lm-deluge 0.0.33__py3-none-any.whl → 0.0.34__py3-none-any.whl

lm_deluge/api_requests/gemini.py

@@ -1,7 +1,7 @@
 import json
 import os
 import warnings
-
+from typing import Any
 from aiohttp import ClientResponse
 
 from lm_deluge.request_context import RequestContext
@@ -37,11 +37,12 @@ async def _build_gemini_request(
 
     # Handle reasoning models (thinking)
     if model.reasoning_model:
-        thinking_config = None
+        thinking_config: dict[str, Any] | None = None
         effort = sampling_params.reasoning_effort
         if effort is None or effort == "none":
+            budget = 128 if "2.5-pro" in model.id else 0
             # Explicitly disable thoughts when no effort is requested
-            thinking_config = {"includeThoughts": False, "thinkingBudget": 0}
+            thinking_config = {"includeThoughts": False, "thinkingBudget": budget}
         else:
             thinking_config = {"includeThoughts": True}
             if effort in {"low", "medium", "high"} and "flash" in model.id:

lm_deluge/client.py

@@ -773,10 +773,54 @@ class _LLMClient(BaseModel):
 
 # Clean factory function with perfect IDE support
 @overload
-def LLMClient(model_names: str, **kwargs) -> _LLMClient: ...
+def LLMClient(
+    model_names: str,
+    *,
+    max_requests_per_minute: int = 1_000,
+    max_tokens_per_minute: int = 100_000,
+    max_concurrent_requests: int = 225,
+    sampling_params: list[SamplingParams] | None = None,
+    model_weights: list[float] | Literal["uniform", "dynamic"] = "uniform",
+    max_attempts: int = 5,
+    request_timeout: int = 30,
+    cache: Any = None,
+    extra_headers: dict[str, str] | None = None,
+    temperature: float = 0.75,
+    top_p: float = 1.0,
+    json_mode: bool = False,
+    max_new_tokens: int = 512,
+    reasoning_effort: Literal["low", "medium", "high", None] = None,
+    logprobs: bool = False,
+    top_logprobs: int | None = None,
+    force_local_mcp: bool = False,
+    progress: Literal["rich", "tqdm", "manual"] = "rich",
+) -> _LLMClient: ...
+
+
+@overload
+def LLMClient(
+    model_names: list[str],
+    *,
+    max_requests_per_minute: int = 1_000,
+    max_tokens_per_minute: int = 100_000,
+    max_concurrent_requests: int = 225,
+    sampling_params: list[SamplingParams] | None = None,
+    model_weights: list[float] | Literal["uniform", "dynamic"] = "uniform",
+    max_attempts: int = 5,
+    request_timeout: int = 30,
+    cache: Any = None,
+    extra_headers: dict[str, str] | None = None,
+    temperature: float = 0.75,
+    top_p: float = 1.0,
+    json_mode: bool = False,
+    max_new_tokens: int = 512,
+    reasoning_effort: Literal["low", "medium", "high", None] = None,
+    logprobs: bool = False,
+    top_logprobs: int | None = None,
+    force_local_mcp: bool = False,
+    progress: Literal["rich", "tqdm", "manual"] = "rich",
+) -> _LLMClient: ...
 
-@overload  
-def LLMClient(model_names: list[str], **kwargs) -> _LLMClient: ...
 
 def LLMClient(
     model_names: str | list[str] = "gpt-4.1-mini",
@@ -802,18 +846,18 @@ def LLMClient(
 ) -> _LLMClient:
     """
     Create an LLMClient with model_names as a positional argument.
-    
+
     Args:
         model_names: Model name(s) to use - can be a single string or list of strings
         **kwargs: All other LLMClient configuration options (keyword-only)
-    
+
     Returns:
         Configured LLMClient instance
     """
     # Handle default for mutable argument
     if sampling_params is None:
         sampling_params = []
-    
+
     # Simply pass everything to the Pydantic constructor
     return _LLMClient(
         model_names=model_names,

lm_deluge/models.py

@@ -1275,7 +1275,7 @@ def register_model(
     reasoning_model: bool = False,
     regions: list[str] | dict[str, int] = field(default_factory=list),
     tokens_per_minute: int | None = None,
-    requests_per_minute: int | None = None
+    requests_per_minute: int | None = None,
 ) -> APIModel:
     """Register a model configuration and return the created APIModel."""
     model = APIModel(
@@ -1292,7 +1292,7 @@ def register_model(
         reasoning_model=reasoning_model,
         regions=regions,
         tokens_per_minute=tokens_per_minute,
-        requests_per_minute=requests_per_minute
+        requests_per_minute=requests_per_minute,
     )
     registry[model.id] = model
     return model

{lm_deluge-0.0.33.dist-info → lm_deluge-0.0.34.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: lm_deluge
-Version: 0.0.33
+Version: 0.0.34
 Summary: Python utility for using LLM API models.
 Author-email: Benjamin Anderson <ben@trytaylor.ai>
 Requires-Python: >=3.10

