xinference 0.14.4.post1__py3-none-any.whl → 0.15.1__py3-none-any.whl

xinference/model/llm/transformers/cogvlm2.py

@@ -12,7 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import logging
-import time
 import uuid
 from concurrent.futures import ThreadPoolExecutor
 from typing import Dict, Iterator, List, Optional, Tuple, Union
@@ -21,17 +20,14 @@ import torch
 
 from ....core.scheduler import InferenceRequest
 from ....model.utils import select_device
-from ....types import (
-    ChatCompletion,
-    ChatCompletionChunk,
-    ChatCompletionMessage,
-    Completion,
-    CompletionChoice,
-    CompletionChunk,
-    CompletionUsage,
-)
+from ....types import ChatCompletion, ChatCompletionChunk, CompletionChunk
 from ..llm_family import LLMFamilyV1, LLMSpecV1
-from ..utils import _decode_image
+from ..utils import (
+    _decode_image,
+    generate_chat_completion,
+    generate_completion_chunk,
+    parse_messages,
+)
 from .core import PytorchChatModel, PytorchGenerateConfig
 from .utils import get_max_src_len
 
@@ -139,9 +135,7 @@ class CogVLM2Model(PytorchChatModel):
                 )
         return content, None
 
-    def _history_content_to_cogvlm2(
-        self, system_prompt: str, chat_history: List[ChatCompletionMessage]
-    ):
+    def _history_content_to_cogvlm2(self, system_prompt: str, chat_history: List[Dict]):
         query = system_prompt
         history: List[Tuple] = []
         pixel_values = None
@@ -163,7 +157,7 @@ class CogVLM2Model(PytorchChatModel):
         self,
         prompt: Union[str, List[Dict]],
         system_prompt: Optional[str] = None,
-        chat_history: Optional[List[ChatCompletionMessage]] = None,
+        chat_history: Optional[List[Dict]] = None,
     ):
         content, image = self._message_content_to_cogvlm2(prompt)
 
@@ -184,12 +178,12 @@ class CogVLM2Model(PytorchChatModel):
 
     def chat(
         self,
-        prompt: Union[str, List[Dict]],
-        system_prompt: Optional[str] = None,
-        chat_history: Optional[List[ChatCompletionMessage]] = None,
+        messages: List[Dict],
         generate_config: Optional[PytorchGenerateConfig] = None,
     ) -> Union[ChatCompletion, Iterator[ChatCompletionChunk]]:
-        system_prompt = system_prompt if system_prompt else ""
+        system_prompt = ""
+        if messages[0]["role"] == "system":
+            system_prompt = messages[0]["content"]
         stream = generate_config.get("stream", False) if generate_config else False
 
         sanitized_config = {
@@ -199,6 +193,7 @@ class CogVLM2Model(PytorchChatModel):
             else 512,
         }
 
+        prompt, _, chat_history = parse_messages(messages)
         query, image, history = self.get_query_and_history(
             prompt, system_prompt=system_prompt, chat_history=chat_history
         )
@@ -236,21 +231,7 @@ class CogVLM2Model(PytorchChatModel):
                 response = self._tokenizer.decode(outputs[0])
                 response = response.split("<|end_of_text|>")[0]
 
-            chunk = Completion(
-                id=str(uuid.uuid1()),
-                object="text_completion",
-                created=int(time.time()),
-                model=self.model_uid,
-                choices=[
-                    CompletionChoice(
-                        index=0, text=response, finish_reason="stop", logprobs=None
-                    )
-                ],
-                usage=CompletionUsage(
-                    prompt_tokens=-1, completion_tokens=-1, total_tokens=-1
-                ),
-            )
-            return self._to_chat_completion(chunk)
+            return generate_chat_completion(self.model_uid, response)
 
     def _streaming_chat_response(
         self, inputs: Dict, config: Dict
@@ -277,36 +258,26 @@ class CogVLM2Model(PytorchChatModel):
 
