Skip to content

Async Providers

llm_client.providers.async_providers

Async support for LLM providers.

Classes

AsyncGeminiProvider

Bases: BaseProvider, AsyncProviderMixin

Async Gemini provider via OpenAI compatibility.

Source code in llm_client/providers/async_providers.py 
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
class AsyncGeminiProvider(BaseProvider, AsyncProviderMixin):
    """Async Gemini provider via OpenAI compatibility."""

    def _initialize_client(self, **kwargs: Any) -> None:
        """Initialize async Gemini client."""
        if not self.is_available():
            raise ProviderNotAvailableError("gemini", "openai")

        api_key = kwargs.get("api_key")
        if not api_key:
            raise APIKeyNotFoundError("gemini", "GEMINI_API_KEY")

        self.client = AsyncOpenAI(
            api_key=api_key,
            base_url="https://generativelanguage.googleapis.com/v1beta/openai/",
        )

    def _chat_completion_impl(self, messages: list[dict[str, str]]) -> str:
        """Sync method raises error - use async version."""
        raise RuntimeError(
            "AsyncGeminiProvider only supports async methods. " "Use achat_completion() instead."
        )

    async def _achat_completion_impl(self, messages: list[dict[str, str]]) -> str:
        """Execute async chat completion with Gemini."""
        if not self.client:
            raise RuntimeError("Gemini client not initialized")

        response = await self.client.chat.completions.create(
            model=self.llm,
            messages=messages,
            temperature=self.temperature,
            max_tokens=self.max_tokens,
        )
        return response.choices[0].message.content

    async def _achat_completion_with_tools_impl(
        self,
        messages: list[dict[str, str]],
        tools: list[dict],
        tool_choice: str | dict | None = None,
    ) -> dict:
        """Execute async chat completion with tools using Gemini."""
        if not self.client:
            raise RuntimeError("Gemini client not initialized")

        kwargs = {
            "model": self.llm,
            "messages": messages,
            "temperature": self.temperature,
            "max_tokens": self.max_tokens,
            "tools": tools,
        }

        if tool_choice is not None:
            kwargs["tool_choice"] = tool_choice

        response = await self.client.chat.completions.create(**kwargs)

        choice = response.choices[0]
        return {
            "content": choice.message.content,
            "tool_calls": (
                [
                    {
                        "id": tc.id,
                        "type": tc.type,
                        "function": {"name": tc.function.name, "arguments": tc.function.arguments},
                    }
                    for tc in (choice.message.tool_calls or [])
                ]
                if choice.message.tool_calls
                else None
            ),
        }

    async def _achat_completion_with_files_impl(
        self,
        messages: list[dict[str, str]],
        files: list[str] | None = None,
    ) -> str:
        """Execute async chat completion with files using Gemini."""
        if not self.client:
            raise RuntimeError("Gemini client not initialized")

        enhanced_messages = messages.copy()

        if files:
            file_data = prepare_files_for_provider(files, "gemini")

            if enhanced_messages and enhanced_messages[-1]["role"] == "user":
                last_msg = enhanced_messages[-1]
                if isinstance(last_msg["content"], str):
                    text_content = last_msg["content"]
                    last_msg["content"] = [{"type": "text", "text": text_content}]

                last_msg["content"].extend(file_data)
            else:
                enhanced_messages.append({"role": "user", "content": file_data})

        response = await self.client.chat.completions.create(
            model=self.llm,
            messages=enhanced_messages,
            temperature=self.temperature,
            max_tokens=self.max_tokens,
        )

        return response.choices[0].message.content

    async def _achat_completion_stream_impl(
        self, messages: list[dict[str, str]]
    ) -> AsyncIterator[str]:
        """Stream async chat completion with Gemini."""
        if not self.client:
            raise RuntimeError("Gemini client not initialized")

        stream = await self.client.chat.completions.create(
            model=self.llm,
            messages=messages,
            temperature=self.temperature,
            max_tokens=self.max_tokens,
            stream=True,
        )

        async for chunk in stream:
            if chunk.choices[0].delta.content is not None:
                yield chunk.choices[0].delta.content

    @staticmethod
    def get_default_model() -> str:
        """Get default Gemini model."""
        return "gemini-2.0-flash-exp"

    @staticmethod
    def is_available() -> bool:
        """Check if AsyncOpenAI (for Gemini) is available."""
        return AsyncOpenAI is not None
Functions
get_default_model() staticmethod

Get default Gemini model.

