pass
+
+    def _get_provider(self, provider_name: ModelProviderName) -> ChatModelProvider:
+        _provider = self._provider_instances.get(provider_name)
+        if not _provider:
+            Provider = self._get_provider_class(provider_name)
+            settings = Provider.default_settings.copy(deep=True)
+            settings.budget = self._budget
+            settings.configuration.extra_request_headers.update(
+                self._settings.configuration.extra_request_headers
+            )
+            if settings.credentials is None:
+                try:
+                    Credentials = settings.__fields__["credentials"].type_
+                    settings.credentials = Credentials.from_env()
+                except ValidationError as e:
+                    raise ValueError(
+                        f"{provider_name} is unavailable: can't load credentials"
+                    ) from e
+
+            self._provider_instances[provider_name] = _provider = Provider(
+                settings=settings, logger=self._logger
+            )
+            _provider._budget = self._budget  # Object binding not preserved by Pydantic
+        return _provider
+
+    @classmethod
+    def _get_model_provider_class(
+        cls, model_name: ModelName
+    ) -> type[AnthropicProvider | OpenAIProvider]:
+        return cls._get_provider_class(CHAT_MODELS[model_name].provider_name)
+
+    @classmethod
+    def _get_provider_class(
+        cls, provider_name: ModelProviderName
+    ) -> type[AnthropicProvider | OpenAIProvider]:
+        try:
+            return {
+                ModelProviderName.ANTHROPIC: AnthropicProvider,
+                ModelProviderName.OPENAI: OpenAIProvider,
+            }[provider_name]
+        except KeyError:
+            raise ValueError(f"{provider_name} is not a known provider") from None
+
+    def __repr__(self):
+        return f"{self.__class__.__name__}()"

autogpts/autogpt/autogpt/core/resource/model_providers/openai.py

@@ -42,6 +42,8 @@ from autogpt.core.resource.model_providers.schema import (
 from autogpt.core.utils.json_schema import JSONSchema
 from autogpt.core.utils.json_utils import json_loads
 
+from .utils import validate_tool_calls
+
 _T = TypeVar("_T")
 _P = ParamSpec("_P")
 
@@ -298,6 +300,7 @@ class OpenAIProvider(
         budget=ModelProviderBudget(),
     )
 
+    _settings: OpenAISettings
     _configuration: OpenAIConfiguration
     _credentials: OpenAICredentials
     _budget: ModelProviderBudget
@@ -312,11 +315,7 @@ class OpenAIProvider(
         if not settings.credentials:
             settings.credentials = OpenAICredentials.from_env()
 
-        self._settings = settings
+        super(OpenAIProvider, self).__init__(settings=settings, logger=logger)
-
-        self._configuration = settings.configuration
-        self._credentials = settings.credentials
-        self._budget = settings.budget
 
         if self._credentials.api_type == "azure":
             from openai import AsyncAzureOpenAI
@@ -329,8 +328,6 @@ class OpenAIProvider(
 
             self._client = AsyncOpenAI(**self._credentials.get_api_access_kwargs())
 
-        self._logger = logger or logging.getLogger(__name__)
-
     async def get_available_models(self) -> list[ChatModelInfo]:
         _models = (await self._client.models.list()).data
         return [OPEN_AI_MODELS[m.id] for m in _models if m.id in OPEN_AI_MODELS]
@@ -398,6 +395,7 @@ class OpenAIProvider(
         completion_parser: Callable[[AssistantChatMessage], _T] = lambda _: None,
         functions: Optional[list[CompletionModelFunction]] = None,
         max_output_tokens: Optional[int] = None,
+        prefill_response: str = "",  # not supported by OpenAI
         **kwargs,
     ) -> ChatModelResponse[_T]:
         """Create a completion using the OpenAI API and parse it."""
@@ -432,6 +430,10 @@ class OpenAIProvider(
             )
             parse_errors += _errors
 
+            # Validate tool calls
+            if not parse_errors and tool_calls and functions:
+                parse_errors += validate_tool_calls(tool_calls, functions)
+
             assistant_msg = AssistantChatMessage(
                 content=_assistant_msg.content,
                 tool_calls=tool_calls or None,

autogpts/autogpt/autogpt/core/resource/model_providers/schema.py

@@ -1,8 +1,10 @@
 import abc
 import enum
+import logging
 import math
 from collections import defaultdict
 from typing import (
+    TYPE_CHECKING,
     Any,
     Callable,
     ClassVar,
@@ -28,6 +30,9 @@ from autogpt.core.resource.schema import (
 from autogpt.core.utils.json_schema import JSONSchema
 from autogpt.logs.utils import fmt_kwargs
 
