diff --git a/Changelog.md b/Changelog.md
index 473cc69..4119790 100644
--- a/Changelog.md
+++ b/Changelog.md
@@ -1,5 +1,11 @@
 # Changelog
 
+## 4.0.0
+
+- Turned all NamedTuples into Dataclasses. Even if this is technically a breaking change
+  probably not much source code is actually affected as constructing instances still behaves
+  the same.
+
 ## 3.5.1
 
 - Fix failing serialization of Prompt-based Documents in QA requests.
diff --git a/aleph_alpha_client/aleph_alpha_client.py b/aleph_alpha_client/aleph_alpha_client.py
index e294957..6c4f357 100644
--- a/aleph_alpha_client/aleph_alpha_client.py
+++ b/aleph_alpha_client/aleph_alpha_client.py
@@ -214,7 +214,7 @@ def _build_query_parameters(self) -> Mapping[str, str]:
     def _build_json_body(
         self, request: AnyRequest, model: Optional[str]
     ) -> Mapping[str, Any]:
-        json_body = request.to_json()
+        json_body = dict(request.to_json())
 
         if model is not None:
             json_body["model"] = model
@@ -721,7 +721,7 @@ def _build_query_parameters(self) -> Mapping[str, str]:
     def _build_json_body(
         self, request: AnyRequest, model: Optional[str]
     ) -> Mapping[str, Any]:
-        json_body = request.to_json()
+        json_body = dict(request.to_json())
 
         if model is not None:
             json_body["model"] = model
diff --git a/aleph_alpha_client/completion.py b/aleph_alpha_client/completion.py
index fbca704..ac48523 100644
--- a/aleph_alpha_client/completion.py
+++ b/aleph_alpha_client/completion.py
@@ -1,9 +1,11 @@
-from typing import Any, Dict, List, Mapping, NamedTuple, Optional, Sequence
+from dataclasses import asdict, dataclass
+from typing import Any, Dict, List, Mapping, Optional, Sequence
 
 from aleph_alpha_client.prompt import Prompt
 
 
-class CompletionRequest(NamedTuple):
+@dataclass(frozen=True)
+class CompletionRequest:
     """
     Describes a completion request
 
@@ -212,13 +214,17 @@ class CompletionRequest(NamedTuple):
     repetition_penalties_include_completion: bool = True
     raw_completion: bool = False
 
-    def to_json(self) -> Dict[str, Any]:
+    def to_json(self) -> Mapping[str, Any]:
         payload = {k: v for k, v in self._asdict().items() if v is not None}
         payload["prompt"] = self.prompt.to_json()
         return payload
 
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
+
 
-class CompletionResult(NamedTuple):
+@dataclass(frozen=True)
+class CompletionResult:
     log_probs: Optional[Sequence[Mapping[str, Optional[float]]]] = None
     completion: Optional[str] = None
     completion_tokens: Optional[Sequence[str]] = None
@@ -235,8 +241,12 @@ def from_json(json: Dict[str, Any]) -> "CompletionResult":
             raw_completion=json.get("raw_completion"),
         )
 
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
 
-class CompletionResponse(NamedTuple):
+
+@dataclass(frozen=True)
+class CompletionResponse:
     model_version: str
     completions: Sequence[CompletionResult]
     optimized_prompt: Optional[Prompt] = None
@@ -259,3 +269,6 @@ def to_json(self) -> Mapping[str, Any]:
             **self._asdict(),
             "completions": [completion._asdict() for completion in self.completions],
         }
+
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
diff --git a/aleph_alpha_client/detokenization.py b/aleph_alpha_client/detokenization.py
index e63a5aa..52bd974 100644
--- a/aleph_alpha_client/detokenization.py
+++ b/aleph_alpha_client/detokenization.py
@@ -1,7 +1,9 @@
-from typing import Any, Dict, List, NamedTuple, Optional, Sequence
+from dataclasses import dataclass, asdict
+from typing import Any, Dict, List, Mapping, Optional, Sequence
 
 
-class DetokenizationRequest(NamedTuple):
+@dataclass(frozen=True)
+class DetokenizationRequest:
     """Describes a detokenization request.
 