         completion_id = str(uuid.uuid1())
         for new_text in streamer:
-            chunk = CompletionChunk(
-                id=completion_id,
-                object="text_completion",
-                created=int(time.time()),
-                model=self.model_uid,
-                choices=[
-                    CompletionChoice(
-                        index=0, text=new_text, finish_reason=None, logprobs=None
-                    )
-                ],
-                usage=CompletionUsage(
-                    prompt_tokens=-1, completion_tokens=-1, total_tokens=-1
-                ),
+            yield generate_completion_chunk(
+                chunk_text=new_text,
+                finish_reason=None,
+                chunk_id=completion_id,
+                model_uid=self.model_uid,
+                prompt_tokens=-1,
+                completion_tokens=-1,
+                total_tokens=-1,
             )
-            yield chunk
-
-        completion_choice = CompletionChoice(
-            text="", index=0, logprobs=None, finish_reason="stop"
-        )
-        chunk = CompletionChunk(
-            id=completion_id,
-            object="text_completion",
-            created=int(time.time()),
-            model=self.model_uid,
-            choices=[completion_choice],
-            usage=CompletionUsage(
-                prompt_tokens=-1, completion_tokens=-1, total_tokens=-1
-            ),
+        yield generate_completion_chunk(
+            chunk_text=None,
+            finish_reason="stop",
+            chunk_id=completion_id,
+            model_uid=self.model_uid,
+            prompt_tokens=-1,
+            completion_tokens=-1,
+            total_tokens=-1,
+            has_choice=True,
+            has_content=False,
         )
-        yield chunk
 
     @staticmethod
     def build_position_ids(x, attention_mask=None):
@@ -341,7 +312,9 @@ class CogVLM2Model(PytorchChatModel):
     def get_dtype(self):
         return self._torch_type
 
-    def _get_full_prompt(self, prompt, system_prompt, chat_history, tools):
+    def _get_full_prompt(self, messages: List[Dict], tools):
+        prompt, system_prompt, chat_history = parse_messages(messages)
+        system_prompt = system_prompt or ""
         query, image, history = self.get_query_and_history(
             prompt, system_prompt=system_prompt, chat_history=chat_history
         )

xinference/model/llm/transformers/cogvlm2_video.py

@@ -12,28 +12,22 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import logging
-import time
 import uuid
 from concurrent.futures import ThreadPoolExecutor
 from typing import Dict, Iterator, List, Optional, Tuple, Union
 
 import torch
 
-from ....core.scheduler import InferenceRequest
 from ....model.utils import select_device
-from ....types import (
-    ChatCompletion,
-    ChatCompletionChunk,
-    ChatCompletionMessage,
-    Completion,
-    CompletionChoice,
-    CompletionChunk,
-    CompletionUsage,
-)
+from ....types import ChatCompletion, ChatCompletionChunk, CompletionChunk
 from ..llm_family import LLMFamilyV1, LLMSpecV1
-from ..utils import _decode_image
+from ..utils import (
+    _decode_image,
+    generate_chat_completion,
+    generate_completion_chunk,
+    parse_messages,
+)
 from .core import PytorchChatModel, PytorchGenerateConfig
-from .utils import get_max_src_len
 
 logger = logging.getLogger(__name__)
 
@@ -170,9 +164,7 @@ class CogVLM2VideoModel(PytorchChatModel):
             return text, images, video
         return content, [], None
 
-    def _history_content_to_cogvlm2(
-        self, system_prompt: str, chat_history: List[ChatCompletionMessage]
-    ):
+    def _history_content_to_cogvlm2(self, system_prompt: str, chat_history: List[Dict]):
         query = system_prompt
         history: List[Tuple] = []
         pixel_values = None
@@ -202,7 +194,7 @@ class CogVLM2VideoModel(PytorchChatModel):
         self,
         prompt: Union[str, List[Dict]],
         system_prompt: Optional[str] = None,
-        chat_history: Optional[List[ChatCompletionMessage]] = None,
+        chat_history: Optional[List[Dict]] = None,
     ):
         content, image, video = self._message_content_to_cogvlm2(prompt)
 
@@ -237,12 +229,12 @@ class CogVLM2VideoModel(PytorchChatModel):
 
     def chat(
         self,
-        prompt: Union[str, List[Dict]],
-        system_prompt: Optional[str] = None,
-        chat_history: Optional[List[ChatCompletionMessage]] = None,
+        messages: List[Dict],
         generate_config: Optional[PytorchGenerateConfig] = None,
     ) -> Union[ChatCompletion, Iterator[ChatCompletionChunk]]:
-        system_prompt = system_prompt if system_prompt else ""
+        system_prompt = ""
+        if messages[0]["role"] == "system":
+            system_prompt = messages[0]["content"]
         stream = generate_config.get("stream", False) if generate_config else False
 
         sanitized_config = {
@@ -252,6 +244,7 @@ class CogVLM2VideoModel(PytorchChatModel):
             else 512,
         }
 
