gllm-inference-binary 0.5.20__cp313-cp313-manylinux_2_31_x86_64.whl → 0.5.22__cp313-cp313-manylinux_2_31_x86_64.whl

gllm_inference/lm_invoker/azure_openai_lm_invoker.pyi

@@ -160,17 +160,17 @@ class AzureOpenAILMInvoker(OpenAILMInvoker):
         ```
 
     Reasoning:
-        Azure OpenAI\'s o-series models are classified as reasoning models. Reasoning models think before they answer,
-        producing a long internal chain of thought before responding to the user. Reasoning models excel in
-        complex problem solving, coding, scientific reasoning, and multi-step planning for agentic workflows.
+        Azure OpenAI\'s GPT-5 models and o-series models are classified as reasoning models. Reasoning models think
+        before they answer, producing a long internal chain of thought before responding to the user. Reasoning models
+        excel in complex problem solving, coding, scientific reasoning, and multi-step planning for agentic workflows.
 
         The reasoning effort of reasoning models can be set via the `reasoning_effort` parameter. This parameter
-        will guide the models on how many reasoning tokens it should generate before creating a response to the prompt.
+        will guide the models on how many reasoning tokens it should generate before creating a response.
         Available options include:
-        1. "low": Favors speed and economical token usage.
-        2. "medium": Favors a balance between speed and reasoning accuracy.
-        3. "high": Favors more complete reasoning at the cost of more tokens generated and slower responses.
-        When not set, the reasoning effort will be equivalent to `medium` by default.
+        1. "minimal": Favors the least amount of reasoning, only supported for GPT-5 models onwards.
+        2. "low": Favors speed and economical token usage.
+        3. "medium": Favors a balance between speed and reasoning accuracy.
+        4. "high": Favors more complete reasoning at the cost of more tokens generated and slower responses.
 
         Azure OpenAI doesn\'t expose the raw reasoning tokens. However, the summary of the reasoning tokens can still be
         generated. The summary level can be set via the `reasoning_summary` parameter. Available options include:

gllm_inference/lm_invoker/openai_lm_invoker.pyi

@@ -159,16 +159,17 @@ class OpenAILMInvoker(BaseLMInvoker):
         ```
 
     Reasoning:
-        OpenAI\'s o-series models are classified as reasoning models. Reasoning models think before they answer,
-        producing a long internal chain of thought before responding to the user. Reasoning models excel in
-        complex problem solving, coding, scientific reasoning, and multi-step planning for agentic workflows.
+        OpenAI\'s GPT-5 models and o-series models are classified as reasoning models. Reasoning models think before
+        they answer, producing a long internal chain of thought before responding to the user. Reasoning models
+        excel in complex problem solving, coding, scientific reasoning, and multi-step planning for agentic workflows.
 
         The reasoning effort of reasoning models can be set via the `reasoning_effort` parameter. This parameter
         will guide the models on how many reasoning tokens it should generate before creating a response.
         Available options include:
-        1. "low": Favors speed and economical token usage.
-        2. "medium": Favors a balance between speed and reasoning accuracy.
-        3. "high": Favors more complete reasoning at the cost of more tokens generated and slower responses.
+        1. "minimal": Favors the least amount of reasoning, only supported for GPT-5 models onwards.
+        2. "low": Favors speed and economical token usage.
+        3. "medium": Favors a balance between speed and reasoning accuracy.
+        4. "high": Favors more complete reasoning at the cost of more tokens generated and slower responses.
         When not set, the reasoning effort will be equivalent to `medium` by default.
 
         OpenAI doesn\'t expose the raw reasoning tokens. However, the summary of the reasoning tokens can still be

gllm_inference/lm_invoker/schema/openai.pyi

@@ -83,6 +83,7 @@ class ReasoningEffort(StrEnum):
     HIGH = 'high'
     MEDIUM = 'medium'
     LOW = 'low'
+    MINIMAL = 'minimal'
 
 class ReasoningSummary(StrEnum):
     """Defines the reasoning summary for reasoning models."""