     Parameters
@@ -14,12 +16,16 @@ class DetokenizationRequest(NamedTuple):
 
     token_ids: Sequence[int]
 
-    def to_json(self) -> Dict[str, Any]:
+    def to_json(self) -> Mapping[str, Any]:
         payload = self._asdict()
         return payload
 
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
 
-class DetokenizationResponse(NamedTuple):
+
+@dataclass(frozen=True)
+class DetokenizationResponse:
     result: str
 
     @staticmethod
diff --git a/aleph_alpha_client/embedding.py b/aleph_alpha_client/embedding.py
index a5637d3..d0fdb41 100644
--- a/aleph_alpha_client/embedding.py
+++ b/aleph_alpha_client/embedding.py
@@ -1,9 +1,10 @@
+from dataclasses import asdict, dataclass
 from enum import Enum
 from typing import (
     Any,
     Dict,
     List,
-    NamedTuple,
+    Mapping,
     Optional,
     Sequence,
     Tuple,
@@ -11,7 +12,8 @@
 from aleph_alpha_client.prompt import Prompt
 
 
-class EmbeddingRequest(NamedTuple):
+@dataclass(frozen=True)
+class EmbeddingRequest:
     """
     Embeds a text and returns vectors that can be used for downstream tasks (e.g. semantic similarity) and models (e.g. classifiers).
 
@@ -73,13 +75,18 @@ class EmbeddingRequest(NamedTuple):
     contextual_control_threshold: Optional[float] = None
     control_log_additive: Optional[bool] = True
 
-    def to_json(self) -> Dict[str, Any]:
-        payload = self._asdict()
-        payload["prompt"] = self.prompt.to_json()
-        return payload
+    def to_json(self) -> Mapping[str, Any]:
+        return {
+            **self._asdict(),
+            "prompt": self.prompt.to_json(),
+        }
 
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
 
-class EmbeddingResponse(NamedTuple):
+
+@dataclass(frozen=True)
+class EmbeddingResponse:
     model_version: str
     embeddings: Optional[Dict[Tuple[str, str], List[float]]]
     tokens: Optional[List[str]]
@@ -120,7 +127,8 @@ class SemanticRepresentation(Enum):
     Query = "query"
 
 
-class SemanticEmbeddingRequest(NamedTuple):
+@dataclass(frozen=True)
+class SemanticEmbeddingRequest:
     """
     Embeds a text and returns vectors that can be used for downstream tasks (e.g. semantic similarity) and models (e.g. classifiers).
 
@@ -181,14 +189,19 @@ class SemanticEmbeddingRequest(NamedTuple):
     contextual_control_threshold: Optional[float] = None
     control_log_additive: Optional[bool] = True
 
-    def to_json(self) -> Dict[str, Any]:
-        payload = self._asdict()
-        payload["representation"] = self.representation.value
-        payload["prompt"] = self.prompt.to_json()
-        return payload
+    def to_json(self) -> Mapping[str, Any]:
+        return {
+            **self._asdict(),
+            "representation": self.representation.value,
+            "prompt": self.prompt.to_json(),
+        }
+
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
 
 
-class BatchSemanticEmbeddingRequest(NamedTuple):
+@dataclass(frozen=True)
+class BatchSemanticEmbeddingRequest:
     """
     Embeds multiple multi-modal prompts and returns their embeddings in the same order as they were supplied.
 
