lm-deluge 0.0.56__tar.gz → 0.0.58__tar.gz

{lm_deluge-0.0.56/src/lm_deluge.egg-info → lm_deluge-0.0.58}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: lm_deluge
-Version: 0.0.56
+Version: 0.0.58
 Summary: Python utility for using LLM API models.
 Author-email: Benjamin Anderson <ben@trytaylor.ai>
 Requires-Python: >=3.10

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/pyproject.toml

@@ -3,7 +3,7 @@ requires = ["setuptools", "wheel"]
 
 [project]
 name = "lm_deluge"
-version = "0.0.56"
+version = "0.0.58"
 authors = [{ name = "Benjamin Anderson", email = "ben@trytaylor.ai" }]
 description = "Python utility for using LLM API models."
 readme = "README.md"

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/src/lm_deluge/__init__.py

@@ -1,7 +1,7 @@
 from .client import APIResponse, LLMClient, SamplingParams
 from .file import File
 from .prompt import Conversation, Message
-from .tool import Tool
+from .tool import Tool, ToolParams
 
 # dotenv.load_dotenv() - don't do this, fucks with other packages
 
@@ -12,5 +12,6 @@ __all__ = [
     "Conversation",
     "Message",
     "Tool",
+    "ToolParams",
     "File",
 ]

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/src/lm_deluge/api_requests/anthropic.py

@@ -72,7 +72,7 @@ def _build_anthropic_request(
         request_json["system"] = system_message
 
     # handle temp + top_p for opus 4.1/sonnet 4.5
-    if model.name in ["claude-sonnet-4-5-20250929", "claude-opus-4-1-20250805"]:
+    if "4-1" in model.name or "4-5" in model.name:
         if "temperature" in request_json and "top_p" in request_json:
             request_json.pop("top_p")
 

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/src/lm_deluge/client.py

@@ -369,7 +369,7 @@ class _LLMClient(BaseModel):
         tools: list[Tool | dict | MCPServer] | None = ...,
         cache: CachePattern | None = ...,
         use_responses_api: bool = ...,
-    ) -> list[APIResponse | None]: ...
+    ) -> list[APIResponse]: ...
 
     async def process_prompts_async(
         self,
@@ -380,7 +380,7 @@ class _LLMClient(BaseModel):
         tools: list[Tool | dict | MCPServer] | None = None,
         cache: CachePattern | None = None,
         use_responses_api: bool = False,
-    ) -> list[APIResponse | None] | list[str | None] | dict[str, int]:
+    ) -> list[APIResponse] | list[str | None] | dict[str, int]:
         """Process multiple prompts asynchronously using the start_nowait/wait_for_all backend.
 
         This implementation creates all tasks upfront and waits for them to complete,
@@ -516,28 +516,40 @@ class _LLMClient(BaseModel):
         tools: list[Tool | dict | MCPServer] | None = None,
         cache: CachePattern | None = None,
         use_responses_api: bool = False,
-    ) -> APIResponse | None:
+    ) -> APIResponse:
         task_id = self.start_nowait(
             prompt, tools=tools, cache=cache, use_responses_api=use_responses_api
         )
         return await self.wait_for(task_id)
 
-    async def wait_for(self, task_id: int) -> APIResponse | None:
+    async def wait_for(self, task_id: int) -> APIResponse:
         task = self._tasks.get(task_id)
         if task:
             return await task
-        return self._results.get(task_id)
+        res = self._results.get(task_id)
+        if res:
+            return res
+        else:
+            return APIResponse(
+                id=-1,
+                model_internal="",
+                prompt=Conversation([]),
+                sampling_params=SamplingParams(),
+                status_code=500,
+                is_error=True,
+                error_message="Task not found",
+            )
 
     async def wait_for_all(
         self, task_ids: Sequence[int] | None = None
-    ) -> list[APIResponse | None]:
+    ) -> list[APIResponse]:
         if task_ids is None:
             task_ids = list(self._tasks.keys())
         return [await self.wait_for(tid) for tid in task_ids]
 
     async def as_completed(
         self, task_ids: Sequence[int] | None = None
-    ) -> AsyncGenerator[tuple[int, APIResponse | None], None]:
+    ) -> AsyncGenerator[tuple[int, APIResponse], None]:
         """Yield ``(task_id, result)`` pairs as tasks complete.
 
