File size: 6,295 Bytes
1397957
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
from typing import Dict, Any, List, Optional, AsyncGenerator
import os
import json

from .provider import BaseProvider, ModelInfo, Message, StreamChunk, ToolCall


class OpenAIProvider(BaseProvider):
    
    def __init__(self, api_key: Optional[str] = None):
        self._api_key = api_key or os.environ.get("OPENAI_API_KEY")
        self._client = None
    
    @property
    def id(self) -> str:
        return "openai"
    
    @property
    def name(self) -> str:
        return "OpenAI"
    
    @property
    def models(self) -> Dict[str, ModelInfo]:
        return {
            "gpt-4o": ModelInfo(
                id="gpt-4o",
                name="GPT-4o",
                provider_id="openai",
                context_limit=128000,
                output_limit=16384,
                supports_tools=True,
                supports_streaming=True,
                cost_input=2.5,
                cost_output=10.0,
            ),
            "gpt-4o-mini": ModelInfo(
                id="gpt-4o-mini",
                name="GPT-4o Mini",
                provider_id="openai",
                context_limit=128000,
                output_limit=16384,
                supports_tools=True,
                supports_streaming=True,
                cost_input=0.15,
                cost_output=0.6,
            ),
            "o1": ModelInfo(
                id="o1",
                name="o1",
                provider_id="openai",
                context_limit=200000,
                output_limit=100000,
                supports_tools=True,
                supports_streaming=True,
                cost_input=15.0,
                cost_output=60.0,
            ),
        }
    
    def _get_client(self):
        if self._client is None:
            try:
                from openai import AsyncOpenAI
                self._client = AsyncOpenAI(api_key=self._api_key)
            except ImportError:
                raise ImportError("openai package is required. Install with: pip install openai")
        return self._client
    
    async def stream(
        self,
        model_id: str,
        messages: List[Message],
        tools: Optional[List[Dict[str, Any]]] = None,
        system: Optional[str] = None,
        temperature: Optional[float] = None,
        max_tokens: Optional[int] = None,
    ) -> AsyncGenerator[StreamChunk, None]:
        client = self._get_client()
        
        openai_messages = []
        
        if system:
            openai_messages.append({"role": "system", "content": system})
        
        for msg in messages:
            content = msg.content
            if isinstance(content, str):
                openai_messages.append({"role": msg.role, "content": content})
            else:
                openai_messages.append({
                    "role": msg.role,
                    "content": [{"type": c.type, "text": c.text} for c in content if c.text]
                })
        
        kwargs: Dict[str, Any] = {
            "model": model_id,
            "messages": openai_messages,
            "stream": True,
        }
        
        if max_tokens:
            kwargs["max_tokens"] = max_tokens
        
        if temperature is not None:
            kwargs["temperature"] = temperature
        
        if tools:
            kwargs["tools"] = [
                {
                    "type": "function",
                    "function": {
                        "name": t["name"],
                        "description": t.get("description", ""),
                        "parameters": t.get("parameters", t.get("input_schema", {}))
                    }
                }
                for t in tools
            ]
        
        tool_calls: Dict[int, Dict[str, Any]] = {}
        usage_data = None
        finish_reason = None
        
        async for chunk in await client.chat.completions.create(**kwargs):
            if chunk.choices and chunk.choices[0].delta:
                delta = chunk.choices[0].delta
                
                if delta.content:
                    yield StreamChunk(type="text", text=delta.content)
                
                if delta.tool_calls:
                    for tc in delta.tool_calls:
                        idx = tc.index
                        if idx not in tool_calls:
                            tool_calls[idx] = {
                                "id": tc.id or "",
                                "name": tc.function.name if tc.function else "",
                                "arguments": ""
                            }
                        
                        if tc.id:
                            tool_calls[idx]["id"] = tc.id
                        if tc.function:
                            if tc.function.name:
                                tool_calls[idx]["name"] = tc.function.name
                            if tc.function.arguments:
                                tool_calls[idx]["arguments"] += tc.function.arguments
            
            if chunk.choices and chunk.choices[0].finish_reason:
                finish_reason = chunk.choices[0].finish_reason
            
            if chunk.usage:
                usage_data = {
                    "input_tokens": chunk.usage.prompt_tokens,
                    "output_tokens": chunk.usage.completion_tokens,
                }
        
        for tc_data in tool_calls.values():
            try:
                args = json.loads(tc_data["arguments"]) if tc_data["arguments"] else {}
            except json.JSONDecodeError:
                args = {}
            yield StreamChunk(
                type="tool_call",
                tool_call=ToolCall(
                    id=tc_data["id"],
                    name=tc_data["name"],
                    arguments=args
                )
            )
        
        stop_reason = self._map_stop_reason(finish_reason)
        yield StreamChunk(type="done", usage=usage_data, stop_reason=stop_reason)
    
    def _map_stop_reason(self, openai_finish_reason: Optional[str]) -> str:
        mapping = {
            "stop": "end_turn",
            "tool_calls": "tool_calls",
            "length": "max_tokens",
            "content_filter": "end_turn",
        }
        return mapping.get(openai_finish_reason or "", "end_turn")