Source code in llm_client/providers/async_providers.py 
566
567
568
569
@staticmethod
def get_default_model() -> str:
    """Get default Gemini model."""
    return "gemini-2.0-flash-exp"
is_available() staticmethod

Check if AsyncOpenAI (for Gemini) is available.

Source code in llm_client/providers/async_providers.py 
571
572
573
574
@staticmethod
def is_available() -> bool:
    """Check if AsyncOpenAI (for Gemini) is available."""
    return AsyncOpenAI is not None

AsyncGroqProvider

Bases: BaseProvider, AsyncProviderMixin

Async Groq provider.

Source code in llm_client/providers/async_providers.py 
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
class AsyncGroqProvider(BaseProvider, AsyncProviderMixin):
    """Async Groq provider."""

    def _initialize_client(self, **kwargs: Any) -> None:
        """Initialize async Groq client."""
        if not self.is_available():
            raise ProviderNotAvailableError("groq", "groq")

        api_key = kwargs.get("api_key")
        if not api_key:
            raise APIKeyNotFoundError("groq", "GROQ_API_KEY")

        self.client = AsyncGroq(api_key=api_key)

    def _chat_completion_impl(self, messages: list[dict[str, str]]) -> str:
        """Sync method raises error - use async version."""
        raise RuntimeError(
            "AsyncGroqProvider only supports async methods. " "Use achat_completion() instead."
        )

    async def _achat_completion_impl(self, messages: list[dict[str, str]]) -> str:
        """Execute async chat completion with Groq."""
        if not self.client:
            raise RuntimeError("Groq client not initialized")

        response = await self.client.chat.completions.create(
            model=self.llm,
            messages=messages,
            temperature=self.temperature,
            max_tokens=self.max_tokens,
        )
        return response.choices[0].message.content

    async def _achat_completion_with_tools_impl(
        self,
        messages: list[dict[str, str]],
        tools: list[dict],
        tool_choice: str | dict | None = None,
    ) -> dict:
        """Execute async chat completion with tools using Groq."""
        if not self.client:
            raise RuntimeError("Groq client not initialized")

        kwargs = {
            "model": self.llm,
            "messages": messages,
            "temperature": self.temperature,
            "max_tokens": self.max_tokens,
            "tools": tools,
        }

        if tool_choice is not None:
            kwargs["tool_choice"] = tool_choice

        response = await self.client.chat.completions.create(**kwargs)

        choice = response.choices[0]
        return {
            "content": choice.message.content,
            "tool_calls": (
                [
                    {
                        "id": tc.id,
                        "type": tc.type,
                        "function": {"name": tc.function.name, "arguments": tc.function.arguments},
                    }
                    for tc in (choice.message.tool_calls or [])
                ]
                if choice.message.tool_calls
                else None
            ),
        }

    async def _achat_completion_with_files_impl(
        self,
        messages: list[dict[str, str]],
        files: list[str] | None = None,
    ) -> str:
        """Execute async chat completion with files using Groq."""
        if not self.client:
            raise RuntimeError("Groq client not initialized")

        if files:
            for file_path in files:
                if detect_file_type(file_path) != "image":
                    raise ValueError("Groq only supports image files for vision models")

        enhanced_messages = messages.copy()

        if files:
            file_data = prepare_files_for_provider(files, "groq")

            if enhanced_messages and enhanced_messages[-1]["role"] == "user":
                last_msg = enhanced_messages[-1]
                if isinstance(last_msg["content"], str):
                    text_content = last_msg["content"]
                    last_msg["content"] = [{"type": "text", "text": text_content}]

                last_msg["content"].extend(file_data)
            else:
                enhanced_messages.append({"role": "user", "content": file_data})

        response = await self.client.chat.completions.create(
            model=self.llm,
            messages=enhanced_messages,
            temperature=self.temperature,
            max_tokens=self.max_tokens,
        )

        return response.choices[0].message.content

    async def _achat_completion_stream_impl(
        self, messages: list[dict[str, str]]
    ) -> AsyncIterator[str]:
        """Stream async chat completion with Groq."""
        if not self.client:
            raise RuntimeError("Groq client not initialized")

        stream = await self.client.chat.completions.create(
            model=self.llm,
            messages=messages,
            temperature=self.temperature,
            max_tokens=self.max_tokens,
            stream=True,
        )

        async for chunk in stream:
            if chunk.choices[0].delta.content is not None:
                yield chunk.choices[0].delta.content

    @staticmethod
    def get_default_model() -> str:
        """Get default Groq model."""
        return "moonshotai/kimi-k2-instruct-0905"

    @staticmethod
    def is_available() -> bool:
        """Check if AsyncGroq is available."""
        return AsyncGroq is not None
Functions
get_default_model() staticmethod

Get default Groq model.