@@ -246,17 +259,22 @@ class BatchSemanticEmbeddingRequest(NamedTuple):
     contextual_control_threshold: Optional[float] = None
     control_log_additive: Optional[bool] = True
 
-    def to_json(self) -> Dict[str, Any]:
-        payload = self._asdict()
-        payload["representation"] = self.representation.value
-        payload["prompts"] = [prompt.to_json() for prompt in self.prompts]
-        return payload
+    def to_json(self) -> Mapping[str, Any]:
+        return {
+            **self._asdict(),
+            "representation": self.representation.value,
+            "prompts": [prompt.to_json() for prompt in self.prompts],
+        }
+
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
 
 
 EmbeddingVector = List[float]
 
 
-class SemanticEmbeddingResponse(NamedTuple):
+@dataclass(frozen=True)
+class SemanticEmbeddingResponse:
     """
     Response of a semantic embedding request
 
@@ -275,10 +293,15 @@ class SemanticEmbeddingResponse(NamedTuple):
 
     @staticmethod
     def from_json(json: Dict[str, Any]) -> "SemanticEmbeddingResponse":
-        return SemanticEmbeddingResponse(**json)
+        return SemanticEmbeddingResponse(
+            model_version=json["model_version"],
+            embedding=json["embedding"],
+            message=json.get("message"),
+        )
 
 
-class BatchSemanticEmbeddingResponse(NamedTuple):
+@dataclass(frozen=True)
+class BatchSemanticEmbeddingResponse:
     """
     Response of a batch semantic embedding request
 
@@ -294,7 +317,9 @@ class BatchSemanticEmbeddingResponse(NamedTuple):
 
     @staticmethod
     def from_json(json: Dict[str, Any]) -> "BatchSemanticEmbeddingResponse":
-        return BatchSemanticEmbeddingResponse(**json)
+        return BatchSemanticEmbeddingResponse(
+            model_version=json["model_version"], embeddings=json["embeddings"]
+        )
 
     @staticmethod
     def _from_model_version_and_embeddings(
diff --git a/aleph_alpha_client/evaluation.py b/aleph_alpha_client/evaluation.py
index 8d9377e..3f8ebf1 100644
--- a/aleph_alpha_client/evaluation.py
+++ b/aleph_alpha_client/evaluation.py
@@ -1,13 +1,15 @@
+from dataclasses import dataclass, asdict
 from typing import (
     Any,
     Dict,
-    NamedTuple,
+    Mapping,
     Optional,
 )
 from aleph_alpha_client.prompt import Prompt
 
 
-class EvaluationRequest(NamedTuple):
+@dataclass(frozen=True)
+class EvaluationRequest:
     """
     Evaluates the model's likelihood to produce a completion given a prompt.
 
@@ -39,13 +41,15 @@ class EvaluationRequest(NamedTuple):
     contextual_control_threshold: Optional[float] = None
     control_log_additive: Optional[bool] = True
 
-    def to_json(self) -> Dict[str, Any]:
-        payload = self._asdict()
-        payload["prompt"] = self.prompt.to_json()
-        return payload
+    def to_json(self) -> Mapping[str, Any]:
+        return {**self._asdict(), "prompt": self.prompt.to_json()}
 
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
 
-class EvaluationResponse(NamedTuple):
+
+@dataclass(frozen=True)
+class EvaluationResponse:
     model_version: str
     message: Optional[str]
     result: Dict[str, Any]
diff --git a/aleph_alpha_client/explanation.py b/aleph_alpha_client/explanation.py
index e2ca6ef..9c962c9 100644
--- a/aleph_alpha_client/explanation.py
+++ b/aleph_alpha_client/explanation.py
@@ -1,10 +1,10 @@
+from dataclasses import dataclass
 from enum import Enum
 from typing import (
     Any,
     List,
     Dict,
     Mapping,
-    NamedTuple,
     Optional,
     Union,
 )
@@ -34,7 +34,8 @@ def to_json(self) -> str:
         return self.value
 
