Add Fireworks AI as new model provider (#8428)

api/core/model_runtime/model_providers/_position.yaml

@@ -37,3 +37,4 @@
 - siliconflow
 - perfxcloud
 - zhinao
+- fireworks

api/core/model_runtime/model_providers/fireworks/_common.py

@@ -0,0 +1,52 @@
+from collections.abc import Mapping
+
+import openai
+
+from core.model_runtime.errors.invoke import (
+    InvokeAuthorizationError,
+    InvokeBadRequestError,
+    InvokeConnectionError,
+    InvokeError,
+    InvokeRateLimitError,
+    InvokeServerUnavailableError,
+)
+
+
+class _CommonFireworks:
+    def _to_credential_kwargs(self, credentials: Mapping) -> dict:
+        """
+        Transform credentials to kwargs for model instance
+
+        :param credentials:
+        :return:
+        """
+        credentials_kwargs = {
+            "api_key": credentials["fireworks_api_key"],
+            "base_url": "https://api.fireworks.ai/inference/v1",
+            "max_retries": 1,
+        }
+
+        return credentials_kwargs
+
+    @property
+    def _invoke_error_mapping(self) -> dict[type[InvokeError], list[type[Exception]]]:
+        """
+        Map model invoke error to unified error
+        The key is the error type thrown to the caller
+        The value is the error type thrown by the model,
+        which needs to be converted into a unified error type for the caller.
+
+        :return: Invoke error mapping
+        """
+        return {
+            InvokeConnectionError: [openai.APIConnectionError, openai.APITimeoutError],
+            InvokeServerUnavailableError: [openai.InternalServerError],
+            InvokeRateLimitError: [openai.RateLimitError],
+            InvokeAuthorizationError: [openai.AuthenticationError, openai.PermissionDeniedError],
+            InvokeBadRequestError: [
+                openai.BadRequestError,
+                openai.NotFoundError,
+                openai.UnprocessableEntityError,
+                openai.APIError,
+            ],
+        }

api/core/model_runtime/model_providers/fireworks/fireworks.py

@@ -0,0 +1,27 @@
+import logging
+
+from core.model_runtime.entities.model_entities import ModelType
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.__base.model_provider import ModelProvider
+
+logger = logging.getLogger(__name__)
+
+
+class FireworksProvider(ModelProvider):
+    def validate_provider_credentials(self, credentials: dict) -> None:
+        """
+        Validate provider credentials
+        if validate failed, raise exception
+
+        :param credentials: provider credentials, credentials form defined in `provider_credential_schema`.
+        """
+        try:
+            model_instance = self.get_model_instance(ModelType.LLM)
+            model_instance.validate_credentials(
+                model="accounts/fireworks/models/llama-v3p1-8b-instruct", credentials=credentials
+            )
+        except CredentialsValidateFailedError as ex:
+            raise ex
+        except Exception as ex:
+            logger.exception(f"{self.get_provider_schema().provider} credentials validate failed")
+            raise ex

api/core/model_runtime/model_providers/fireworks/fireworks.yaml

@@ -0,0 +1,29 @@
+provider: fireworks
+label:
+  zh_Hans: Fireworks AI
+  en_US: Fireworks AI
+icon_small:
+  en_US: icon_s_en.svg
+icon_large:
+  en_US: icon_l_en.svg
+background: "#FCFDFF"
+help:
+  title:
+    en_US: Get your API Key from Fireworks AI
+    zh_Hans: 从 Fireworks AI 获取 API Key
+  url:
+    en_US: https://fireworks.ai/account/api-keys
+supported_model_types:
+  - llm
+configurate_methods:
+  - predefined-model
+provider_credential_schema:
+  credential_form_schemas:
+    - variable: fireworks_api_key
+      label:
+        en_US: API Key
+      type: secret-input
+      required: true
+      placeholder:
+        zh_Hans: 在此输入您的 API Key
+        en_US: Enter your API Key

api/core/model_runtime/model_providers/fireworks/llm/_position.yaml

@@ -0,0 +1,16 @@
+- llama-v3p1-405b-instruct
+- llama-v3p1-70b-instruct
+- llama-v3p1-8b-instruct
+- llama-v3-70b-instruct
+- mixtral-8x22b-instruct
+- mixtral-8x7b-instruct
+- firefunction-v2
+- firefunction-v1
+- gemma2-9b-it
+- llama-v3-70b-instruct-hf
+- llama-v3-8b-instruct
+- llama-v3-8b-instruct-hf
+- mixtral-8x7b-instruct-hf
+- mythomax-l2-13b
+- phi-3-vision-128k-instruct
+- yi-large

