gpt4free

from __future__ import annotations

import json
import requests
from aiohttp import ClientSession

from ...typing import AsyncResult, Messages, MediaListType
from ..base_provider import AsyncGeneratorProvider, ProviderModelMixin
from ...providers.response import FinishReason, Usage, Reasoning, ToolCalls
from ...tools.media import render_messages
from ...requests import sse_stream, raise_for_status
from ...errors import ResponseError, ModelNotFoundError, MissingAuthError
from ..helper import format_media_prompt
from ... import debug


class Puter(AsyncGeneratorProvider, ProviderModelMixin):
    label = "Puter.js"
    url = "https://docs.puter.com/playground"
    login_url = "https://github.com/HeyPuter/puter-cli"
    models_endpoint = "https://api.puter.com/puterai/chat/models/"
    api_endpoint = "https://api.puter.com/drivers/call"
    working = True
    active_by_default = True
    needs_auth = True
    quota_url = "https://api.puter.com/metering/usage"
    supports_native_tools = True

    default_model = "gpt-5.1"
    default_vision_model = default_model
    model_aliases = {
        ### mistral_models ###
        "mixtral-8x22b": "open-mixtral-8x22b",
        "pixtral-large": "pixtral-large-latest",
        ### openrouter_models ###
        # llama
        "llama-2-70b": "openrouter:meta-llama/llama-2-70b-chat",
        "llama-3-8b": "openrouter:meta-llama/llama-3-8b-instruct",
        "llama-3-70b": "openrouter:meta-llama/llama-3-70b-instruct",
        "llama-3.1-8b": [
            "openrouter:meta-llama/llama-3.1-8b-instruct:free",
            "openrouter:meta-llama/llama-3.1-8b-instruct",
        ],
        "llama-3.1-70b": "openrouter:meta-llama/llama-3.1-70b-instruct",
        "llama-3.1-405b": [
            "openrouter:meta-llama/llama-3.1-405b:free",
            "openrouter:meta-llama/llama-3.1-405b",
            "openrouter:meta-llama/llama-3.1-405b-instruct",
        ],
        "llama-3.2-1b": [
            "openrouter:meta-llama/llama-3.2-1b-instruct:free",
            "openrouter:meta-llama/llama-3.2-1b-instruct",
        ],
        "llama-3.2-3b": [
            "openrouter:meta-llama/llama-3.2-3b-instruct:free",
            "openrouter:meta-llama/llama-3.2-3b-instruct",
        ],
        "llama-3.2-11b": [
            "openrouter:meta-llama/llama-3.2-11b-vision-instruct:free",
            "openrouter:meta-llama/llama-3.2-11b-vision-instruct",
        ],
        "llama-3.2-90b": "openrouter:meta-llama/llama-3.2-90b-vision-instruct",
        "llama-3.3-8b": "openrouter:meta-llama/llama-3.3-8b-instruct:free",
        "llama-3.3-70b": [
            "openrouter:meta-llama/llama-3.3-70b-instruct:free",
            "openrouter:meta-llama/llama-3.3-70b-instruct",
        ],
        "llama-4-maverick": [
            "openrouter:meta-llama/llama-4-maverick:free",
            "openrouter:meta-llama/llama-4-maverick",
        ],
        "llama-4-scout": [
            "openrouter:meta-llama/llama-4-scout:free",
            "openrouter:meta-llama/llama-4-scout",
        ],
        # google (gemini)
        "gemini-1.5-flash": [
            "gemini-1.5-flash",
            "openrouter:google/gemini-flash-1.5",
            "gemini-flash-1.5-8b",
        ],
        "gemini-1.5-8b-flash": "openrouter:google/gemini-flash-1.5-8b",
        "gemini-1.5-pro": "openrouter:google/gemini-pro-1.5",
        "gemini-2.0-flash": [
            "gemini-2.0-flash",
            "openrouter:google/gemini-2.0-flash-lite-001",
            "openrouter:google/gemini-2.0-flash-001",
            "openrouter:google/gemini-2.0-flash-exp:free",
        ],
        "gemini-2.5-pro": [
            "openrouter:google/gemini-2.5-pro-preview",
            "openrouter:google/gemini-2.5-pro-exp-03-25",
        ],
        "gemini-2.5-flash": "openrouter:google/gemini-2.5-flash-preview",
        "gemini-2.5-flash-thinking": "openrouter:google/gemini-2.5-flash-preview:thinking",