+        prompt, _, chat_history = parse_messages(messages)
         query, image, video, history = self.get_query_and_history(
             prompt, system_prompt=system_prompt, chat_history=chat_history
         )
@@ -292,21 +285,7 @@ class CogVLM2VideoModel(PytorchChatModel):
                 response = self._tokenizer.decode(outputs[0])
                 response = response.split("<|end_of_text|>")[0]
 
-            chunk = Completion(
-                id=str(uuid.uuid1()),
-                object="text_completion",
-                created=int(time.time()),
-                model=self.model_uid,
-                choices=[
-                    CompletionChoice(
-                        index=0, text=response, finish_reason="stop", logprobs=None
-                    )
-                ],
-                usage=CompletionUsage(
-                    prompt_tokens=-1, completion_tokens=-1, total_tokens=-1
-                ),
-            )
-            return self._to_chat_completion(chunk)
+            return generate_chat_completion(self.model_uid, response)
 
     def _streaming_chat_response(
         self, inputs: Dict, config: Dict
@@ -333,192 +312,23 @@ class CogVLM2VideoModel(PytorchChatModel):
 
         completion_id = str(uuid.uuid1())
         for new_text in streamer:
-            chunk = CompletionChunk(
-                id=completion_id,
-                object="text_completion",
-                created=int(time.time()),
-                model=self.model_uid,
-                choices=[
-                    CompletionChoice(
-                        index=0, text=new_text, finish_reason=None, logprobs=None
-                    )
-                ],
-                usage=CompletionUsage(
-                    prompt_tokens=-1, completion_tokens=-1, total_tokens=-1
-                ),
+            yield generate_completion_chunk(
+                chunk_text=new_text,
+                finish_reason=None,
+                chunk_id=completion_id,
+                model_uid=self.model_uid,
+                prompt_tokens=-1,
+                completion_tokens=-1,
+                total_tokens=-1,
             )
-            yield chunk
-
-        completion_choice = CompletionChoice(
-            text="", index=0, logprobs=None, finish_reason="stop"
-        )
-        chunk = CompletionChunk(
-            id=completion_id,
-            object="text_completion",
-            created=int(time.time()),
-            model=self.model_uid,
-            choices=[completion_choice],
-            usage=CompletionUsage(
-                prompt_tokens=-1, completion_tokens=-1, total_tokens=-1
-            ),
-        )
-        yield chunk
-
-    @staticmethod
-    def build_position_ids(x, attention_mask=None):
-        """
-        Copied from https://huggingface.co/THUDM/cogvlm2-llama3-chinese-chat-19B-int4/blob/main/modeling_cogvlm.py
-        """
-        # Fix: 参考官方开源代码
-        if attention_mask is not None:
-            tmp = x.clone()
-            tmp[~(attention_mask.bool())] = -1
-        else:
-            tmp = x.clone()
-        # image boi eoi token as LANGUAGE_TOKEN_TYPE
-        is_boi_eoi = torch.zeros_like(x, dtype=torch.bool)
-        is_boi_eoi[:, 1:] |= (tmp[:, 1:] == VISION_TOKEN_TYPE) & (
-            tmp[:, :-1] == LANGUAGE_TOKEN_TYPE
-        )
-        is_boi_eoi[:, 0] |= tmp[:, 0] == VISION_TOKEN_TYPE
-        is_boi_eoi[:, :-1] |= (tmp[:, :-1] == VISION_TOKEN_TYPE) & (
-            tmp[:, 1:] == LANGUAGE_TOKEN_TYPE
+        yield generate_completion_chunk(
+            chunk_text=None,
+            finish_reason="stop",
+            chunk_id=completion_id,
+            model_uid=self.model_uid,
+            prompt_tokens=-1,
+            completion_tokens=-1,
+            total_tokens=-1,
+            has_choice=True,
+            has_content=False,
         )
-        is_boi_eoi[:, -1] |= tmp[:, -1] == VISION_TOKEN_TYPE
-        tmp[is_boi_eoi] = LANGUAGE_TOKEN_TYPE
-        # final position ids
-        y = torch.zeros_like(x, dtype=torch.long)
-        y[:, 1:] = (tmp[:, 1:] == LANGUAGE_TOKEN_TYPE) | (
-            (tmp[:, 1:] == VISION_TOKEN_TYPE) & (tmp[:, :-1] == LANGUAGE_TOKEN_TYPE)
-        )
-        y = y.cumsum(dim=-1)
-        return y
-
-    def get_dtype(self):
-        return self._torch_type
-
-    def _get_full_prompt(self, prompt, system_prompt, chat_history, tools):
-        query, image, video, history = self.get_query_and_history(
-            prompt, system_prompt=system_prompt, chat_history=chat_history
-        )
-
-        if video:
-            image = [video]