{lm_deluge-0.0.33.dist-info → lm_deluge-0.0.34.dist-info}/RECORD

@@ -2,14 +2,14 @@ lm_deluge/__init__.py,sha256=mAztMuxINmh7dGbYnT8tsmw1eryQAvd0jpY8yHzd0EE,315
 lm_deluge/agent.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 lm_deluge/batches.py,sha256=vJXVnuuGkIQnXoDPODPERrvdG9X1Ov1jnXExnPe6ZAc,21772
 lm_deluge/cache.py,sha256=VB1kv8rM2t5XWPR60uhszFcxLDnVKOe1oA5hYjVDjIo,4375
-lm_deluge/client.py,sha256=82Q0nWmOhYgFQojlsxMXJyfTNmgpNn0QpDb_MjPj6-g,32833
+lm_deluge/client.py,sha256=mTC_gxydu1JBtjXcEp8_GuAj4U6cAvZzAQjj4_0gCt0,34287
 lm_deluge/config.py,sha256=H1tQyJDNHGFuwxqQNL5Z-CjWAC0luHSBA3iY_pxmACM,932
 lm_deluge/embed.py,sha256=CO-TOlC5kOTAM8lcnicoG4u4K664vCBwHF1vHa-nAGg,13382
 lm_deluge/errors.py,sha256=oHjt7YnxWbh-eXMScIzov4NvpJMo0-2r5J6Wh5DQ1tk,209
 lm_deluge/file.py,sha256=FGomcG8s2go_55Z2CChflHgmU-UqgFftgFY8c7f_G70,5631
 lm_deluge/gemini_limits.py,sha256=V9mpS9JtXYz7AY6OuKyQp5TuIMRH1BVv9YrSNmGmHNA,1569
 lm_deluge/image.py,sha256=Qpa0k5yXfrpSaHzVUwW_TEn7yEgmwzYGL17Sa7-KhSA,7729
-lm_deluge/models.py,sha256=WkbGoaWUJLRwB9ug9RBjKL-ykSOmf8oKEzcWh2dcCO4,51758
+lm_deluge/models.py,sha256=L1vL24I74QNL7AgAGSmUMNFW9gSMBc8xinDBcQXu158,51760
 lm_deluge/prompt.py,sha256=cfwzCAmT-1K0v7SfEMUrxpBkJGgf7IFlWfNLJrCcoBM,37025
 lm_deluge/request_context.py,sha256=o33LSEwnK6YPhZeulUoSE_VrdKCXiCQa0tjjixK2K6M,2540
 lm_deluge/rerank.py,sha256=-NBAJdHz9OB-SWWJnHzkFmeVO4wR6lFV7Vw-SxG7aVo,11457
@@ -21,7 +21,7 @@ lm_deluge/api_requests/anthropic.py,sha256=d22ainIrH2PgOgQZcygFZK-cvs6O4XCJCnyx2
 lm_deluge/api_requests/base.py,sha256=EVHNFtlttKbN7Tt1MnLaO-NjvKHPSV5CqlRv-OnpVAE,5593
 lm_deluge/api_requests/bedrock.py,sha256=FZMhF590JzJtAYDugbDtG93RhPt5efWZ0Wn4V8U8Dgw,11031
 lm_deluge/api_requests/common.py,sha256=BZ3vRO5TB669_UsNKugkkuFSzoLHOYJIKt4nV4sf4vc,422
-lm_deluge/api_requests/gemini.py,sha256=W4NjQ0buBsdS7RYpzDahrXNQWMzDHRMLNRSphCOmIqg,7685
+lm_deluge/api_requests/gemini.py,sha256=tXk6AfioN7xv7B_HYw7Va7kQsm0hLJhSZfYNP6hAwgM,7792
 lm_deluge/api_requests/mistral.py,sha256=S_LpOfCGbCVEROH_od3P-tYeNYTKFMamMTL-c_wFCBI,4597
 lm_deluge/api_requests/openai.py,sha256=hsJIMRO4wpalrczD0bVc--RWFu2BoXEp0USAwRlLQEA,21763
 lm_deluge/api_requests/response.py,sha256=FtkVYk_rDH93Kj9pqbB-l7a4dQHzVr6ivKL9khYKLbs,5966
@@ -48,8 +48,8 @@ lm_deluge/util/logprobs.py,sha256=UkBZakOxWluaLqHrjARu7xnJ0uCHVfLGHJdnYlEcutk,11
 lm_deluge/util/spatial.py,sha256=BsF_UKhE-x0xBirc-bV1xSKZRTUhsOBdGqsMKme20C8,4099
 lm_deluge/util/validation.py,sha256=hz5dDb3ebvZrZhnaWxOxbNSVMI6nmaOODBkk0htAUhs,1575
 lm_deluge/util/xml.py,sha256=Ft4zajoYBJR3HHCt2oHwGfymGLdvp_gegVmJ-Wqk4Ck,10547
-lm_deluge-0.0.33.dist-info/licenses/LICENSE,sha256=uNNXGXPCw2TC7CUs7SEBkA-Mz6QBQFWUUEWDMgEs1dU,1058
-lm_deluge-0.0.33.dist-info/METADATA,sha256=USqmVPTkUJkLSrMeChb-PyxELn6aMkBB-XK9UL8a2zs,13295
-lm_deluge-0.0.33.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-lm_deluge-0.0.33.dist-info/top_level.txt,sha256=hqU-TJX93yBwpgkDtYcXyLr3t7TLSCCZ_reytJjwBaE,10
-lm_deluge-0.0.33.dist-info/RECORD,,
+lm_deluge-0.0.34.dist-info/licenses/LICENSE,sha256=uNNXGXPCw2TC7CUs7SEBkA-Mz6QBQFWUUEWDMgEs1dU,1058
+lm_deluge-0.0.34.dist-info/METADATA,sha256=7vzU_xBUX93r35eUF08MmB0jhBN2SrYH4yhj7snPi2g,13295
+lm_deluge-0.0.34.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+lm_deluge-0.0.34.dist-info/top_level.txt,sha256=hqU-TJX93yBwpgkDtYcXyLr3t7TLSCCZ_reytJjwBaE,10
+lm_deluge-0.0.34.dist-info/RECORD,,