        # google (gemma)
        "gemma-2-9b": [
            "openrouter:google/gemma-2-9b-it:free",
            "openrouter:google/gemma-2-9b-it",
        ],
        "gemma-2-27b": "openrouter:google/gemma-2-27b-it",
        "gemma-3-1b": "openrouter:google/gemma-3-1b-it:free",
        "gemma-3-4b": [
            "openrouter:google/gemma-3-4b-it:free",
            "openrouter:google/gemma-3-4b-it",
        ],
        "gemma-3-12b": [
            "openrouter:google/gemma-3-12b-it:free",
            "openrouter:google/gemma-3-12b-it",
        ],
        "gemma-3-27b": [
            "openrouter:google/gemma-3-27b-it:free",
            "openrouter:google/gemma-3-27b-it",
        ],
        # nousresearch
        "hermes-2-dpo": "openrouter:nousresearch/nous-hermes-2-mixtral-8x7b-dpo",
        "hermes-2-pro": "openrouter:nousresearch/hermes-2-pro-llama-3-8b",
        "hermes-3-70b": "openrouter:nousresearch/hermes-3-llama-3.1-70b",
        "hermes-3-405b": "openrouter:nousresearch/hermes-3-llama-3.1-405b",
        "deephermes-3-8b": "openrouter:nousresearch/deephermes-3-llama-3-8b-preview:free",
        "deephermes-3-24b": "openrouter:nousresearch/deephermes-3-mistral-24b-preview:free",
        # microsoft
        "phi-3-mini": "openrouter:microsoft/phi-3-mini-128k-instruct",
        "phi-3-medium": "openrouter:microsoft/phi-3-medium-128k-instruct",
        "phi-3.5-mini": "openrouter:microsoft/phi-3.5-mini-128k-instruct",
        "phi-4": "openrouter:microsoft/phi-4",
        "phi-4-multimodal": "openrouter:microsoft/phi-4-multimodal-instruct",
        "phi-4-reasoning": "openrouter:microsoft/phi-4-reasoning:free",
        "phi-4-reasoning-plus": [
            "openrouter:microsoft/phi-4-reasoning-plus:free",
            "openrouter:microsoft/phi-4-reasoning-plus",
        ],
        "wizardlm-2-8x22b": "openrouter:microsoft/wizardlm-2-8x22b",
        "mai-ds-r1": "openrouter:microsoft/mai-ds-r1:free",
        # anthropic
        "claude-3.7-sonnet": [
            "claude-3-7-sonnet-20250219",
            "claude-3-7-sonnet-latest",
            "openrouter:anthropic/claude-3.7-sonnet",
            "openrouter:anthropic/claude-3.7-sonnet:beta",
        ],
        "claude-3.7-sonnet-thinking": "openrouter:anthropic/claude-3.7-sonnet:thinking",
        "claude-3.5-haiku": [
            "openrouter:anthropic/claude-3.5-haiku:beta",
            "openrouter:anthropic/claude-3.5-haiku",
            "openrouter:anthropic/claude-3.5-haiku-20241022:beta",
            "openrouter:anthropic/claude-3.5-haiku-20241022",
        ],
        "claude-3.5-sonnet": [
            "claude-3-5-sonnet-20241022",
            "claude-3-5-sonnet-latest",
            "claude-3-5-sonnet-20240620",
            "openrouter:anthropic/claude-3.5-sonnet-20240620:beta",
            "openrouter:anthropic/claude-3.5-sonnet-20240620",
            "openrouter:anthropic/claude-3.5-sonnet:beta",
            "openrouter:anthropic/claude-3.5-sonnet",
        ],
        "claude-3-haiku": [
            "claude-3-haiku-20240307",
            "openrouter:anthropic/claude-3-haiku:beta",
            "openrouter:anthropic/claude-3-haiku",
        ],
        "claude-3-opus": [
            "openrouter:anthropic/claude-3-opus:beta",
            "openrouter:anthropic/claude-3-opus",
        ],
        "claude-3-sonnet": [
            "openrouter:anthropic/claude-3-sonnet:beta",
            "openrouter:anthropic/claude-3-sonnet",
        ],
        "claude-2.1": [
            "openrouter:anthropic/claude-2.1:beta",
            "openrouter:anthropic/claude-2.1",
        ],
        "claude-2": [
            "openrouter:anthropic/claude-2:beta",
            "openrouter:anthropic/claude-2",
        ],
        "claude-2.0": [
            "openrouter:anthropic/claude-2.0:beta",
            "openrouter:anthropic/claude-2.0",
        ],
        # rekaai
        "reka-flash": "openrouter:rekaai/reka-flash-3:free",