         Args:
@@ -561,7 +573,9 @@ class _LLMClient(BaseModel):
         for task in list(tasks_map.keys()):
             if task.done():
                 tid = tasks_map.pop(task)
-                yield tid, self._results.get(tid, await task)
+                task_result = self._results.get(tid, await task)
+                assert task_result
+                yield tid, task_result
 
         while tasks_map:
             done, _ = await asyncio.wait(
@@ -569,7 +583,9 @@ class _LLMClient(BaseModel):
             )
             for task in done:
                 tid = tasks_map.pop(task)
-                yield tid, self._results.get(tid, await task)
+                task_result = self._results.get(tid, await task)
+                assert task_result
+                yield tid, task_result
 
     async def stream(
         self,
@@ -618,23 +634,20 @@ class _LLMClient(BaseModel):
                     mcp_tools = await tool.to_tools()
                     expanded_tools.extend(mcp_tools)
 
-        last_response: APIResponse | None = None
+        response: APIResponse | None = None
 
         for _ in range(max_rounds):
-            responses = await self.process_prompts_async(
-                [conversation],
+            response = await self.start(
+                conversation,
                 tools=tools,  # type: ignore
-                return_completions_only=False,
-                show_progress=show_progress,
             )
 
-            last_response = responses[0]
-            if last_response is None or last_response.content is None:
+            if response is None or response.content is None:
                 break
 
-            conversation = conversation.with_message(last_response.content)
+            conversation = conversation.with_message(response.content)
 
-            tool_calls = last_response.content.tool_calls
+            tool_calls = response.content.tool_calls
             if not tool_calls:
                 break
 
@@ -657,12 +670,12 @@ class _LLMClient(BaseModel):
                 if not isinstance(result, (str, dict, list)):
                     result = str(result)
 
-                conversation.add_tool_result(call.id, result)  # type: ignore
+                conversation.with_tool_result(call.id, result)  # type: ignore
 
-        if last_response is None:
+        if response is None:
             raise RuntimeError("model did not return a response")
 
-        return conversation, last_response
+        return conversation, response
 
     def run_agent_loop_sync(
         self,

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/src/lm_deluge/models/__init__.py

@@ -38,9 +38,9 @@ class APIModel:
     supports_responses: bool = False
     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
-    gpus: list[str] | None = None
+    # tokens_per_minute: int | None = None
+    # requests_per_minute: int | None = None
+    # gpus: list[str] | None = None
 
     @classmethod
     def from_registry(cls, name: str):
@@ -97,8 +97,8 @@ def register_model(
     supports_responses: bool = False,
     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,
+    # tokens_per_minute: int | None = None,
+    # requests_per_minute: int | None = None,
 ) -> APIModel:
     """Register a model configuration and return the created APIModel."""
     model = APIModel(
@@ -116,8 +116,8 @@ def register_model(
         supports_responses=supports_responses,
         reasoning_model=reasoning_model,
         regions=regions,
-        tokens_per_minute=tokens_per_minute,
-        requests_per_minute=requests_per_minute,
+        # tokens_per_minute=tokens_per_minute,
+        # requests_per_minute=requests_per_minute,
     )
     registry[model.id] = model
     return model

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/src/lm_deluge/models/anthropic.py

