Fix LMAreana provider

g4f/Provider/Cloudflare.py

@@ -36,7 +36,7 @@ def clean_name(name: str) -> str:
 class Cloudflare(AsyncGeneratorProvider, ProviderModelMixin, AuthFileMixin):
     label = "Cloudflare AI"
     url = "https://playground.ai.cloudflare.com"
-    working = has_curl_cffi
+    working = False
     use_nodriver = True
     active_by_default = True
     api_endpoint = "https://playground.ai.cloudflare.com/api/inference"

g4f/Provider/needs_auth/LMArena.py

@@ -1,11 +1,12 @@
 from __future__ import annotations
 
-import uuid
 import json
 import asyncio
 import os
 import requests
 import json
+import time
+import secrets
 
 try:
     import curl_cffi
@@ -26,11 +27,26 @@ from ...requests import StreamSession, get_args_from_nodriver, raise_for_status,
 from ...errors import ModelNotFoundError, CloudflareError, MissingAuthError, MissingRequirementsError
 from ...providers.response import FinishReason, Usage, JsonConversation, ImageResponse, Reasoning, PlainTextResponse, JsonRequest
 from ...tools.media import merge_media
-from ...integration import uuid
 from ..base_provider import AsyncGeneratorProvider, ProviderModelMixin, AuthFileMixin
 from ..helper import get_last_user_message
 from ... import debug
 
+def uuid7():
+    """
+    Generate a UUIDv7 using Unix epoch (milliseconds since 1970-01-01)
+    matching the browser's implementation.
+    """
+    timestamp_ms = int(time.time() * 1000)
+    rand_a = secrets.randbits(12)
+    rand_b = secrets.randbits(62)
+    
+    uuid_int = timestamp_ms << 80
+    uuid_int |= (0x7000 | rand_a) << 64
+    uuid_int |= (0x8000000000000000 | rand_b)
+    
+    hex_str = f"{uuid_int:032x}"
+    return f"{hex_str[0:8]}-{hex_str[8:12]}-{hex_str[12:16]}-{hex_str[16:20]}-{hex_str[20:32]}"
+
 models = [
     {'id': '812c93cc-5f88-4cff-b9ca-c11a26599b0e', 'publicName': 'qwen3-max-preview',
      'capabilities': {'inputCapabilities': {'text': True}, 'outputCapabilities': {'text': True}},
@@ -485,7 +501,8 @@ class LMArena(AsyncGeneratorProvider, ProviderModelMixin, AuthFileMixin):
     label = "LMArena"
     url = "https://lmarena.ai"
     share_url = None
-    api_endpoint = "https://lmarena.ai/nextjs-api/stream/create-evaluation"
+    create_evaluation = "https://lmarena.ai/nextjs-api/stream/create-evaluation"
+    post_to_evaluation = "https://lmarena.ai/nextjs-api/stream/post-to-evaluation/{id}"
     working = True
     active_by_default = True
     use_stream_timeout = False
@@ -637,19 +654,21 @@ class LMArena(AsyncGeneratorProvider, ProviderModelMixin, AuthFileMixin):
             else:
                 raise ModelNotFoundError(f"Model '{model}' is not supported by LMArena provider.")
             