{gllm_inference_binary-0.5.20.dist-info → gllm_inference_binary-0.5.22.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gllm-inference-binary
-Version: 0.5.20
+Version: 0.5.22
 Summary: A library containing components related to model inferences in Gen AI applications.
 Author: Henry Wicaksono
 Author-email: henry.wicaksono@gdplabs.id
@@ -23,7 +23,7 @@ Requires-Dist: aiohttp (>=3.12.14,<4.0.0)
 Requires-Dist: anthropic (>=0.60.0,<0.61.0) ; extra == "anthropic"
 Requires-Dist: filetype (>=1.2.0,<2.0.0)
 Requires-Dist: gllm-core-binary (>=0.3.0,<0.4.0)
-Requires-Dist: google-genai (==1.20.0) ; extra == "google"
+Requires-Dist: google-genai (>=1.26.0,<2.0.0) ; extra == "google"
 Requires-Dist: httpx (>=0.28.0,<0.29.0)
 Requires-Dist: huggingface-hub (>=0.30.0,<0.31.0) ; extra == "huggingface"
 Requires-Dist: jinja2 (>=3.1.4,<4.0.0)

{gllm_inference_binary-0.5.20.dist-info → gllm_inference_binary-0.5.22.dist-info}/RECORD

@@ -34,7 +34,7 @@ gllm_inference/exceptions/error_parser.pyi,sha256=4RkVfS2Fl9kjz_h2bK9eoAeI-Y-VkH
 gllm_inference/exceptions/exceptions.pyi,sha256=5YRackwVNvyOJjOtiVszqu8q87s8ioXTa-XwaYmeiC4,4643
 gllm_inference/lm_invoker/__init__.pyi,sha256=NmQSqObPjevEP1KbbrNnaz4GMh175EVPERZ19vK5Emc,1202
 gllm_inference/lm_invoker/anthropic_lm_invoker.pyi,sha256=MsF3OmDo0L9aEHuTJYTgsoDILi2B_IgKtPpDcDMduWc,14925
-gllm_inference/lm_invoker/azure_openai_lm_invoker.pyi,sha256=AoEC8GdPW2LAyiCfH7CoSGHRTlVUcteVx0WTIwkBljI,14527
+gllm_inference/lm_invoker/azure_openai_lm_invoker.pyi,sha256=94SgOYmMW-hI3TeGRMslplC5xZPxKs3M7CMcAuAZyO0,14545
 gllm_inference/lm_invoker/bedrock_lm_invoker.pyi,sha256=IuLxgCThOSBHx1AXqnhL6yVu5_JV6hAeGBWWm5P1JCo,12423
 gllm_inference/lm_invoker/datasaur_lm_invoker.pyi,sha256=QS84w3WpD3Oyl5HdxrucsadCmsHE8gn6Ewl3l01DCgI,9203
 gllm_inference/lm_invoker/google_lm_invoker.pyi,sha256=aPOlaw2rexUDhbMFaXnuKqOT7lqeKxjfeToe9LjwEUw,16787
@@ -42,14 +42,14 @@ gllm_inference/lm_invoker/langchain_lm_invoker.pyi,sha256=skcbX34ZosGMFli3SLmGIT
 gllm_inference/lm_invoker/litellm_lm_invoker.pyi,sha256=_c56ewpEQ-Ywj5ofFzRYBvQgefR7Q_WkcQt97lnIFgg,13128
 gllm_inference/lm_invoker/lm_invoker.pyi,sha256=uKKNom4kjH7xUnOsqh9DBRumo244U-f6EjR5kLncEys,7823
 gllm_inference/lm_invoker/openai_compatible_lm_invoker.pyi,sha256=_hOAde_Faph3JoGYh7zLch6BRc2Lam8PXZvi5-PkL-E,14938
-gllm_inference/lm_invoker/openai_lm_invoker.pyi,sha256=y9IiSvffDVA7cQh24nFZa2qbl-YOBT8A0rBWUcH8I6c,19531