-
-        input_by_model: dict = self._model.build_conversation_input_ids(  # type: ignore
-            self._tokenizer,
-            query=query,
-            history=history,
-            images=image,
-            template_version="chat",
-        )
-        return {
-            "input_ids": input_by_model["input_ids"],  # seq_len
-            "token_type_ids": input_by_model["token_type_ids"],  # seq_len
-            "attention_mask": input_by_model["attention_mask"],  # seq_len
-            "images": input_by_model["images"],
-        }
-
-    def prepare_sanitize_generate_config(self, req: InferenceRequest):
-        """
-        See https://huggingface.co/THUDM/cogvlm2-llama3-chat-19B/blob/main/generation_config.json
-        """
-        raw_config = req.inference_kwargs.get("raw_params", {})
-        temperature = raw_config.get("temperature", None)
-        if temperature is None:
-            raw_config["temperature"] = 0.6
-        top_p = raw_config.get("top_p", None)
-        if top_p is None:
-            raw_config["top_p"] = 0.9
-        return raw_config
-
-    def build_prefill_kwargs(self, prompts: List, req_list: List[InferenceRequest]):
-        context_len = self.get_context_len()
-        assert isinstance(prompts[0], dict)
-        images = []
-        max_length = float("-inf")
-        for i, feature in enumerate(prompts):
-            req = req_list[i]
-            if "images" in feature:
-                images.append(feature.pop("images", None))
-            max_src_len = get_max_src_len(context_len, req)
-            input_ids = feature["input_ids"][-max_src_len:]
-            req.prompt_tokens = input_ids.tolist()
-            feature["input_ids"] = input_ids
-            feature["token_type_ids"] = feature["token_type_ids"][-max_src_len:]
-            feature["attention_mask"] = feature["attention_mask"][-max_src_len:]
-            req.extra_kwargs["attention_mask_seq_len"] = feature[
-                "attention_mask"
-            ].shape[0]
-            max_length = max(len(input_ids), max_length)
-
-        def pad_to_max_length_internal(feature, max_len, idx):
-            padding_length = max_len - len(feature["input_ids"])
-            req_list[idx].padding_len = padding_length
-            feature["input_ids"] = torch.cat(
-                [torch.full((padding_length,), 0), feature["input_ids"]]
-            )
-            feature["token_type_ids"] = torch.cat(
-                [
-                    torch.zeros(padding_length, dtype=torch.long),
-                    feature["token_type_ids"],
-                ]
-            )
-            feature["attention_mask"] = torch.cat(
-                [
-                    torch.zeros(padding_length, dtype=torch.long),
-                    feature["attention_mask"],
-                ]
-            )
-            return feature
-
-        features = [
-            pad_to_max_length_internal(feature, max_length, i)
-            for i, feature in enumerate(prompts)
-        ]
-        batch = {
-            key: torch.stack([feature[key] for feature in features])
-            for key in features[0].keys()
-        }
-
-        position_ids = self.build_position_ids(batch["token_type_ids"])
-        batch["position_ids"] = position_ids
-
-        for i in range(len(prompts)):
-            req = req_list[i]
-            req.extra_kwargs["max_position_id"] = position_ids[i : i + 1, -1].item()
-
-        if images:
-            batch["images"] = images
-
-        batch = recur_move_to(
-            batch, self._device, lambda x: isinstance(x, torch.Tensor)
-        )
-        dtype = self.get_dtype()
-        if dtype:
-            batch = recur_move_to(
-                batch,
-                dtype,
-                lambda x: isinstance(x, torch.Tensor) and torch.is_floating_point(x),
-            )
-        return batch
-
-    def build_decode_token_type_ids(
-        self, batch_size: int, seq_length: int, reqs: List[InferenceRequest]
-    ):
-        token_type_ids = torch.full(
-            (batch_size, 1), fill_value=1, dtype=torch.long, device=self._device
-        )
-        return token_type_ids
-
-    def build_decode_position_ids(
-        self, batch_size: int, seq_length: int, reqs: List[InferenceRequest]
-    ):
-        tmp = []
-        for r in reqs:
-            r.extra_kwargs["max_position_id"] += 1
-            tmp.append(r.extra_kwargs["max_position_id"])
-        position_ids = torch.as_tensor(
-            tmp, device=self._device, dtype=torch.long
-        ).unsqueeze(1)
-        return position_ids