 
-class CustomGranularity(NamedTuple):
+@dataclass(frozen=True)
+class CustomGranularity:
     """
     Allows for passing a custom delimiter to determine the granularity to
     to explain the prompt by. The text of the prompt will be split by the
@@ -58,7 +59,7 @@ class PromptGranularity(Enum):
     Sentence = "sentence"
     Paragraph = "paragraph"
 
-    def to_json(self):
+    def to_json(self) -> Mapping[str, Any]:
         return {"type": self.value}
 
 
@@ -90,7 +91,8 @@ def to_json(self) -> str:
         return self.value
 
 
-class ExplanationRequest(NamedTuple):
+@dataclass(frozen=True)
+class ExplanationRequest:
     """
     Describes an Explanation request you want to make agains the API.
 
@@ -177,7 +179,7 @@ class ExplanationRequest(NamedTuple):
     postprocessing: Optional[ExplanationPostprocessing] = None
     normalize: Optional[bool] = None
 
-    def to_json(self) -> Dict[str, Any]:
+    def to_json(self) -> Mapping[str, Any]:
         payload: Dict[str, Any] = {
             "prompt": self.prompt.to_json(),
             "target": self.target,
@@ -206,7 +208,8 @@ def to_json(self) -> Dict[str, Any]:
         return payload
 
 
-class TextScore(NamedTuple):
+@dataclass(frozen=True)
+class TextScore:
     start: int
     length: int
     score: float
@@ -220,7 +223,8 @@ def from_json(score: Any) -> "TextScore":
         )
 
 
-class TextScoreWithRaw(NamedTuple):
+@dataclass(frozen=True)
+class TextScoreWithRaw:
     start: int
     length: int
     score: float
@@ -236,7 +240,8 @@ def from_text_score(score: TextScore, prompt: Text) -> "TextScoreWithRaw":
         )
 
 
-class ImageScore(NamedTuple):
+@dataclass(frozen=True)
+class ImageScore:
     left: float
     top: float
     width: float
@@ -254,7 +259,8 @@ def from_json(score: Any) -> "ImageScore":
         )
 
 
-class TargetScore(NamedTuple):
+@dataclass(frozen=True)
+class TargetScore:
     start: int
     length: int
     score: float
@@ -268,7 +274,8 @@ def from_json(score: Any) -> "TargetScore":
         )
 
 
-class TargetScoreWithRaw(NamedTuple):
+@dataclass(frozen=True)
+class TargetScoreWithRaw:
     start: int
     length: int
     score: float
@@ -284,7 +291,8 @@ def from_target_score(score: TargetScore, target: str) -> "TargetScoreWithRaw":
         )
 
 
-class TokenScore(NamedTuple):
+@dataclass(frozen=True)
+class TokenScore:
     score: float
 
     @staticmethod
@@ -294,7 +302,8 @@ def from_json(score: Any) -> "TokenScore":
         )
 
 
-class ImagePromptItemExplanation(NamedTuple):
+@dataclass(frozen=True)
+class ImagePromptItemExplanation:
     """
     Explains the importance of an image prompt item.
     The amount of items in the "scores" array depends on the granularity setting.
@@ -328,7 +337,8 @@ def in_pixels(self, prompt_item: PromptItem) -> "ImagePromptItemExplanation":
         )
 
 
-class TextPromptItemExplanation(NamedTuple):
+@dataclass(frozen=True)
+class TextPromptItemExplanation:
     """
     Explains the importance of a text prompt item.
     The amount of items in the "scores" array depends on the granularity setting.
@@ -355,7 +365,8 @@ def with_text(self, prompt: Text) -> "TextPromptItemExplanation":
         )
 
 
-class TargetPromptItemExplanation(NamedTuple):
+@dataclass(frozen=True)
+class TargetPromptItemExplanation:
     """
     Explains the importance of text in the target string that came before the currently
     to-be-explained target token. The amount of items in the "scores" array depends on the
@@ -383,7 +394,8 @@ def with_text(self, prompt: str) -> "TargetPromptItemExplanation":
         )
 
 
-class TokenPromptItemExplanation(NamedTuple):
+@dataclass(frozen=True)
+class TokenPromptItemExplanation:
     """Explains the importance of a request prompt item of type "token_ids".
     Will contain one floating point importance value for each token in the same order as in the original prompt.
     """
@@ -397,7 +409,8 @@ def from_json(item: Dict[str, Any]) -> "TokenPromptItemExplanation":
         )
 
 
-class Explanation(NamedTuple):
+@dataclass(frozen=True)
+class Explanation:
     """
     Explanations for a given portion of the target.
 