+if TYPE_CHECKING:
+    from jsonschema import ValidationError
+
 
 class ModelProviderService(str, enum.Enum):
     """A ModelService describes what kind of service the model provides."""
@@ -39,6 +44,7 @@ class ModelProviderService(str, enum.Enum):
 
 class ModelProviderName(str, enum.Enum):
     OPENAI = "openai"
+    ANTHROPIC = "anthropic"
 
 
 class ChatMessage(BaseModel):
@@ -100,6 +106,12 @@ class AssistantChatMessage(ChatMessage):
     tool_calls: Optional[list[AssistantToolCall]] = None
 
 
+class ToolResultMessage(ChatMessage):
+    role: Literal[ChatMessage.Role.TOOL] = ChatMessage.Role.TOOL
+    is_error: bool = False
+    tool_call_id: str
+
+
 class AssistantChatMessageDict(TypedDict, total=False):
     role: str
     content: str
@@ -146,6 +158,30 @@ class CompletionModelFunction(BaseModel):
         )
         return f"{self.name}: {self.description}. Params: ({params})"
 
+    def validate_call(
+        self, function_call: AssistantFunctionCall
+    ) -> tuple[bool, list["ValidationError"]]:
+        """
+        Validates the given function call against the function's parameter specs
+
+        Returns:
+            bool: Whether the given set of arguments is valid for this command
+            list[ValidationError]: Issues with the set of arguments (if any)
+
+        Raises:
+            ValueError: If the function_call doesn't call this function
+        """
+        if function_call.name != self.name:
+            raise ValueError(
+                f"Can't validate {function_call.name} call using {self.name} spec"
+            )
+
+        params_schema = JSONSchema(
+            type=JSONSchema.Type.OBJECT,
+            properties={name: spec for name, spec in self.parameters.items()},
+        )
+        return params_schema.validate_object(function_call.arguments)
+
 
 class ModelInfo(BaseModel):
     """Struct for model information.
@@ -229,7 +265,7 @@ class ModelProviderBudget(ProviderBudget):
 class ModelProviderSettings(ProviderSettings):
     resource_type: ResourceType = ResourceType.MODEL
     configuration: ModelProviderConfiguration
-    credentials: ModelProviderCredentials
+    credentials: Optional[ModelProviderCredentials] = None
     budget: Optional[ModelProviderBudget] = None
 
 
@@ -238,9 +274,28 @@ class ModelProvider(abc.ABC):
 
     default_settings: ClassVar[ModelProviderSettings]
 
+    _settings: ModelProviderSettings
     _configuration: ModelProviderConfiguration
+    _credentials: Optional[ModelProviderCredentials] = None
     _budget: Optional[ModelProviderBudget] = None
 
+    _logger: logging.Logger
+
+    def __init__(
+        self,
+        settings: Optional[ModelProviderSettings] = None,
+        logger: Optional[logging.Logger] = None,
+    ):
+        if not settings:
+            settings = self.default_settings.copy(deep=True)
+
+        self._settings = settings
+        self._configuration = settings.configuration
+        self._credentials = settings.credentials
+        self._budget = settings.budget
+
+        self._logger = logger or logging.getLogger(self.__module__)
+
     @abc.abstractmethod
     def count_tokens(self, text: str, model_name: str) -> int:
         ...
@@ -358,6 +413,7 @@ class ChatModelProvider(ModelProvider):
         completion_parser: Callable[[AssistantChatMessage], _T] = lambda _: None,
         functions: Optional[list[CompletionModelFunction]] = None,
         max_output_tokens: Optional[int] = None,
+        prefill_response: str = "",
         **kwargs,
     ) -> ChatModelResponse[_T]:
         ...