xinference/model/llm/transformers/core.py

@@ -16,7 +16,7 @@ import json
 import logging
 import os
 from functools import lru_cache
-from typing import Iterable, Iterator, List, Optional, Tuple, Union
+from typing import Dict, Iterable, Iterator, List, Optional, Tuple, Union
 
 import torch
 
@@ -29,7 +29,6 @@ from ....device_utils import (
 from ....types import (
     ChatCompletion,
     ChatCompletionChunk,
-    ChatCompletionMessage,
     Completion,
     CompletionChoice,
     CompletionChunk,
@@ -52,8 +51,6 @@ NON_DEFAULT_MODEL_LIST: List[str] = [
     "chatglm3-128k",
     "glm4-chat",
     "glm4-chat-1m",
-    "llama-2",
-    "llama-2-chat",
     "internlm2-chat",
     "internlm2.5-chat",
     "qwen-vl-chat",
@@ -67,6 +64,13 @@ NON_DEFAULT_MODEL_LIST: List[str] = [
     "MiniCPM-Llama3-V-2_5",
     "MiniCPM-V-2.6",
     "glm-4v",
+    "qwen2-vl-instruct",
+    "qwen2-audio",
+    "qwen2-audio-instruct",
+    "deepseek-v2",
+    "deepseek-v2-chat",
+    "deepseek-v2.5",
+    "deepseek-v2-chat-0628",
 ]
 
 
@@ -615,12 +619,17 @@ class PytorchModel(LLM):
                 r.error_msg = str(e)
 
     def get_builtin_stop_token_ids(self) -> Tuple:
-        return (
-            tuple(self.model_family.prompt_style.stop_token_ids)
-            if self.model_family.prompt_style
-            and self.model_family.prompt_style.stop_token_ids
-            else tuple()
-        )
+        from ..utils import get_stop_token_ids_from_config_file
+
+        stop_token_ids = get_stop_token_ids_from_config_file(self.model_path)
+        if stop_token_ids is not None:
+            return tuple(stop_token_ids)
+        else:
+            return (
+                tuple(self.model_family.stop_token_ids)
+                if self.model_family.stop_token_ids
+                else tuple()
+            )
 
     def handle_batch_inference_results(self, req_list: List[InferenceRequest]):
         for req in req_list:
@@ -693,20 +702,13 @@ class PytorchChatModel(PytorchModel, ChatModelMixin):
         generate_config: Optional[PytorchGenerateConfig],
     ) -> PytorchGenerateConfig:
         generate_config = super()._sanitize_generate_config(generate_config)
-        if (
-            (not generate_config.get("stop"))
-            and self.model_family.prompt_style
-            and self.model_family.prompt_style.stop
-        ):
-            generate_config["stop"] = self.model_family.prompt_style.stop.copy()
+        if (not generate_config.get("stop")) and self.model_family.stop is not None:
+            generate_config["stop"] = self.model_family.stop.copy()
         if (
             generate_config.get("stop_token_ids", None) is None
-            and self.model_family.prompt_style
-            and self.model_family.prompt_style.stop_token_ids
+            and self.model_family.stop_token_ids is not None
         ):
-            generate_config[
-                "stop_token_ids"
-            ] = self.model_family.prompt_style.stop_token_ids.copy()
+            generate_config["stop_token_ids"] = self.model_family.stop_token_ids.copy()
 