+gllm_inference/lm_invoker/openai_lm_invoker.pyi,sha256=wPTJr5DkXpoXpxw3MoaqEnzAOUanBRGUu954KdKDaVU,19649
 gllm_inference/lm_invoker/schema/__init__.pyi,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 gllm_inference/lm_invoker/schema/anthropic.pyi,sha256=zNbm4RV454dBEEUUN-Vyl_4cO628wUMPZyrO27O_DfM,991
 gllm_inference/lm_invoker/schema/bedrock.pyi,sha256=FOUMZkBi6KRa__mYoy3FNJ5sP0EC6rgLuhoijnwelIg,927
 gllm_inference/lm_invoker/schema/datasaur.pyi,sha256=aA4DhTXIezwLvFzphR24a5ueVln2FCBIloP9Hbt3iz4,230
 gllm_inference/lm_invoker/schema/google.pyi,sha256=AIsNgq0ZZuicHmx4bL7z6q-946T05nWts3HUeA8hhHQ,505
 gllm_inference/lm_invoker/schema/langchain.pyi,sha256=l2kHU7S3vmG3-NCt8B26krp_i4Br3waES_CekkgrKSA,409
-gllm_inference/lm_invoker/schema/openai.pyi,sha256=w_XSzOohX_7vnR81hVXdgDbBqk737LmMAN_AvODGk2I,1902
+gllm_inference/lm_invoker/schema/openai.pyi,sha256=9KjOJMnDyPs4hsysD8qFEMObUkbnxp6U9PmRIiUa3h4,1926
 gllm_inference/lm_invoker/schema/openai_compatible.pyi,sha256=m3bL2hVpxI_crURIi1bGDUqMy1Z5OgKBVU_-BkhX1mg,1166
 gllm_inference/lm_invoker/schema/xai.pyi,sha256=cWnbJmDtllqRH3NXpQbiXgkNBcUXr8ksDSDywcgJebE,632
 gllm_inference/lm_invoker/xai_lm_invoker.pyi,sha256=wPzjSLLiPb4DB1gJjtCs6URPUN0JCDcIxeNRsmD6tbo,15420
@@ -96,8 +96,8 @@ gllm_inference/utils/io_utils.pyi,sha256=7kUTacHAVRYoemFUOjCH7-Qmw-YsQGd6rGYxjf_
 gllm_inference/utils/langchain.pyi,sha256=VluQiHkGigDdqLUbhB6vnXiISCP5hHqV0qokYY6dC1A,1164
 gllm_inference/utils/validation.pyi,sha256=toxBtRp-VItC_X7sNi-GDd7sjibBdWMrR0q01OI2D7k,385
 gllm_inference.build/.gitignore,sha256=aEiIwOuxfzdCmLZe4oB1JsBmCUxwG8x-u-HBCV9JT8E,1
-gllm_inference.cpython-313-x86_64-linux-gnu.so,sha256=jwwiGpOGCRVrW31cK29huqEf_dxu0n9CAPNl3EMKsmM,4367528
+gllm_inference.cpython-313-x86_64-linux-gnu.so,sha256=hm1eT48i51NCfmmEu05NArSwwkFBUu516r2GAW_O5lY,4367528
 gllm_inference.pyi,sha256=uxl1voKdn19LurAHKEZLWbq9ryPO4UkJ1Nk1MM8IL34,3636
-gllm_inference_binary-0.5.20.dist-info/METADATA,sha256=Uq49Bd6PIOb23ezYyti27LutHIe-fqCtcpC9rbv2PNE,4608
-gllm_inference_binary-0.5.20.dist-info/WHEEL,sha256=qGYSeeDMRvGsNMRKS15OK05VQRV6Z0DMQkqDjYiypg0,110
-gllm_inference_binary-0.5.20.dist-info/RECORD,,
+gllm_inference_binary-0.5.22.dist-info/METADATA,sha256=cYrD49wnleGgC4XFM7sECpvEUXxaoewJ9DOBqCCQdwc,4615
+gllm_inference_binary-0.5.22.dist-info/WHEEL,sha256=qGYSeeDMRvGsNMRKS15OK05VQRV6Z0DMQkqDjYiypg0,110
+gllm_inference_binary-0.5.22.dist-info/RECORD,,