@@ -482,7 +495,8 @@ def with_text_from_prompt(self, prompt: Prompt, target: str) -> "Explanation":
         )
 
 
-class ExplanationResponse(NamedTuple):
+@dataclass(frozen=True)
+class ExplanationResponse:
     """
     The top-level response data structure that will be returned from an explanation request.
 
diff --git a/aleph_alpha_client/prompt.py b/aleph_alpha_client/prompt.py
index fdab870..242cd78 100644
--- a/aleph_alpha_client/prompt.py
+++ b/aleph_alpha_client/prompt.py
@@ -8,7 +8,6 @@
     Dict,
     List,
     Mapping,
-    NamedTuple,
     Optional,
     Sequence,
     Tuple,
@@ -42,7 +41,8 @@ def to_json(self) -> str:
         return self.value
 
 
-class TokenControl(NamedTuple):
+@dataclass(frozen=True)
+class TokenControl:
     """
     Used for Attention Manipulation, for a given token index, you can supply
     the factor you want to adjust the attention by.
@@ -69,7 +69,8 @@ def to_json(self) -> Mapping[str, Any]:
         return {"index": self.pos, "factor": self.factor}
 
 
-class Tokens(NamedTuple):
+@dataclass(frozen=True)
+class Tokens:
     """
     A list of token ids to be sent as part of a prompt.
 
@@ -108,7 +109,8 @@ def from_token_ids(token_ids: Sequence[int]) -> "Tokens":
         return Tokens(token_ids, [])
 
 
-class TextControl(NamedTuple):
+@dataclass(frozen=True)
+class TextControl:
     """
     Attention manipulation for a Text PromptItem.
 
@@ -141,7 +143,7 @@ class TextControl(NamedTuple):
     factor: float
     token_overlap: Optional[ControlTokenOverlap] = None
 
-    def to_json(self) -> Dict[str, Any]:
+    def to_json(self) -> Mapping[str, Any]:
         payload: Dict[str, Any] = {
             "start": self.start,
             "length": self.length,
@@ -152,7 +154,8 @@ def to_json(self) -> Dict[str, Any]:
         return payload
 
 
-class Text(NamedTuple):
+@dataclass(frozen=True)
+class Text:
     """
     A Text-prompt including optional controls for attention manipulation.
 
@@ -186,7 +189,8 @@ def from_text(text: str) -> "Text":
         return Text(text, [])
 
 
-class Cropping(NamedTuple):
+@dataclass(frozen=True)
+class Cropping:
     """
     Describes a quadratic crop of the file.
     """
@@ -196,7 +200,8 @@ class Cropping(NamedTuple):
     size: int
 
 
-class ImageControl(NamedTuple):
+@dataclass(frozen=True)
+class ImageControl:
     """
     Attention manipulation for an Image PromptItem.
 