        # cohere
        "command-r7b": "openrouter:cohere/command-r7b-12-2024",
        "command-r-plus": [
            "openrouter:cohere/command-r-plus-08-2024",
            "openrouter:cohere/command-r-plus",
            "openrouter:cohere/command-r-plus-04-2024",
        ],
        "command": "openrouter:cohere/command",
        "command-r": [
            "openrouter:cohere/command-r-08-2024",
            "openrouter:cohere/command-r",
            "openrouter:cohere/command-r-03-2024",
        ],
        "command-a": "openrouter:cohere/command-a",
        # qwen
        "qwq-32b": [
            "openrouter:qwen/qwq-32b-preview",
            "openrouter:qwen/qwq-32b:free",
            "openrouter:qwen/qwq-32b",
        ],
        "qwen-vl-plus": "openrouter:qwen/qwen-vl-plus",
        "qwen-vl-max": "openrouter:qwen/qwen-vl-max",
        "qwen-turbo": "openrouter:qwen/qwen-turbo",
        "qwen-2.5-vl-72b": [
            "openrouter:qwen/qwen2.5-vl-72b-instruct:free",
            "openrouter:qwen/qwen2.5-vl-72b-instruct",
        ],
        "qwen-plus": "openrouter:qwen/qwen-plus",
        "qwen-max": "openrouter:qwen/qwen-max",
        "qwen-2.5-coder-32b": [
            "openrouter:qwen/qwen-2.5-coder-32b-instruct:free",
            "openrouter:qwen/qwen-2.5-coder-32b-instruct",
        ],
        "qwen-2.5-7b": [
            "openrouter:qwen/qwen-2.5-7b-instruct:free",
            "openrouter:qwen/qwen-2.5-7b-instruct",
        ],
        "qwen-2.5-72b": [
            "openrouter:qwen/qwen-2.5-72b-instruct:free",
            "openrouter:qwen/qwen-2.5-72b-instruct",
        ],
        "qwen-2.5-vl-7b": [
            "openrouter:qwen/qwen-2.5-vl-7b-instruct:free",
            "openrouter:qwen/qwen-2.5-vl-7b-instruct",
        ],
        "qwen-2-72b": "openrouter:qwen/qwen-2-72b-instruct",
        "qwen-3-0.6b": "openrouter:qwen/qwen3-0.6b-04-28:free",
        "qwen-3-1.7b": "openrouter:qwen/qwen3-1.7b:free",
        "qwen-3-4b": "openrouter:qwen/qwen3-4b:free",
        "qwen-3-30b": [
            "openrouter:qwen/qwen3-30b-a3b:free",
            "openrouter:qwen/qwen3-30b-a3b",
        ],
        "qwen-3-8b": ["openrouter:qwen/qwen3-8b:free", "openrouter:qwen/qwen3-8b"],
        "qwen-3-14b": ["openrouter:qwen/qwen3-14b:free", "openrouter:qwen/qwen3-14b"],
        "qwen-3-32b": ["openrouter:qwen/qwen3-32b:free", "openrouter:qwen/qwen3-32b"],
        "qwen-3-235b": [
            "openrouter:qwen/qwen3-235b-a22b:free",
            "openrouter:qwen/qwen3-235b-a22b",
        ],
        "qwen-2.5-coder-7b": "openrouter:qwen/qwen2.5-coder-7b-instruct",
        "qwen-2.5-vl-3b": "openrouter:qwen/qwen2.5-vl-3b-instruct:free",
        "qwen-2.5-vl-32b": [
            "openrouter:qwen/qwen2.5-vl-32b-instruct:free",
            "openrouter:qwen/qwen2.5-vl-32b-instruct",
        ],
        # deepseek
        "deepseek-prover-v2": [
            "openrouter:deepseek/deepseek-prover-v2:free",
            "openrouter:deepseek/deepseek-prover-v2",
        ],
        "deepseek-v3": "openrouter:deepseek/deepseek-v3-base:free",
        "deepseek-v3-0324": [
            "deepseek-chat",
            "openrouter:deepseek/deepseek-chat-v3-0324:free",
            "openrouter:deepseek/deepseek-chat-v3-0324",
        ],
        "deepseek-r1-zero": "openrouter:deepseek/deepseek-r1-zero:free",
        "deepseek-r1-distill-llama-8b": "openrouter:deepseek/deepseek-r1-distill-llama-8b",
        "deepseek-r1-distill-qwen-1.5b": "openrouter:deepseek/deepseek-r1-distill-qwen-1.5b",
        "deepseek-r1-distill-qwen-32b": [
            "openrouter:deepseek/deepseek-r1-distill-qwen-32b:free",
            "openrouter:deepseek/deepseek-r1-distill-qwen-32b",
        ],
        "deepseek-r1-distill-qwen-14b": [
            "openrouter:deepseek/deepseek-r1-distill-qwen-14b:free",
            "openrouter:deepseek/deepseek-r1-distill-qwen-14b",
        ],
        "deepseek-r1-distill-llama-70b": [