Source code in llm_client/providers/async_providers.py 
427
428
429
430
@staticmethod
def get_default_model() -> str:
    """Get default Groq model."""
    return "moonshotai/kimi-k2-instruct-0905"
is_available() staticmethod

Check if AsyncGroq is available.

Source code in llm_client/providers/async_providers.py 
432
433
434
435
@staticmethod
def is_available() -> bool:
    """Check if AsyncGroq is available."""
    return AsyncGroq is not None

AsyncOpenAIProvider

Bases: BaseProvider, AsyncProviderMixin

Async OpenAI provider.

Source code in llm_client/providers/async_providers.py 
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
class AsyncOpenAIProvider(BaseProvider, AsyncProviderMixin):
    """Async OpenAI provider."""

    def _initialize_client(self, **kwargs: Any) -> None:
        """Initialize async OpenAI client."""
        if not self.is_available():
            raise ProviderNotAvailableError("openai", "openai")

        api_key = kwargs.get("api_key")
        if not api_key:
            raise APIKeyNotFoundError("openai", "OPENAI_API_KEY")

        self.client = AsyncOpenAI(api_key=api_key)

    def _chat_completion_impl(self, messages: list[dict[str, str]]) -> str:
        """Sync method raises error - use async version."""
        raise RuntimeError(
            "AsyncOpenAIProvider only supports async methods. " "Use achat_completion() instead."
        )

    async def _achat_completion_impl(self, messages: list[dict[str, str]]) -> str:
        """Execute async chat completion with OpenAI."""
        if not self.client:
            raise RuntimeError("OpenAI client not initialized")

        response = await self.client.chat.completions.create(
            model=self.llm,
            messages=messages,
            temperature=self.temperature,
            max_tokens=self.max_tokens,
        )
        return response.choices[0].message.content

    async def _achat_completion_with_tools_impl(
        self,
        messages: list[dict[str, str]],
        tools: list[dict],
        tool_choice: str | dict | None = None,
    ) -> dict:
        """Execute async chat completion with tools using OpenAI."""
        if not self.client:
            raise RuntimeError("OpenAI client not initialized")

        kwargs = {
            "model": self.llm,
            "messages": messages,
            "temperature": self.temperature,
            "max_tokens": self.max_tokens,
            "tools": tools,
        }

        if tool_choice is not None:
            kwargs["tool_choice"] = tool_choice

        response = await self.client.chat.completions.create(**kwargs)

        choice = response.choices[0]
        return {
            "content": choice.message.content,
            "tool_calls": (
                [
                    {
                        "id": tc.id,
                        "type": tc.type,
                        "function": {"name": tc.function.name, "arguments": tc.function.arguments},
                    }
                    for tc in (choice.message.tool_calls or [])
                ]
                if choice.message.tool_calls
                else None
            ),
        }

    async def _achat_completion_with_files_impl(
        self,
        messages: list[dict[str, str]],
        files: list[str] | None = None,
    ) -> str:
        """Execute async chat completion with files using OpenAI."""
        if not self.client:
            raise RuntimeError("OpenAI client not initialized")

        enhanced_messages = messages.copy()

        if files:
            file_data = prepare_files_for_provider(files, "openai")

            if enhanced_messages and enhanced_messages[-1]["role"] == "user":
                last_msg = enhanced_messages[-1]
                if isinstance(last_msg["content"], str):
                    text_content = last_msg["content"]
                    last_msg["content"] = [{"type": "text", "text": text_content}]

                last_msg["content"].extend(file_data)
            else:
                enhanced_messages.append({"role": "user", "content": file_data})

        response = await self.client.chat.completions.create(
            model=self.llm,
            messages=enhanced_messages,
            temperature=self.temperature,
            max_tokens=self.max_tokens,
        )

        return response.choices[0].message.content

    async def _achat_completion_stream_impl(
        self, messages: list[dict[str, str]]
    ) -> AsyncIterator[str]:
        """Stream async chat completion with OpenAI."""
        if not self.client:
            raise RuntimeError("OpenAI client not initialized")

        stream = await self.client.chat.completions.create(
            model=self.llm,
            messages=messages,
            temperature=self.temperature,
            max_tokens=self.max_tokens,
            stream=True,
        )

        async for chunk in stream:
            if chunk.choices[0].delta.content is not None:
                yield chunk.choices[0].delta.content

    @staticmethod
    def get_default_model() -> str:
        """Get default OpenAI model."""
        return "gpt-4o-mini"

    @staticmethod
    def is_available() -> bool:
        """Check if AsyncOpenAI is available."""
        return AsyncOpenAI is not None
Functions
get_default_model() staticmethod

Get default OpenAI model.