autogpts/autogpt/autogpt/core/resource/model_providers/utils.py

@@ -0,0 +1,71 @@
+from typing import Any
+
+from .schema import AssistantToolCall, CompletionModelFunction
+
+
+class InvalidFunctionCallError(Exception):
+    def __init__(self, name: str, arguments: dict[str, Any], message: str):
+        self.message = message
+        self.name = name
+        self.arguments = arguments
+        super().__init__(message)
+
+    def __str__(self) -> str:
+        return f"Invalid function call for {self.name}: {self.message}"
+
+
+def validate_tool_calls(
+    tool_calls: list[AssistantToolCall], functions: list[CompletionModelFunction]
+) -> list[InvalidFunctionCallError]:
+    """
+    Validates a list of tool calls against a list of functions.
+
+    1. Tries to find a function matching each tool call
+    2. If a matching function is found, validates the tool call's arguments,
+    reporting any resulting errors
+    2. If no matching function is found, an error "Unknown function X" is reported
+    3. A list of all errors encountered during validation is returned
+
+    Params:
+        tool_calls: A list of tool calls to validate.
+        functions: A list of functions to validate against.
+
+    Returns:
+        list[InvalidFunctionCallError]: All errors encountered during validation.
+    """
+    errors: list[InvalidFunctionCallError] = []
+    for tool_call in tool_calls:
+        function_call = tool_call.function
+
+        if function := next(
+            (f for f in functions if f.name == function_call.name),
+            None,
+        ):
+            is_valid, validation_errors = function.validate_call(function_call)
+            if not is_valid:
+                fmt_errors = [
+                    f"{'.'.join(str(p) for p in f.path)}: {f.message}"
+                    if f.path
+                    else f.message
+                    for f in validation_errors
+                ]
+                errors.append(
+                    InvalidFunctionCallError(
+                        name=function_call.name,
+                        arguments=function_call.arguments,
+                        message=(
+                            "The set of arguments supplied is invalid:\n"
+                            + "\n".join(fmt_errors)
+                        ),
+                    )
+                )
+        else:
+            errors.append(
+                InvalidFunctionCallError(
+                    name=function_call.name,
+                    arguments=function_call.arguments,
+                    message=f"Unknown function {function_call.name}",
+                )
+            )
+
+    return errors

autogpts/autogpt/autogpt/llm/providers/openai.py

@@ -14,7 +14,7 @@ logger = logging.getLogger(__name__)
 T = TypeVar("T", bound=Callable)
 
 
-def get_openai_command_specs(
+def function_specs_from_commands(
     commands: Iterable[Command],
 ) -> list[CompletionModelFunction]:
     """Get OpenAI-consumable function specs for the agent's available commands.

autogpts/autogpt/autogpt/models/command.py

@@ -3,8 +3,6 @@ from __future__ import annotations
 import inspect
 from typing import Any, Callable
 
-from autogpt.core.utils.json_schema import JSONSchema
-
 from .command_parameter import CommandParameter
 from .context_item import ContextItem
 
@@ -42,20 +40,6 @@ class Command:
     def is_async(self) -> bool:
         return inspect.iscoroutinefunction(self.method)
 
-    def validate_args(self, args: dict[str, Any]):
-        """
-        Validates the given arguments against the command's parameter specifications
-
-        Returns:
-            bool: Whether the given set of arguments is valid for this command
-            list[ValidationError]: Issues with the set of arguments (if any)
-        """
-        params_schema = JSONSchema(
-            type=JSONSchema.Type.OBJECT,
-            properties={p.name: p.spec for p in self.parameters},
-        )
-        return params_schema.validate_object(args)
-
     def _parameters_match(
         self, func: Callable, parameters: list[CommandParameter]
     ) -> bool:

autogpts/autogpt/poetry.lock

@@ -167,6 +167,30 @@ files = [
 [package.dependencies]
 frozenlist = ">=1.1.0"
 
+[[package]]
+name = "anthropic"
+version = "0.25.1"
+description = "The official Python library for the anthropic API"
+optional = false
+python-versions = ">=3.7"
+files = [
+    {file = "anthropic-0.25.1-py3-none-any.whl", hash = "sha256:95d0cedc2a4b5beae3a78f9030aea4001caea5f46c6d263cce377c891c594e71"},
+    {file = "anthropic-0.25.1.tar.gz", hash = "sha256:0c01b30b77d041a8d07c532737bae69da58086031217150008e4541f52a64bd9"},
+]
+
+[package.dependencies]
+anyio = ">=3.5.0,<5"
+distro = ">=1.7.0,<2"
+httpx = ">=0.23.0,<1"
+pydantic = ">=1.9.0,<3"
+sniffio = "*"
+tokenizers = ">=0.13.0"
+typing-extensions = ">=4.7,<5"
+
+[package.extras]
+bedrock = ["boto3 (>=1.28.57)", "botocore (>=1.31.57)"]
+vertex = ["google-auth (>=2,<3)"]
+
 [[package]]
 name = "anyio"
 version = "4.2.0"
@@ -7234,4 +7258,4 @@ benchmark = ["agbenchmark"]
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.10"
-content-hash = "e6eab5c079d53f075ce701e86a2007e7ebeb635ac067d25f555bfea363bcc630"
+content-hash = "ad1e3c4706465733d04ddab975af630975bd528efce152c1da01eded53069eca"

autogpts/autogpt/pyproject.toml

@@ -22,6 +22,7 @@ serve = "autogpt.app.cli:serve"
 
