Merge pull request #228 from MorpheusAIs/feature/update-pricing

update pricing

Author: alex-sandrk

models/prod_model_price.json

@@ -10,11 +10,18 @@
     "kimi-k2.5": { "input": "0.60", "output": "3.00" },
     "kimi-k2-thinking": { "input": "0.60", "output": "3.00" },
 
+    "gemma-4-31b": { "input": "0.15", "output": "0.40" },
+    "gemma-4-26b-a4b": { "input": "0.15", "output": "0.40" },
+
     "qwen3-235b": { "input": "0.40", "output": "3.00" },
     "qwen-3-235b": { "input": "0.40", "output": "3.00" },
     "qwen3-coder-480b-a35b-instruct": { "input": "0.70", "output": "2.80" },
     "qwen3-coder-480b-a35b": { "input": "0.70", "output": "2.80" },
     "qwen3-next-80b": { "input": "0.15", "output": "1.50" },
+    "qwen35-35b-a3b": { "input": "0.30", "output": "1.25" },
+    "qwen35-9b": { "input": "0.05", "output": "0.15" },
+
+    "arcee-trinity-large-thinking": { "input": "0.30", "output": "1.00" },
 
     "minimax-m2.5": { "input": "0.30", "output": "1.20" },
 

models/prod_rate_limit.json

@@ -17,9 +17,7 @@
       "tpm": 1000000,
       "models": [
         "llama-3.2-3b",
-        "llama-3.2-3b:web",
-        "qwen3-4b",
-        "qwen3-4b:web"
+        "llama-3.2-3b:web"
       ],
       "priority": 25,
       "description": "Small models with high throughput limits"
@@ -35,6 +33,10 @@
         "mistral-31-24b:web",
         "qwen3-next-80b",
         "qwen3-next-80b:web",
+        "qwen35-35b-a3b",
+        "qwen35-35b-a3b:web",
+        "qwen35-9b",
+        "qwen35-9b:web",
         "venice-uncensored",
         "venice-uncensored:web"
       ],
@@ -58,6 +60,12 @@
         "kimi-k2.5:web",
         "kimi-k2-thinking",
         "kimi-k2-thinking:web",
+        "gemma-4-31b",
+        "gemma-4-31b:web",
+        "gemma-4-26b-a4b",
+        "gemma-4-26b-a4b:web",
+        "arcee-trinity-large-thinking",
+        "arcee-trinity-large-thinking:web",
         "qwen3-235b",
         "qwen3-235b:web",
         "qwen3-coder-480b-a35b-instruct",