@@ -10,6 +10,18 @@ ANTHROPIC_MODELS = {
     #                                                                   ░███
     #                                                                   █████
     #
+    "claude-4.5-haiku": {
+        "id": "claude-4.5-haiku",
+        "name": "claude-haiku-4-5-20251001",
+        "api_base": "https://api.anthropic.com/v1",
+        "api_key_env_var": "ANTHROPIC_API_KEY",
+        "supports_json": False,
+        "api_spec": "anthropic",
+        "input_cost": 1.0,
+        "cached_input_cost": 0.10,
+        "cache_write_cost": 1.25,
+        "output_cost": 3.0,
+    },
     "claude-4.5-sonnet": {
         "id": "claude-4.5-sonnet",
         "name": "claude-sonnet-4-5-20250929",
@@ -21,8 +33,6 @@ ANTHROPIC_MODELS = {
         "cached_input_cost": 0.30,
         "cache_write_cost": 3.75,
         "output_cost": 15.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
     },
     "claude-4.1-opus": {
         "id": "claude-4.1-opus",
@@ -35,8 +45,6 @@ ANTHROPIC_MODELS = {
         "cached_input_cost": 1.50,
         "cache_write_cost": 18.75,
         "output_cost": 75.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
         "reasoning_model": True,
     },
     "claude-4-opus": {
@@ -50,8 +58,6 @@ ANTHROPIC_MODELS = {
         "cached_input_cost": 1.50,
         "cache_write_cost": 18.75,
         "output_cost": 75.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
         "reasoning_model": True,
     },
     "claude-4-sonnet": {
@@ -65,8 +71,6 @@ ANTHROPIC_MODELS = {
         "cached_input_cost": 0.30,
         "cache_write_cost": 3.75,
         "output_cost": 15.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
     },
     "claude-3.7-sonnet": {
         "id": "claude-3.7-sonnet",
@@ -79,8 +83,6 @@ ANTHROPIC_MODELS = {
         "cached_input_cost": 0.30,
         "cache_write_cost": 3.75,
         "output_cost": 15.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
         "reasoning_model": True,
     },
     "claude-3.6-sonnet": {
@@ -94,8 +96,6 @@ ANTHROPIC_MODELS = {
         "cached_input_cost": 0.30,
         "cache_write_cost": 3.75,
         "output_cost": 15.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
     },
     "claude-3.5-sonnet": {
         "id": "claude-3.5-sonnet",
@@ -108,8 +108,6 @@ ANTHROPIC_MODELS = {
         "cached_input_cost": 0.30,
         "cache_write_cost": 3.75,
         "output_cost": 15.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
     },
     "claude-3-opus": {
         "id": "claude-3-opus",
@@ -120,8 +118,6 @@ ANTHROPIC_MODELS = {
         "api_spec": "anthropic",
         "input_cost": 15.0,
         "output_cost": 75.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
     },
     "claude-3.5-haiku": {
         "id": "claude-3.5-haiku",
@@ -134,8 +130,6 @@ ANTHROPIC_MODELS = {
         "cached_input_cost": 0.08,
         "cache_write_cost": 1.00,
         "output_cost": 4.00,
-        "requests_per_minute": 20_000,
-        "tokens_per_minute": 4_000_000,  # supposed to be this but they fucked up
     },
     "claude-3-haiku": {
         "id": "claude-3-haiku",
@@ -148,7 +142,5 @@ ANTHROPIC_MODELS = {
         "cache_write_cost": 0.30,
         "cached_input_cost": 0.03,
         "output_cost": 1.25,
-        "requests_per_minute": 10_000,
-        "tokens_per_minute": 4_000_000,  # supposed to be this but they fucked up
     },
 }

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/src/lm_deluge/models/bedrock.py