            "openrouter:deepseek/deepseek-r1-distill-llama-70b:free",
            "openrouter:deepseek/deepseek-r1-distill-llama-70b",
        ],
        "deepseek-r1": [
            "deepseek-reasoner",
            "openrouter:deepseek/deepseek-r1:free",
            "openrouter:deepseek/deepseek-r1",
        ],
        "deepseek-chat": [
            "deepseek-chat",
            "openrouter:deepseek/deepseek-chat:free",
            "openrouter:deepseek/deepseek-chat",
        ],
        "deepseek-coder": ["openrouter:deepseek/deepseek-coder"],
        # inflection
        "inflection-3-productivity": "openrouter:inflection/inflection-3-productivity",
        "inflection-3-pi": "openrouter:inflection/inflection-3-pi",
        # x-ai
        "grok-3-mini": "openrouter:x-ai/grok-3-mini-beta",
        "grok-3-beta": "openrouter:x-ai/grok-3-beta",
        "grok-2": ["openrouter:x-ai/grok-2-vision-1212", "openrouter:x-ai/grok-2-1212"],
        "grok": [
            "openrouter:x-ai/grok-vision-beta",
            "openrouter:x-ai/grok-2-vision-1212",
            "openrouter:x-ai/grok-2-1212",
            "grok-beta",
            "grok-vision-beta",
            "openrouter:x-ai/grok-beta",
            "openrouter:x-ai/grok-3-beta",
            "openrouter:x-ai/grok-3-mini-beta",
        ],
        "grok-beta": [
            "grok-beta",
            "grok-vision-beta",
            "openrouter:x-ai/grok-beta",
            "openrouter:x-ai/grok-3-beta",
        ],
        # perplexity
        "sonar-reasoning-pro": "openrouter:perplexity/sonar-reasoning-pro",
        "sonar-pro": "openrouter:perplexity/sonar-pro",
        "sonar-deep-research": "openrouter:perplexity/sonar-deep-research",
        "r1-1776": "openrouter:perplexity/r1-1776",
        "sonar-reasoning": "openrouter:perplexity/sonar-reasoning",
        "sonar": "openrouter:perplexity/sonar",
        "llama-3.1-sonar-small-online": "openrouter:perplexity/llama-3.1-sonar-small-128k-online",
        "llama-3.1-sonar-large-online": "openrouter:perplexity/llama-3.1-sonar-large-128k-online",
        # nvidia
        "nemotron-49b": [
            "openrouter:nvidia/llama-3.3-nemotron-super-49b-v1:free",
            "openrouter:nvidia/llama-3.3-nemotron-super-49b-v1",
        ],
        "nemotron-70b": "openrouter:nvidia/llama-3.1-nemotron-70b-instruct",
        "nemotron-253b": "openrouter:nvidia/llama-3.1-nemotron-ultra-253b-v1:free",
        # thudm
        "glm-4": [
            "openrouter:thudm/glm-4-32b:free",
            "openrouter:thudm/glm-4-32b",
            "openrouter:thudm/glm-4-9b:free",
        ],
        "glm-4-32b": ["openrouter:thudm/glm-4-32b:free", "openrouter:thudm/glm-4-32b"],
        "glm-z1-32b": [
            "openrouter:thudm/glm-z1-32b:free",
            "openrouter:thudm/glm-z1-32b",
        ],
        "glm-4-9b": "openrouter:thudm/glm-4-9b:free",
        "glm-z1-9b": "openrouter:thudm/glm-z1-9b:free",
        "glm-z1-rumination-32b": "openrouter:thudm/glm-z1-rumination-32b",
        # minimax
        "minimax": "openrouter:minimax/minimax-01",
        # cognitivecomputations
        "dolphin-3.0-r1-24b": "openrouter:cognitivecomputations/dolphin3.0-r1-mistral-24b:free",
        "dolphin-3.0-24b": "openrouter:cognitivecomputations/dolphin3.0-mistral-24b:free",
        "dolphin-8x22b": "openrouter:cognitivecomputations/dolphin-mixtral-8x22b",
        # agentica-org
        "deepcoder-14b": "openrouter:agentica-org/deepcoder-14b-preview:free",
        # moonshotai
        "kimi-vl-thinking": "openrouter:moonshotai/kimi-vl-a3b-thinking:free",
        "moonlight-16b": "openrouter:moonshotai/moonlight-16b-a3b-instruct:free",
        # featherless
        "qwerky-72b": "openrouter:featherless/qwerky-72b:free",
        # liquid
        "lfm-7b": "openrouter:liquid/lfm-7b",
        "lfm-3b": "openrouter:liquid/lfm-3b",
        "lfm-40b": "openrouter:liquid/lfm-40b",
    }