@@ -261,7 +266,8 @@ def to_json(self) -> Mapping[str, Any]:
         return payload
 
 
-class Image(NamedTuple):
+@dataclass(frozen=True)
+class Image:
     """
     An image send as part of a prompt to a model. The image is represented as
     base64.
@@ -395,7 +401,7 @@ def _get_url(cls, url: str) -> bytes:
         response.raise_for_status()
         return response.content
 
-    def to_json(self) -> Dict[str, Any]:
+    def to_json(self) -> Mapping[str, Any]:
         """
         A dict if serialized to JSON is suitable as a prompt element
         """
diff --git a/aleph_alpha_client/qa.py b/aleph_alpha_client/qa.py
index ecf0411..d8e8891 100644
--- a/aleph_alpha_client/qa.py
+++ b/aleph_alpha_client/qa.py
@@ -1,10 +1,12 @@
-from typing import Any, Dict, Mapping, NamedTuple, Optional, Sequence
+from dataclasses import asdict, dataclass
+from typing import Any, Dict, Mapping, Optional, Sequence
 
 from aleph_alpha_client.document import Document
 
 
-class QaRequest(NamedTuple):
-    """DEPRECATED: `QaRequest` is deprecated and will be removed in the next major release. New
+@dataclass(frozen=True)
+class QaRequest:
+    """DEPRECATED: `QaRequest` is deprecated and will be removed in the future. New
     methods of processing Q&A tasks will be provided before this is removed.
 
     Answers a question about a prompt.
@@ -30,18 +32,24 @@ class QaRequest(NamedTuple):
     documents: Sequence[Document]
     max_answers: Optional[int] = None
 
-    def to_json(self) -> Dict[str, Any]:
-        payload = self._asdict()
-        payload["documents"] = [
-            document._to_serializable_document() for document in self.documents
-        ]
+    def to_json(self) -> Mapping[str, Any]:
+        payload = {
+            **self._asdict(),
+            "documents": [
+                document._to_serializable_document() for document in self.documents
+            ],
+        }
         if self.max_answers is None:
             del payload["max_answers"]
         return payload
 
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
+
 
-class QaAnswer(NamedTuple):
-    """DEPRECATED: `QaAnswer` is deprecated and will be removed in the next major release. New
+@dataclass(frozen=True)
+class QaAnswer:
+    """DEPRECATED: `QaAnswer` is deprecated and will be removed in the future. New
     methods of processing Q&A tasks will be provided before this is removed.
     """
 
@@ -50,8 +58,9 @@ class QaAnswer(NamedTuple):
     evidence: str
 
 
-class QaResponse(NamedTuple):
-    """DEPRECATED: `QaResponse` is deprecated and will be removed in the next major release. New
+@dataclass(frozen=True)
+class QaResponse:
+    """DEPRECATED: `QaResponse` is deprecated and will be removed in the future. New
     methods of processing Q&A tasks will be provided before this is removed.
     """
 
diff --git a/aleph_alpha_client/summarization.py b/aleph_alpha_client/summarization.py
index a86a12b..84809dc 100644
--- a/aleph_alpha_client/summarization.py
+++ b/aleph_alpha_client/summarization.py
@@ -1,10 +1,12 @@
-from typing import Any, Dict, Mapping, NamedTuple, Sequence
+from dataclasses import asdict, dataclass
+from typing import Any, Dict, Mapping, Sequence
 
 from aleph_alpha_client.document import Document
 
 
-class SummarizationRequest(NamedTuple):
-    """DEPRECATED: `SummarizationRequest` is deprecated and will be removed in the next major release. New
+@dataclass(frozen=True)
+class SummarizationRequest:
+    """DEPRECATED: `SummarizationRequest` is deprecated and will be removed in the future. New
     methods of processing Summarization tasks will be provided before this is removed.
 
     Summarizes a document.
@@ -35,14 +37,16 @@ class SummarizationRequest(NamedTuple):
     document: Document
     disable_optimizations: bool = False
 
-    def to_json(self) -> Dict[str, Any]:
-        payload = self._asdict()
-        payload["document"] = self.document._to_serializable_document()
-        return payload
+    def to_json(self) -> Mapping[str, Any]:
+        return {**self._asdict(), "document": self.document._to_serializable_document()}
 