@@ -16,8 +16,6 @@ BEDROCK_MODELS = {
         "api_spec": "bedrock",
         "input_cost": 0.25,
         "output_cost": 1.25,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 8_000_000,
     },
     "claude-3.5-haiku-bedrock": {
         "id": "claude-3.5-haiku-bedrock",
@@ -28,8 +26,6 @@ BEDROCK_MODELS = {
         "api_spec": "bedrock",
         "input_cost": 0.25,
         "output_cost": 1.25,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 8_000_000,
     },
     "claude-3.5-sonnet-bedrock": {
         "id": "claude-3.5-sonnet-bedrock",
@@ -40,8 +36,6 @@ BEDROCK_MODELS = {
         "api_spec": "bedrock",
         "input_cost": 3.0,
         "output_cost": 15.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
         "reasoning_model": False,
     },
     "claude-3.6-sonnet-bedrock": {
@@ -53,8 +47,6 @@ BEDROCK_MODELS = {
         "api_spec": "bedrock",
         "input_cost": 3.0,
         "output_cost": 15.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
         "reasoning_model": False,
     },
     "claude-3.7-sonnet-bedrock": {
@@ -66,8 +58,6 @@ BEDROCK_MODELS = {
         "api_spec": "bedrock",
         "input_cost": 3.0,
         "output_cost": 15.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
         "reasoning_model": True,
     },
     "claude-4-sonnet-bedrock": {
@@ -79,8 +69,6 @@ BEDROCK_MODELS = {
         "api_spec": "bedrock",
         "input_cost": 3.0,
         "output_cost": 15.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
         "reasoning_model": True,
     },
     "claude-4-opus-bedrock": {
@@ -92,8 +80,6 @@ BEDROCK_MODELS = {
         "api_spec": "bedrock",
         "input_cost": 3.0,
         "output_cost": 15.0,
-        "requests_per_minute": 4_000,
-        "tokens_per_minute": 400_000,
         "reasoning_model": True,
     },
     # GPT-OSS on AWS Bedrock

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/src/lm_deluge/models/cohere.py

@@ -15,8 +15,6 @@ COHERE_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.5,
         "output_cost": 1.5,
-        "requests_per_minute": 10_000,
-        "tokens_per_minute": None,
     },
     "aya-expanse-32b": {
         "id": "aya-expanse-32b",
@@ -26,8 +24,6 @@ COHERE_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.5,
         "output_cost": 1.5,
-        "requests_per_minute": 10_000,
-        "tokens_per_minute": None,
     },
     "aya-vision-8b": {
         "id": "aya-vision-8b",
@@ -37,8 +33,6 @@ COHERE_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.5,
         "output_cost": 1.5,
-        "requests_per_minute": 10_000,
-        "tokens_per_minute": None,
     },
     "aya-vision-32b": {
         "id": "aya-vision-32b",
@@ -48,8 +42,6 @@ COHERE_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.5,
         "output_cost": 1.5,
-        "requests_per_minute": 10_000,
-        "tokens_per_minute": None,
     },
     "command-a": {
         "id": "command-a",
@@ -59,8 +51,6 @@ COHERE_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.5,
         "output_cost": 1.5,
-        "requests_per_minute": 10_000,
-        "tokens_per_minute": None,
     },
     "command-r-7b": {
         "id": "command-r-cohere",
@@ -70,8 +60,6 @@ COHERE_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.5,
         "output_cost": 1.5,
-        "requests_per_minute": 10_000,
-        "tokens_per_minute": None,
     },
     "command-r": {
         "id": "command-r",
@@ -81,8 +69,6 @@ COHERE_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.5,
         "output_cost": 1.5,
-        "requests_per_minute": 10_000,
-        "tokens_per_minute": None,
     },
     "command-r-plus": {
         "id": "command-r-plus",
@@ -92,7 +78,5 @@ COHERE_MODELS = {
         "api_spec": "openai",
         "input_cost": 3.0,
         "output_cost": 15.0,
-        "requests_per_minute": 10_000,
-        "tokens_per_minute": None,
     },
 }

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/src/lm_deluge/models/google.py