api/core/model_runtime/model_providers/fireworks/llm/firefunction-v1.yaml

@@ -0,0 +1,46 @@
+model: accounts/fireworks/models/firefunction-v1
+label:
+  zh_Hans: Firefunction V1
+  en_US: Firefunction V1
+model_type: llm
+features:
+  - agent-thought
+  - tool-call
+model_properties:
+  mode: chat
+  context_size: 32768
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+  - name: max_tokens
+    use_template: max_tokens
+  - name: context_length_exceeded_behavior
+    default: None
+    label:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    help:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    type: string
+    options:
+      - None
+      - truncate
+      - error
+  - name: response_format
+    use_template: response_format
+pricing:
+  input: '0.5'
+  output: '0.5'
+  unit: '0.000001'
+  currency: USD

api/core/model_runtime/model_providers/fireworks/llm/firefunction-v2.yaml

@@ -0,0 +1,46 @@
+model: accounts/fireworks/models/firefunction-v2
+label:
+  zh_Hans: Firefunction V2
+  en_US: Firefunction V2
+model_type: llm
+features:
+  - agent-thought
+  - tool-call
+model_properties:
+  mode: chat
+  context_size: 8192
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+  - name: max_tokens
+    use_template: max_tokens
+  - name: context_length_exceeded_behavior
+    default: None
+    label:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    help:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    type: string
+    options:
+      - None
+      - truncate
+      - error
+  - name: response_format
+    use_template: response_format
+pricing:
+  input: '0.9'
+  output: '0.9'
+  unit: '0.000001'
+  currency: USD

api/core/model_runtime/model_providers/fireworks/llm/gemma2-9b-it.yaml

@@ -0,0 +1,45 @@
+model: accounts/fireworks/models/gemma2-9b-it
+label:
+  zh_Hans: Gemma2 9B Instruct
+  en_US: Gemma2 9B Instruct
+model_type: llm
+features:
+  - agent-thought
+model_properties:
+  mode: chat
+  context_size: 8192
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+  - name: max_tokens
+    use_template: max_tokens
+  - name: context_length_exceeded_behavior
+    default: None
+    label:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    help:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    type: string
+    options:
+      - None
+      - truncate
+      - error
+  - name: response_format
+    use_template: response_format
+pricing:
+  input: '0.2'
+  output: '0.2'
+  unit: '0.000001'
+  currency: USD

api/core/model_runtime/model_providers/fireworks/llm/llama-v3-70b-instruct-hf.yaml

@@ -0,0 +1,46 @@
+model: accounts/fireworks/models/llama-v3-70b-instruct-hf
+label:
+  zh_Hans: Llama3 70B Instruct(HF version)
+  en_US: Llama3 70B Instruct(HF version)
+model_type: llm
+features:
+  - agent-thought
+  - tool-call
+model_properties:
+  mode: chat
+  context_size: 8192
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+  - name: max_tokens
+    use_template: max_tokens
+  - name: context_length_exceeded_behavior
+    default: None
+    label:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    help:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    type: string
+    options:
+      - None
+      - truncate
+      - error
+  - name: response_format
+    use_template: response_format
+pricing:
+  input: '0.9'
+  output: '0.9'
+  unit: '0.000001'
+  currency: USD

api/core/model_runtime/model_providers/fireworks/llm/llama-v3-70b-instruct.yaml

@@ -0,0 +1,46 @@
+model: accounts/fireworks/models/llama-v3-70b-instruct
+label:
+  zh_Hans: Llama3 70B Instruct
+  en_US: Llama3 70B Instruct
+model_type: llm
+features:
+  - agent-thought
+  - tool-call
+model_properties:
+  mode: chat
+  context_size: 8192
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+  - name: max_tokens
+    use_template: max_tokens
+  - name: context_length_exceeded_behavior
+    default: None
+    label:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    help:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    type: string
+    options:
+      - None
+      - truncate
+      - error
+  - name: response_format
+    use_template: response_format
+pricing:
+  input: '0.9'
+  output: '0.9'
+  unit: '0.000001'
+  currency: USD