Source code in llm_client/providers/async_providers.py 
286
287
288
289
@staticmethod
def get_default_model() -> str:
    """Get default OpenAI model."""
    return "gpt-4o-mini"
is_available() staticmethod

Check if AsyncOpenAI is available.

Source code in llm_client/providers/async_providers.py 
291
292
293
294
@staticmethod
def is_available() -> bool:
    """Check if AsyncOpenAI is available."""
    return AsyncOpenAI is not None

AsyncProviderMixin

Mixin for async provider functionality.

This mixin adds async chat completion methods to providers.

Source code in llm_client/providers/async_providers.py 
 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
class AsyncProviderMixin:
    """Mixin for async provider functionality.

    This mixin adds async chat completion methods to providers.
    """

    async def achat_completion(self, messages: list[dict[str, str]]) -> str:
        """Execute an async chat completion request.

        Args:
            messages: List of message dictionaries.

        Returns:
            Generated text response.

        Raises:
            ChatCompletionError: If the API call fails.

        Examples:
            >>> messages = [{"role": "user", "content": "Hello"}]
            >>> response = await provider.achat_completion(messages)
        """
        try:
            return await self._achat_completion_impl(messages)
        except Exception as e:
            raise ChatCompletionError(self.__class__.__name__, e) from e

    async def _achat_completion_impl(self, messages: list[dict[str, str]]) -> str:
        """Internal implementation of async chat completion.

        Must be implemented by concrete providers.

        Args:
            messages: List of message dictionaries.

        Returns:
            Generated text response.
        """
        raise NotImplementedError("Async chat completion not implemented")

    async def achat_completion_with_tools(
        self,
        messages: list[dict[str, str]],
        tools: list[dict],
        tool_choice: str | dict | None = None,
    ) -> dict:
        """Execute async chat completion with tools."""
        try:
            return await self._achat_completion_with_tools_impl(messages, tools, tool_choice)
        except NotImplementedError as err:
            raise NotImplementedError(
                f"{self.__class__.__name__} does not support tool calling"
            ) from err
        except Exception as e:
            raise ChatCompletionError(self.__class__.__name__, e) from e

    async def _achat_completion_with_tools_impl(
        self,
        messages: list[dict[str, str]],
        tools: list[dict],
        tool_choice: str | dict | None = None,
    ) -> dict:
        """Internal async implementation of tool calling."""
        raise NotImplementedError("Async tool calling not implemented")

    async def achat_completion_with_files(
        self,
        messages: list[dict[str, str]],
        files: list[str] | None = None,
    ) -> str:
        """Execute async chat completion with files."""
        try:
            return await self._achat_completion_with_files_impl(messages, files)
        except NotImplementedError as err:
            raise NotImplementedError(
                f"{self.__class__.__name__} does not support file uploads"
            ) from err
        except Exception as e:
            raise ChatCompletionError(self.__class__.__name__, e) from e

    async def _achat_completion_with_files_impl(
        self,
        messages: list[dict[str, str]],
        files: list[str] | None = None,
    ) -> str:
        """Internal async implementation of file upload."""
        raise NotImplementedError("Async file upload not implemented")

    async def achat_completion_stream(self, messages: list[dict[str, str]]) -> AsyncIterator[str]:
        """Stream response tokens asynchronously.

        Args:
            messages: List of message dictionaries.

        Yields:
            Individual tokens or chunks of response text.

        Raises:
            StreamingNotSupportedError: If streaming not supported.
            ChatCompletionError: If the streaming API call fails.

        Examples:
            >>> messages = [{"role": "user", "content": "Tell me a story"}]
            >>> async for chunk in provider.achat_completion_stream(messages):
            ...     print(chunk, end="", flush=True)
        """
        try:
            async for chunk in self._achat_completion_stream_impl(messages):
                yield chunk
        except NotImplementedError as err:
            raise StreamingNotSupportedError(
                self.__class__.__name__, "Provider does not implement async streaming"
            ) from err
        except Exception as e:
            raise ChatCompletionError(self.__class__.__name__, e) from e

    async def _achat_completion_stream_impl(
        self, messages: list[dict[str, str]]
    ) -> AsyncIterator[str]:
        """Internal implementation of async streaming.