@@ -20,8 +20,6 @@ GOOGLE_MODELS = {
         "input_cost": 0.1,
         "cached_input_cost": 0.025,
         "output_cost": 0.4,
-        "requests_per_minute": 20,
-        "tokens_per_minute": 100_000,
         "reasoning_model": False,
     },
     "gemini-2.0-flash-lite-compat": {
@@ -34,8 +32,6 @@ GOOGLE_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.075,
         "output_cost": 0.3,
-        "requests_per_minute": 20,
-        "tokens_per_minute": 100_000,
         "reasoning_model": False,
     },
     "gemini-2.5-pro-compat": {
@@ -49,8 +45,6 @@ GOOGLE_MODELS = {
         "input_cost": 1.25,
         "cached_input_cost": 0.31,
         "output_cost": 10.0,
-        "requests_per_minute": 20,
-        "tokens_per_minute": 100_000,
         "reasoning_model": True,
     },
     "gemini-2.5-flash-compat": {
@@ -64,8 +58,6 @@ GOOGLE_MODELS = {
         "input_cost": 0.3,
         "cached_input_cost": 0.075,
         "output_cost": 2.5,
-        "requests_per_minute": 20,
-        "tokens_per_minute": 100_000,
         "reasoning_model": True,
     },
     "gemini-2.5-flash-lite-compat": {
@@ -79,8 +71,6 @@ GOOGLE_MODELS = {
         "input_cost": 0.1,
         "cached_input_cost": 0.025,
         "output_cost": 0.4,
-        "requests_per_minute": 20,
-        "tokens_per_minute": 100_000,
         "reasoning_model": True,
     },
     # Native Gemini API versions with file support
@@ -95,8 +85,6 @@ GOOGLE_MODELS = {
         "input_cost": 0.1,
         "cached_input_cost": 0.025,
         "output_cost": 0.4,
-        "requests_per_minute": 20,
-        "tokens_per_minute": 100_000,
         "reasoning_model": False,
     },
     "gemini-2.0-flash-lite": {
@@ -109,8 +97,6 @@ GOOGLE_MODELS = {
         "api_spec": "gemini",
         "input_cost": 0.075,
         "output_cost": 0.3,
-        "requests_per_minute": 20,
-        "tokens_per_minute": 100_000,
         "reasoning_model": False,
     },
     "gemini-2.5-pro": {
@@ -124,8 +110,6 @@ GOOGLE_MODELS = {
         "input_cost": 1.25,
         "cached_input_cost": 0.31,
         "output_cost": 10.0,
-        "requests_per_minute": 20,
-        "tokens_per_minute": 100_000,
         "reasoning_model": True,
     },
     "gemini-2.5-flash": {
@@ -139,8 +123,6 @@ GOOGLE_MODELS = {
         "input_cost": 0.3,
         "cached_input_cost": 0.075,
         "output_cost": 2.5,
-        "requests_per_minute": 20,
-        "tokens_per_minute": 100_000,
         "reasoning_model": True,
     },
     "gemini-2.5-flash-lite": {
@@ -154,8 +136,6 @@ GOOGLE_MODELS = {
         "input_cost": 0.1,
         "cached_input_cost": 0.025,
         "output_cost": 0.4,
-        "requests_per_minute": 20,
-        "tokens_per_minute": 100_000,
         "reasoning_model": True,
     },
 }