    @classmethod
    def get_models(cls, **kwargs) -> list[str]:
        if not cls.models:
            try:
                url = cls.models_endpoint
                cls.models = (
                    requests.get(url, timeout=kwargs.get("timeout", 15))
                    .json()
                    .get("models", [])
                )
                cls.models = [
                    model
                    for model in cls.models
                    if model not in ["abuse", "costly", "fake", "model-fallback-test-1"]
                ]
                cls.live += 1
            except Exception as e:
                debug.log(f"PuterJS: Failed to fetch models from API: {e}")
                cls.models = []
            cls.vision_models = []
            for model in cls.models:
                for tag in ["vision", "multimodal", "gpt", "o1", "o3", "o4"]:
                    if tag in model:
                        cls.vision_models.append(model)
        return cls.models

    @classmethod
    async def create_async_generator(
        cls,
        model: str,
        messages: Messages,
        prompt: str = None,
        proxy: str = None,
        stream: bool = True,
        api_key: str = None,
        media: MediaListType = None,
        extra_parameters: list[str] = [
            "temperature",
            "presence_penalty",
            "top_p",
            "frequency_penalty",
            "response_format",
            "tools",
            "parallel_tool_calls",
            "tool_choice",
            "reasoning_effort",
            "logit_bias",
            "voice",
            "modalities",
            "audio",
            "prompt_cache_key",
        ],
        **kwargs,
    ) -> AsyncResult:
        if not api_key:
            raise MissingAuthError("API key is required for Puter.js API")