-            evaluationSessionId = str(uuid.uuid7())
-            userMessageId = str(uuid.uuid7())
-            modelAMessageId = str(uuid.uuid7())
+            if conversation and getattr(conversation, "evaluationSessionId", None):
+                url = cls.post_to_evaluation.format(id=conversation.evaluationSessionId)
+                evaluationSessionId = conversation.evaluationSessionId
+            else:
+                url = cls.create_evaluation
+                evaluationSessionId = str(uuid7())
+            userMessageId = str(uuid7())
+            modelAMessageId = str(uuid7())
             data = {
                 "id": evaluationSessionId,
                 "mode": "direct",
                 "modelAId": model_id,
                 "userMessageId": userMessageId,
                 "modelAMessageId": modelAMessageId,
-                "messages": [
-                    {
-                        "id": userMessageId,
-                        "role": "user",
+                "userMessage": {
                     "content": prompt,
                     "experimental_attachments": [
                             {
@@ -660,33 +679,14 @@ class LMArena(AsyncGeneratorProvider, ProviderModelMixin, AuthFileMixin):
                             for url, name in list(merge_media(media, messages))
                             if isinstance(url, str) and url.startswith("https://")
                         ],
-                        "parentMessageIds": [] if conversation is None else conversation.message_ids,
-                        "participantPosition": "a",
-                        "modelId": None,
-                        "evaluationSessionId": evaluationSessionId,
-                        "status": "pending",
-                        "failureReason": None
                 },
-                    {
-                        "id": modelAMessageId,
-                        "role": "assistant",
-                        "content": "",
-                        "experimental_attachments": [],
-                        "parentMessageIds": [userMessageId],
-                        "participantPosition": "a",
-                        "modelId": model,
-                        "evaluationSessionId": evaluationSessionId,
-                        "status": "pending",
-                        "failureReason": None
-                    }
-                ],
-                "modality": "image" if is_image_model else "chat"
+                "modality": "image" if is_image_model else "chat",
             }
             yield JsonRequest.from_dict(data)
             try:
                 async with StreamSession(**args, timeout=timeout) as session:
                     async with session.post(
-                            cls.api_endpoint,
+                            url,
                             json=data,
                             proxy=proxy
                     ) as response:
@@ -695,9 +695,7 @@ class LMArena(AsyncGeneratorProvider, ProviderModelMixin, AuthFileMixin):
                         async for chunk in response.iter_lines():
                             line = chunk.decode()
                             yield PlainTextResponse(line)
-                            if line.startswith("af:"):
-                                yield JsonConversation(message_ids=[modelAMessageId])
-                            elif line.startswith("a0:"):
+                            if line.startswith("a0:"):
                                 chunk = json.loads(line[3:])
                                 if chunk == "hasArenaError":
                                     raise ModelNotFoundError("LMArena Beta encountered an error: hasArenaError")
@@ -708,6 +706,7 @@ class LMArena(AsyncGeneratorProvider, ProviderModelMixin, AuthFileMixin):
                             elif line.startswith("a2:"):
                                 yield ImageResponse([image.get("image") for image in json.loads(line[3:])], prompt)
                             elif line.startswith("ad:"):
+                                yield JsonConversation(evaluationSessionId=evaluationSessionId)
                                 finish = json.loads(line[3:])
                                 if "finishReason" in finish:
                                     yield FinishReason(finish["finishReason"])

g4f/Provider/openai/har_file.py

@@ -9,8 +9,10 @@ import uuid
 import random
 from urllib.parse import unquote
 from copy import deepcopy
-
-from .crypt import decrypt, encrypt
+try:
+    from .crypt import decrypt, encrypt
+except ImportError:
+    pass
 from ...requests import StreamSession
 from ...cookies import get_cookies_dir
 from ...errors import NoValidHarFileError

g4f/integration/pydantic_ai.py

@@ -4,16 +4,18 @@ from typing import Optional
 from functools import partial
 from dataclasses import dataclass, field
 
-from pydantic_ai.models import Model, KnownModelName, infer_model
-from pydantic_ai.models.openai import OpenAIModel, OpenAISystemPromptRole
+from pydantic_ai import ModelResponsePart, ThinkingPart, ToolCallPart
+from pydantic_ai.models import Model, ModelResponse, KnownModelName, infer_model
+from pydantic_ai.models.openai import OpenAIChatModel, UnexpectedModelBehavior
+from pydantic_ai.models.openai import OpenAISystemPromptRole, _CHAT_FINISH_REASON_MAP, _map_usage, _now_utc, number_to_datetime, split_content_into_text_and_thinking, replace
 
 import pydantic_ai.models.openai
 pydantic_ai.models.openai.NOT_GIVEN = None
 
-from ..client import AsyncClient
+from ..client import AsyncClient, ChatCompletion
 