 [tool.poetry.dependencies]
 python = "^3.10"
+anthropic = "^0.25.1"
 # autogpt-forge = { path = "../forge" }
 autogpt-forge = {git = "https://github.com/Significant-Gravitas/AutoGPT.git", subdirectory = "autogpts/forge"}
 beautifulsoup4 = "^4.12.2"

autogpts/autogpt/tests/conftest.py

@@ -8,9 +8,9 @@ import pytest
 from pytest_mock import MockerFixture
 
 from autogpt.agents.agent import Agent, AgentConfiguration, AgentSettings
-from autogpt.app.main import _configure_openai_provider
+from autogpt.app.main import _configure_llm_provider
 from autogpt.config import AIProfile, Config, ConfigBuilder
-from autogpt.core.resource.model_providers import ChatModelProvider, OpenAIProvider
+from autogpt.core.resource.model_providers import ChatModelProvider
 from autogpt.file_storage.local import (
     FileStorage,
     FileStorageConfiguration,
@@ -73,8 +73,8 @@ def setup_logger(config: Config):
 
 
 @pytest.fixture
-def llm_provider(config: Config) -> OpenAIProvider:
+def llm_provider(config: Config) -> ChatModelProvider:
-    return _configure_openai_provider(config)
+    return _configure_llm_provider(config)
 
 
 @pytest.fixture

autogpts/autogpt/tests/unit/test_config.py

@@ -14,7 +14,6 @@ from pydantic import SecretStr
 
 from autogpt.app.configurator import GPT_3_MODEL, GPT_4_MODEL, apply_overrides_to_config
 from autogpt.config import Config, ConfigBuilder
-from autogpt.core.resource.model_providers.openai import OpenAIModelName
 from autogpt.core.resource.model_providers.schema import (
     ChatModelInfo,
     ModelProviderName,
@@ -39,8 +38,8 @@ async def test_fallback_to_gpt3_if_gpt4_not_available(
     """
     Test if models update to gpt-3.5-turbo if gpt-4 is not available.
     """
-    config.fast_llm = OpenAIModelName.GPT4_TURBO
+    config.fast_llm = GPT_4_MODEL
-    config.smart_llm = OpenAIModelName.GPT4_TURBO
+    config.smart_llm = GPT_4_MODEL
 
     mock_list_models.return_value = asyncio.Future()
     mock_list_models.return_value.set_result(
@@ -56,8 +55,8 @@ async def test_fallback_to_gpt3_if_gpt4_not_available(
         gpt4only=False,
     )
 
-    assert config.fast_llm == "gpt-3.5-turbo"
+    assert config.fast_llm == GPT_3_MODEL
-    assert config.smart_llm == "gpt-3.5-turbo"
+    assert config.smart_llm == GPT_3_MODEL
 
 
 def test_missing_azure_config(config: Config) -> None:
@@ -148,8 +147,7 @@ def test_azure_config(config_with_azure: Config) -> None:
 @pytest.mark.asyncio
 async def test_create_config_gpt4only(config: Config) -> None:
     with mock.patch(
-        "autogpt.core.resource.model_providers.openai."
+        "autogpt.core.resource.model_providers.multi.MultiProvider.get_available_models"
-        "OpenAIProvider.get_available_models"
     ) as mock_get_models:
         mock_get_models.return_value = [
             ChatModelInfo(
@@ -169,8 +167,7 @@ async def test_create_config_gpt4only(config: Config) -> None:
 @pytest.mark.asyncio
 async def test_create_config_gpt3only(config: Config) -> None:
     with mock.patch(
-        "autogpt.core.resource.model_providers.openai."
+        "autogpt.core.resource.model_providers.multi.MultiProvider.get_available_models"
-        "OpenAIProvider.get_available_models"
     ) as mock_get_models:
         mock_get_models.return_value = [
             ChatModelInfo(

docs/content/AutoGPT/configuration/options.md

@@ -7,6 +7,7 @@ Configuration is controlled through the `Config` object. You can set configurati
 - `AI_SETTINGS_FILE`: Location of the AI Settings file relative to the AutoGPT root directory. Default: ai_settings.yaml
 - `AUDIO_TO_TEXT_PROVIDER`: Audio To Text Provider. Only option currently is `huggingface`. Default: huggingface
 - `AUTHORISE_COMMAND_KEY`: Key response accepted when authorising commands. Default: y
+- `ANTHROPIC_API_KEY`: Set this if you want to use Anthropic models with AutoGPT
 - `AZURE_CONFIG_FILE`: Location of the Azure Config file relative to the AutoGPT root directory. Default: azure.yaml
 - `BROWSE_CHUNK_MAX_LENGTH`: When browsing website, define the length of chunks to summarize. Default: 3000
 - `BROWSE_SPACY_LANGUAGE_MODEL`: [spaCy language model](https://spacy.io/usage/models) to use when creating chunks. Default: en_core_web_sm