         return generate_config
 
@@ -725,26 +727,23 @@ class PytorchChatModel(PytorchModel, ChatModelMixin):
 
     def chat(
         self,
-        prompt: str,
-        system_prompt: Optional[str] = None,
-        chat_history: Optional[List[ChatCompletionMessage]] = None,
+        messages: List[Dict],
         generate_config: Optional[PytorchGenerateConfig] = None,
     ) -> Union[ChatCompletion, Iterator[ChatCompletionChunk]]:
         tools = generate_config.pop("tools", []) if generate_config else None
-        full_prompt = self._get_full_prompt(prompt, system_prompt, chat_history, tools)
-
-        generate_config = self._sanitize_generate_config(generate_config)
-        # TODO(codingl2k1): qwen hacky to set stop for function call.
         model_family = self.model_family.model_family or self.model_family.model_name
+        full_context_kwargs = {}
         if tools and model_family in QWEN_TOOL_CALL_FAMILY:
-            stop = generate_config.get("stop")
-            if isinstance(stop, str):
-                generate_config["stop"] = [stop, "Observation:"]
-            elif isinstance(stop, Iterable):
-                assert not isinstance(stop, str)
-                generate_config["stop"] = list(stop) + ["Observation:"]
-            else:
-                generate_config["stop"] = "Observation:"
+            full_context_kwargs["tools"] = tools
+        assert self.model_family.chat_template is not None
+        full_prompt = self.get_full_context(
+            messages,
+            self.model_family.chat_template,
+            tokenizer=self._tokenizer,
+            **full_context_kwargs,
+        )
+
+        generate_config = self._sanitize_generate_config(generate_config)
 
         stream = generate_config.get("stream", False)
         if stream:
@@ -755,22 +754,16 @@ class PytorchChatModel(PytorchModel, ChatModelMixin):
             c = self.generate(full_prompt, generate_config)
             assert not isinstance(c, Iterator)
             if tools:
-                return self._tool_calls_completion(
-                    self.model_family, self.model_uid, c, tools
-                )
+                return self._tool_calls_completion(self.model_family, self.model_uid, c)
             return self._to_chat_completion(c)
 
     def load(self):
         super().load()
 
-    def _get_full_prompt(self, prompt, system_prompt, chat_history, tools):
-        assert self.model_family.prompt_style is not None
-        prompt_style = self.model_family.prompt_style.copy()
-        if system_prompt:
-            prompt_style.system_prompt = system_prompt
-        chat_history = chat_history or []
-        full_prompt = ChatModelMixin.get_prompt(
-            prompt, chat_history, prompt_style, tools=tools
+    def _get_full_prompt(self, messages: List[Dict], tools):
+        assert self.model_family.chat_template is not None
+        full_prompt = self.get_full_context(
+            messages, self.model_family.chat_template, tokenizer=self._tokenizer
         )
         return full_prompt
 
@@ -779,9 +772,7 @@ class PytorchChatModel(PytorchModel, ChatModelMixin):
         for r in req_list:
             try:
                 if not r.stopped and r.is_prefill:
-                    r.full_prompt = self._get_full_prompt(
-                        r.prompt, r.system_prompt, r.chat_history, None
-                    )
+                    r.full_prompt = self._get_full_prompt(r.prompt, None)
             except Exception as e:
                 logger.exception(f"prepare inference error with {e}")
                 r.stopped = True
@@ -790,6 +781,20 @@ class PytorchChatModel(PytorchModel, ChatModelMixin):
     def handle_batch_inference_results(self, req_list: List[InferenceRequest]):
         for req in req_list:
             if req.error_msg is None and req.completion:
+                # The `generate` function can be called for some chat models.
+                # So that we cannot convert completion chunk to chat completion chunk.
+                if req.call_ability == "generate":
+                    results = []
+                    for c in req.completion:
+                        if c == "<bos_stream>":
+                            continue
+                        elif c == "<eos_stream>":
+                            break
+                        else:
+                            results.append(c)
+                    req.completion = results
+                    continue
+
                 if req.stream:
                     results = []
                     for i, c in enumerate(req.completion):