 @dataclass(init=False)
-class AIModel(OpenAIModel):
+class AIModel(OpenAIChatModel):
     """A model that uses the G4F API."""
 
     client: AsyncClient = field(repr=False)
@@ -54,6 +56,61 @@ class AIModel(OpenAIModel):
             return f'g4f:{self._provider}:{self._model_name}'
         return f'g4f:{self._model_name}'
     
+    def _process_response(self, response: ChatCompletion | str) -> ModelResponse:
+        """Process a non-streamed response, and prepare a message to return."""
+        # Although the OpenAI SDK claims to return a Pydantic model (`ChatCompletion`) from the chat completions function:
+        # * it hasn't actually performed validation (presumably they're creating the model with `model_construct` or something?!)
+        # * if the endpoint returns plain text, the return type is a string
+        # Thus we validate it fully here.
+        if not isinstance(response, ChatCompletion):
+            raise UnexpectedModelBehavior('Invalid response from OpenAI chat completions endpoint, expected JSON data')
+
+        if response.created:
+            timestamp = number_to_datetime(response.created)
+        else:
+            timestamp = _now_utc()
+            response.created = int(timestamp.timestamp())
+
+        # Workaround for local Ollama which sometimes returns a `None` finish reason.
+        if response.choices and (choice := response.choices[0]) and choice.finish_reason is None:  # pyright: ignore[reportUnnecessaryComparison]
+            choice.finish_reason = 'stop'
+
+        choice = response.choices[0]
+        items: list[ModelResponsePart] = []
+
+        # The `reasoning` field is only present in gpt-oss via Ollama and OpenRouter.
+        # - https://cookbook.openai.com/articles/gpt-oss/handle-raw-cot#chat-completions-api
+        # - https://openrouter.ai/docs/use-cases/reasoning-tokens#basic-usage-with-reasoning-tokens
+        if reasoning := getattr(choice.message, 'reasoning', None):
+            items.append(ThinkingPart(id='reasoning', content=reasoning, provider_name=self.system))
+
+        # NOTE: We don't currently handle OpenRouter `reasoning_details`:
+        # - https://openrouter.ai/docs/use-cases/reasoning-tokens#preserving-reasoning-blocks
+        # If you need this, please file an issue.
+
+        if choice.message.content:
+            items.extend(
+                (replace(part, id='content', provider_name=self.system) if isinstance(part, ThinkingPart) else part)
+                for part in split_content_into_text_and_thinking(choice.message.content, self.profile.thinking_tags)
+            )
+        if choice.message.tool_calls is not None:
+            for c in choice.message.tool_calls:
+                items.append(ToolCallPart(c.get("function").get("name"), c.get("function").get("arguments"), tool_call_id=c.get("id")))
+
+        raw_finish_reason = choice.finish_reason
+        finish_reason = _CHAT_FINISH_REASON_MAP.get(raw_finish_reason)
+
+        return ModelResponse(
+            parts=items,
+            usage=_map_usage(response, self._provider, "", self._model_name),
+            model_name=response.model,
+            timestamp=timestamp,
+            provider_details=None,
+            provider_response_id=response.id,
+            provider_name=self._provider,
+            finish_reason=finish_reason,
+        )
+
 def new_infer_model(model: Model | KnownModelName, api_key: str = None) -> Model:
     if isinstance(model, Model):
         return model
@@ -69,4 +126,4 @@ def patch_infer_model(api_key: str | None = None):
     import pydantic_ai.models
 
     pydantic_ai.models.infer_model = partial(new_infer_model, api_key=api_key)
-    pydantic_ai.models.AIModel = AIModel
+    pydantic_ai.models.OpenAIChatModel = AIModel