        Can be overridden by concrete providers.

        Args:
            messages: List of message dictionaries.

        Yields:
            Individual tokens or chunks.
        """
        raise NotImplementedError("Async streaming not implemented")
        # Make this a generator
        yield  # pragma: no cover
Functions
achat_completion(messages) async

Execute an async chat completion request.

Parameters:

Name Type Description Default
messages list[dict[str, str]]

List of message dictionaries.

 required

Returns:

Type Description
str

Generated text response.

Raises:

Type Description
ChatCompletionError

If the API call fails.

Examples:

>>> messages = [{"role": "user", "content": "Hello"}]
>>> response = await provider.achat_completion(messages)
Source code in llm_client/providers/async_providers.py 
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
async def achat_completion(self, messages: list[dict[str, str]]) -> str:
    """Execute an async chat completion request.

    Args:
        messages: List of message dictionaries.

    Returns:
        Generated text response.

    Raises:
        ChatCompletionError: If the API call fails.

    Examples:
        >>> messages = [{"role": "user", "content": "Hello"}]
        >>> response = await provider.achat_completion(messages)
    """
    try:
        return await self._achat_completion_impl(messages)
    except Exception as e:
        raise ChatCompletionError(self.__class__.__name__, e) from e
achat_completion_stream(messages) async

Stream response tokens asynchronously.

Parameters:

Name Type Description Default
messages list[dict[str, str]]

List of message dictionaries.

 required

Yields:

Type Description
AsyncIterator[str]

Individual tokens or chunks of response text.

Raises:

Type Description
StreamingNotSupportedError

If streaming not supported.
ChatCompletionError

If the streaming API call fails.

Examples:

>>> messages = [{"role": "user", "content": "Tell me a story"}]
>>> async for chunk in provider.achat_completion_stream(messages):
...     print(chunk, end="", flush=True)
Source code in llm_client/providers/async_providers.py 
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
async def achat_completion_stream(self, messages: list[dict[str, str]]) -> AsyncIterator[str]:
    """Stream response tokens asynchronously.

    Args:
        messages: List of message dictionaries.

    Yields:
        Individual tokens or chunks of response text.

    Raises:
        StreamingNotSupportedError: If streaming not supported.
        ChatCompletionError: If the streaming API call fails.

    Examples:
        >>> messages = [{"role": "user", "content": "Tell me a story"}]
        >>> async for chunk in provider.achat_completion_stream(messages):
        ...     print(chunk, end="", flush=True)
    """
    try:
        async for chunk in self._achat_completion_stream_impl(messages):
            yield chunk
    except NotImplementedError as err:
        raise StreamingNotSupportedError(
            self.__class__.__name__, "Provider does not implement async streaming"
        ) from err
    except Exception as e:
        raise ChatCompletionError(self.__class__.__name__, e) from e
achat_completion_with_files(messages, files=None) async

Execute async chat completion with files.

Source code in llm_client/providers/async_providers.py 
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
async def achat_completion_with_files(
    self,
    messages: list[dict[str, str]],
    files: list[str] | None = None,
) -> str:
    """Execute async chat completion with files."""
    try:
        return await self._achat_completion_with_files_impl(messages, files)
    except NotImplementedError as err:
        raise NotImplementedError(
            f"{self.__class__.__name__} does not support file uploads"
        ) from err
    except Exception as e:
        raise ChatCompletionError(self.__class__.__name__, e) from e
achat_completion_with_tools(messages, tools, tool_choice=None) async

Execute async chat completion with tools.

Source code in llm_client/providers/async_providers.py 
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
async def achat_completion_with_tools(
    self,
    messages: list[dict[str, str]],
    tools: list[dict],
    tool_choice: str | dict | None = None,
) -> dict:
    """Execute async chat completion with tools."""
    try:
        return await self._achat_completion_with_tools_impl(messages, tools, tool_choice)
    except NotImplementedError as err:
        raise NotImplementedError(
            f"{self.__class__.__name__} does not support tool calling"
        ) from err
    except Exception as e:
        raise ChatCompletionError(self.__class__.__name__, e) from e

Functions