        # Check if we need to use a vision model
        if not model and media is not None and len(media) > 0:
            model = cls.default_vision_model

        # Check for image URLs in messages
        if not model:
            for msg in messages:
                if msg["role"] == "user":
                    content = msg.get("content", "")
                    if isinstance(content, list):
                        for item in content:
                            if item.get("type") == "image_url":
                                model = cls.default_vision_model
                                break

        # Get the model name from the user-provided model
        try:
            model = cls.get_model(model)
        except ModelNotFoundError:
            pass

        async with ClientSession() as session:
            headers = {
                "authorization": f"Bearer {api_key}",
                "user-agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/135.0.0.0 Safari/537.36",
                "content-type": "application/json;charset=UTF-8",
                # Set appropriate accept header based on stream mode
                "accept": "text/event-stream" if stream else "*/*",
                "origin": "http://docs.puter.com",
                "sec-fetch-site": "cross-site",
                "sec-fetch-mode": "cors",
                "sec-fetch-dest": "empty",
                "referer": "http://docs.puter.com/",
                "accept-encoding": "gzip",
                "accept-language": "en-US,en;q=0.9",
            }

            json_data = {
                "interface": "puter-image-generation"
                if model in cls.image_models
                else "puter-chat-completion",
                "driver": "ai-chat",
                "test_mode": messages[0]["content"] == "test",
                "method": "generate" if model in cls.image_models else "complete",
                "args": {"prompt": format_media_prompt(messages, prompt)}
                if model in cls.image_models
                else {
                    "messages": list(render_messages(messages, media)),
                    "model": model,
                    "stream": stream,
                    **{
                        param: kwargs.get(param)
                        for param in extra_parameters
                        if param in kwargs
                    },
                },
            }
            async with session.post(
                cls.api_endpoint, headers=headers, json=json_data, proxy=proxy
            ) as response:
                await raise_for_status(response)
                mime_type = response.headers.get("content-type", "")
                if mime_type.startswith("text/plain"):
                    yield await response.text()
                    return
                elif mime_type.startswith("text/event-stream"):
                    reasoning = False
                    async for result in sse_stream(response.content):
                        if "error" in result:
                            raise ResponseError(
                                result["error"].get("message", result["error"])
                            )
                        choices = result.get("choices", [{}])
                        choice = choices.pop() if choices else {}
                        content = choice.get("delta", {}).get("content")
                        reasoning_content = choice.get("delta", {}).get(
                            "reasoning_content"
                        )
                        if reasoning_content:
                            reasoning = True
                            yield Reasoning(reasoning_content)
                        elif content:
                            if reasoning:
                                yield Reasoning(status="")
                                reasoning = False
                            yield content
                        if result.get("usage") is not None:
                            yield Usage(**result["usage"])
                        tool_calls = choice.get("delta", {}).get("tool_calls")
                        if tool_calls:
                            yield ToolCalls(choice["delta"]["tool_calls"])
                        finish_reason = choice.get("finish_reason")
                        if finish_reason:
                            yield FinishReason(finish_reason)
                elif mime_type.startswith("application/json"):
                    result = await response.json()
                    if "choices" in result:
                        choice = result["choices"][0]
                    elif "result" in result:
                        choice = result.get("result", {})
                    else:
                        raise ResponseError(result)
                    message = choice.get("message", {})
                    reasoning = message.get("reasoning")
                    if reasoning:
                        yield Reasoning(reasoning)
                    content = message.get("content", "")
                    if isinstance(content, list):
                        for item in content:
                            if item.get("type") == "text":
                                yield item.get("text", "")
                    elif content:
                        yield content
                    if "tool_calls" in message:
                        yield ToolCalls(message["tool_calls"])
                    if result.get("usage") is not None:
                        yield Usage(**result["usage"])
                    finish_reason = choice.get("finish_reason")
                    if finish_reason:
                        yield FinishReason(finish_reason)
                elif mime_type.startswith("application/x-ndjson"):
                    tools_idx = 0
                    async for line in response.content:
                        data = json.loads(line)
                        if data.get("type") == "text":
                            yield data.get("text", "")
                        elif data.get("type") == "reasoning":
                            yield Reasoning(data.get("reasoning", ""))
                        elif data.get("type") == "tool_use":
                            yield ToolCalls(
                                [
                                    {
                                        "id": tools_idx,
                                        "type": "function",
                                        "id": data.get("id"),
                                        "function": {
                                            "name": data.get("name"),
                                            "arguments": data.get("input"),
                                        },
                                    }
                                ]
                            )
                            tools_idx += 1
                        elif data.get("type") == "tool_calls":
                            yield ToolCalls(data.get("tool_calls", []))
                        elif data.get("type") == "usage":
                            yield Usage.from_dict(data.get("usage", {}))
                else:
                    raise ResponseError(f"Unexpected content type: {mime_type}")