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
 
-class SummarizationResponse(NamedTuple):
-    """DEPRECATED: `SummarizationResponse` is deprecated and will be removed in the next major release. New
+
+@dataclass(frozen=True)
+class SummarizationResponse:
+    """DEPRECATED: `SummarizationResponse` is deprecated and will be removed in the future. New
     methods of processing Summarization tasks will be provided before this is removed.
     """
 
diff --git a/aleph_alpha_client/tokenization.py b/aleph_alpha_client/tokenization.py
index 0ea084a..48944ff 100644
--- a/aleph_alpha_client/tokenization.py
+++ b/aleph_alpha_client/tokenization.py
@@ -1,7 +1,9 @@
-from typing import Any, Dict, NamedTuple, Optional, Sequence
+from dataclasses import asdict, dataclass
+from typing import Any, Dict, Mapping, Optional, Sequence
 
 
-class TokenizationRequest(NamedTuple):
+@dataclass(frozen=True)
+class TokenizationRequest:
     """Describes a tokenization request.
 
     Parameters
@@ -25,15 +27,20 @@ class TokenizationRequest(NamedTuple):
     tokens: bool
     token_ids: bool
 
-    def to_json(self) -> Dict[str, Any]:
-        payload = self._asdict()
-        return payload
+    def to_json(self) -> Mapping[str, Any]:
+        return self._asdict()
 
+    def _asdict(self) -> Mapping[str, Any]:
+        return asdict(self)
 
-class TokenizationResponse(NamedTuple):
+
+@dataclass(frozen=True)
+class TokenizationResponse:
     tokens: Optional[Sequence[str]] = None
     token_ids: Optional[Sequence[int]] = None
 
     @staticmethod
     def from_json(json: Dict[str, Any]) -> "TokenizationResponse":
-        return TokenizationResponse(**json)
+        return TokenizationResponse(
+            tokens=json.get("tokens"), token_ids=json.get("token_ids")
+        )
diff --git a/aleph_alpha_client/version.py b/aleph_alpha_client/version.py
index 0c11bab..ce1305b 100644
--- a/aleph_alpha_client/version.py
+++ b/aleph_alpha_client/version.py
@@ -1 +1 @@
-__version__ = "3.5.1"
+__version__ = "4.0.0"
diff --git a/tests/test_embed.py b/tests/test_embed.py
index 61ef650..fdd7750 100644
--- a/tests/test_embed.py
+++ b/tests/test_embed.py
@@ -115,13 +115,15 @@ async def test_modelname_gets_passed_along_for_async_client(httpserver: HTTPServ
         representation=SemanticRepresentation.Symmetric,
     )
     model_name = "test_model"
-    expected_body = request.to_json()
-    expected_body["model"] = model_name
+    expected_body = {
+        **request.to_json(),
+        "model": model_name,
+    }
     httpserver.expect_ordered_request(
         "/batch_semantic_embed", method="POST", data=json.dumps(expected_body)
     ).respond_with_json({"model_version": "1", "embeddings": []})
     async_client = AsyncClient(token="", host=httpserver.url_for(""), total_retries=1)
-    _resp = await async_client.batch_semantic_embed(request, model=model_name)
+    await async_client.batch_semantic_embed(request, model=model_name)
 
 
 # Client
@@ -212,10 +214,9 @@ def test_modelname_gets_passed_along_for_sync_client(httpserver: HTTPServer):
         representation=SemanticRepresentation.Symmetric,
     )
     model_name = "test_model"
-    expected_body = request.to_json()
-    expected_body["model"] = model_name
+    expected_body = {**request.to_json(), "model": model_name}
     httpserver.expect_ordered_request(
         "/batch_semantic_embed", method="POST", data=json.dumps(expected_body)
     ).respond_with_json({"model_version": "1", "embeddings": []})
     sync_client = Client(token="", host=httpserver.url_for(""), total_retries=1)
-    _resp = sync_client.batch_semantic_embed(request, model=model_name)
+    sync_client.batch_semantic_embed(request, model=model_name)