lm_deluge-0.0.58/src/lm_deluge/models/grok.py

@@ -0,0 +1,82 @@
+XAI_MODELS = {
+    #  .d8888b.                  888
+    # d88P  Y88b                 888
+    # 888    888                 888
+    # 888        888d888 .d88b.  888  888
+    # 888  88888 888P"  d88""88b 888 .88P
+    # 888    888 888    888  888 888888K
+    # Y88b  d88P 888    Y88..88P 888 "88b
+    #  "Y8888P88 888     "Y88P"  888  888
+    "grok-code-fast-1": {
+        "id": "grok-code-fast-1",
+        "name": "grok-code-fast-1",
+        "api_base": "https://api.x.ai/v1",
+        "api_key_env_var": "GROK_API_KEY",
+        "supports_json": True,
+        "supports_logprobs": True,
+        "api_spec": "openai",
+        "input_cost": 0.2,
+        "output_cost": 1.5,
+        "reasoning_model": False,
+    },
+    "grok-4-fast-reasoning": {
+        "id": "grok-4-fast-reasoning",
+        "name": "grok-4-fast-reasoning",
+        "api_base": "https://api.x.ai/v1",
+        "api_key_env_var": "GROK_API_KEY",
+        "supports_json": True,
+        "supports_logprobs": True,
+        "api_spec": "openai",
+        "input_cost": 0.2,
+        "output_cost": 0.5,
+        "reasoning_model": False,
+    },
+    "grok-4-fast-non-reasoning": {
+        "id": "grok-4-fast-non-reasoning",
+        "name": "grok-4-fast-non-reasoning",
+        "api_base": "https://api.x.ai/v1",
+        "api_key_env_var": "GROK_API_KEY",
+        "supports_json": True,
+        "supports_logprobs": True,
+        "api_spec": "openai",
+        "input_cost": 0.2,
+        "output_cost": 0.5,
+        "reasoning_model": False,
+    },
+    "grok-4": {
+        "id": "grok-4",
+        "name": "grok-4-0709",
+        "api_base": "https://api.x.ai/v1",
+        "api_key_env_var": "GROK_API_KEY",
+        "supports_json": True,
+        "supports_logprobs": True,
+        "api_spec": "openai",
+        "input_cost": 2.0,
+        "output_cost": 8.0,
+        "reasoning_model": False,
+    },
+    "grok-3": {
+        "id": "grok-3",
+        "name": "grok-3-latest",
+        "api_base": "https://api.x.ai/v1",
+        "api_key_env_var": "GROK_API_KEY",
+        "supports_json": True,
+        "supports_logprobs": True,
+        "api_spec": "openai",
+        "input_cost": 2.0,
+        "output_cost": 8.0,
+        "reasoning_model": False,
+    },
+    "grok-3-mini": {
+        "id": "grok-3-mini",
+        "name": "grok-3-mini-latest",
+        "api_base": "https://api.x.ai/v1",
+        "api_key_env_var": "GROK_API_KEY",
+        "supports_json": True,
+        "supports_logprobs": True,
+        "api_spec": "openai",
+        "input_cost": 2.0,
+        "output_cost": 8.0,
+        "reasoning_model": True,
+    },
+}

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/src/lm_deluge/models/groq.py

@@ -41,10 +41,10 @@ GROQ_MODELS = {
     },
     "kimi-k2-groq": {
         "id": "kimi-k2-groq",
-        "name": "moonshotai/kimi-k2-instruct",
+        "name": "moonshotai/kimi-k2-instruct-0905",
         "api_base": "https://api.groq.com/openai/v1",
         "api_key_env_var": "GROQ_API_KEY",
-        "supports_json": False,
+        "supports_json": True,
         "api_spec": "openai",
     },
     "gpt-oss-120b-groq": {

{lm_deluge-0.0.56 → lm_deluge-0.0.58}/src/lm_deluge/models/meta.py

@@ -16,8 +16,6 @@ META_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.0,
         "output_cost": 0.0,
-        "requests_per_minute": 3_000,
-        "tokens_per_minute": 1_000_000,
         "reasoning_model": False,
     },
     "llama-4-maverick": {
@@ -30,8 +28,6 @@ META_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.0,
         "output_cost": 0.0,
-        "requests_per_minute": 3_000,
-        "tokens_per_minute": 1_000_000,
         "reasoning_model": False,
     },
     "llama-3.3-70b": {
@@ -44,8 +40,6 @@ META_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.0,
         "output_cost": 0.0,
-        "requests_per_minute": 3_000,
-        "tokens_per_minute": 1_000_000,
         "reasoning_model": False,
     },
     "llama-3.3-8b": {
@@ -58,8 +52,6 @@ META_MODELS = {
         "api_spec": "openai",
         "input_cost": 0.0,
         "output_cost": 0.0,
-        "requests_per_minute": 3_000,
-        "tokens_per_minute": 1_000_000,
         "reasoning_model": False,
     },
 }