feat(llm): added up-to-date models for gemini, claude, openai

Author: daltonnyx

AgentCrew/__init__.py

@@ -1 +1 @@
-__version__ = "0.9.4"
+__version__ = "0.9.5"

AgentCrew/modules/llm/constants.py

@@ -44,6 +44,17 @@
         input_token_price_1m=3.0,
         output_token_price_1m=15.0,
     ),
+    Model(
+        id="claude-sonnet-4-6",
+        provider="claude",
+        name="Claude 4.6 Sonnet",
+        description="Our smartest model for complex agents and coding",
+        capabilities=["thinking", "tool_use", "vision", "stream"],
+        max_context_token=200_000,
+        default=False,
+        input_token_price_1m=3.0,
+        output_token_price_1m=15.0,
+    ),
     Model(
         id="claude-haiku-4-5",
         provider="claude",
@@ -281,6 +292,17 @@
         input_token_price_1m=2,
         output_token_price_1m=12,
     ),
+    Model(
+        id="gemini-3.1-pro-preview",
+        provider="google",
+        name="Gemini 3.1 Pro",
+        max_context_token=1_000_000,
+        description="Google's most intelligent model family to date, built on a foundation of state-of-the-art reasoning",
+        capabilities=["tool_use", "thinking", "vision", "structured_output"],
+        force_sample_params=SampleParam(temperature=1.0),
+        input_token_price_1m=2,
+        output_token_price_1m=12,
+    ),
 ]
 
 _DEEPINFRA_MODELS = [
@@ -327,14 +349,13 @@
         output_token_price_1m=1.1,
     ),
     Model(
-        id="Qwen/Qwen3-235B-A22B-Thinking-2507",
+        id="Qwen/Qwen3-Max-Thinking",
         provider="deepinfra",
-        name="Qwen 3 MoE 235B-22B",
+        name="Qwen 3 Max Thinking",
         description="Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models",
         capabilities=["tool_use", "thinking", "stream", "structured_output"],
-        force_sample_params=SampleParam(
-            temperature=0.6, top_p=0.95, top_k=20, min_p=0.0
-        ),
+        force_sample_params=SampleParam(temperature=0.7, top_p=0.9, top_k=0, min_p=0.0),
+        max_context_token=256_000,
         input_token_price_1m=0.2,
         output_token_price_1m=0.6,
     ),
@@ -421,15 +442,15 @@
         output_token_price_1m=2.0,
     ),
     Model(
-        id="MiniMaxAI/MiniMax-M2",
+        id="MiniMaxAI/MiniMax-M2.5",
         provider="deepinfra",
-        name="MiniMax-M2",
-        description="MiniMax-M2 is a Mini model built for Max coding & agentic workflows with just 10 billion activated parameters",
+        name="MiniMax-M2.5",
+        description="MiniMax-M2.5 is a Mini model built for Max coding & agentic workflows with just 10 billion activated parameters",
         capabilities=["tool_use", "stream", "structured_output"],
         force_sample_params=SampleParam(temperature=0.6),
         max_context_token=262_000,
-        input_token_price_1m=0.254,
-        output_token_price_1m=1.02,
+        input_token_price_1m=0.27,
+        output_token_price_1m=0.95,
     ),
 ]
 _GITHUB_COPILOT_MODELS = [
@@ -465,6 +486,17 @@
         force_sample_params=SampleParam(temperature=1.0),
         output_token_price_1m=0.0,
     ),
+    Model(
+        id="gemini-3.1-pro-preview",
+        provider="github_copilot",
+        name="Gemini 3.1 Pro",
+        description="",
+        capabilities=["tool_use", "vision", "stream"],
+        default=False,
+        input_token_price_1m=0.0,
+        force_sample_params=SampleParam(temperature=1.0),
+        output_token_price_1m=0.0,
+    ),
     Model(
         id="gemini-3-flash-preview",
         provider="github_copilot",
@@ -574,6 +606,18 @@
         input_token_price_1m=0.0,
         output_token_price_1m=0.0,
     ),
+    Model(
+        id="gpt-5.3-codex",
+        provider="copilot_response",
+        name="GPT 5 Codex",
+        description="",
+        capabilities=["tool_use", "vision", "stream", "structured_output"],
+        max_context_token=400_000,
+        default=False,
+        input_token_price_1m=0.0,
+        output_token_price_1m=0.0,
+        endpoint="response",
+    ),
     Model(
         id="gpt-5.1-codex",
         provider="copilot_response",
@@ -631,6 +675,17 @@
         input_token_price_1m=0.0,
         output_token_price_1m=0.0,
     ),
+    Model(
+        id="claude-sonnet-4.6",
+        provider="github_copilot",
+        name="Claude Sonnet 4.6",
+        description="",
+        capabilities=["tool_use", "vision", "stream", "thinking"],
+        default=False,
+        max_context_token=200_000,
+        input_token_price_1m=0.0,
+        output_token_price_1m=0.0,
+    ),
     Model(
         id="claude-opus-41",
         provider="github_copilot",

docker/pyproject.docker.toml

@@ -1,6 +1,6 @@
 [project]
 name = "agentcrew-ai"
-version = "0.9.3"
+version = "0.9.5"
 requires-python = ">=3.12"
 classifiers = [
   "Programming Language :: Python :: 3",

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "agentcrew-ai"
-version = "0.9.4"
+version = "0.9.5"
 requires-python = ">=3.12"
 classifiers = [
   "Programming Language :: Python :: 3",