chore: change default models

Kujtim Hoxha created 3 months ago

Change summary

internal/providers/configs/cerebras.json | 51 ++++++++-----------------
1 file changed, 16 insertions(+), 35 deletions(-)

Detailed changes

internal/providers/configs/cerebras.json

@@ -4,14 +4,12 @@
     "type": "openai",
     "api_key": "$CEREBRAS_API_KEY",
     "api_endpoint": "https://api.cerebras.ai/v1",
-    "default_large_model_id": "qwen-3-235b-a22b-instruct-2507",
+    "default_large_model_id": "qwen-3-coder-480b",
     "default_small_model_id": "qwen-3-32b",
     "models": [
         {
             "id": "llama-4-scout-17b-16e-instruct",
             "name": "Llama 4 Scout",
-            "parameters": "109B",
-            "speed_tokens_per_s": 2600,
             "cost_per_1m_in": 0.65,
             "cost_per_1m_out": 0.85,
             "context_window": 32768,
@@ -22,10 +20,8 @@
         {
             "id": "llama3.1-8b",
             "name": "Llama 3.1 8B",
-            "parameters": "8B",
-            "speed_tokens_per_s": 2200,
-            "cost_per_1m_in": 0.10,
-            "cost_per_1m_out": 0.10,
+            "cost_per_1m_in": 0.1,
+            "cost_per_1m_out": 0.1,
             "context_window": 32768,
             "default_max_tokens": 4000,
             "can_reason": true,
@@ -34,10 +30,8 @@
         {
             "id": "llama-3.3-70b",
             "name": "Llama 3.3 70B",
-            "parameters": "70B",
-            "speed_tokens_per_s": 2100,
             "cost_per_1m_in": 0.85,
-            "cost_per_1m_out": 1.20,
+            "cost_per_1m_out": 1.2,
             "context_window": 128000,
             "default_max_tokens": 4000,
             "can_reason": true,
@@ -46,10 +40,8 @@
         {
             "id": "qwen-3-32b",
             "name": "Qwen 3 32B",
-            "parameters": "32B",
-            "speed_tokens_per_s": 2600,
-            "cost_per_1m_in": 0.40,
-            "cost_per_1m_out": 0.80,
+            "cost_per_1m_in": 0.4,
+            "cost_per_1m_out": 0.8,
             "context_window": 128000,
             "default_max_tokens": 32768,
             "can_reason": true,
@@ -58,10 +50,8 @@
         {
             "id": "llama-4-maverick-17b-128e-instruct",
             "name": "Llama 4 Maverick",
-            "parameters": "17B",
-            "speed_tokens_per_s": 1500,
-            "cost_per_1m_in": 0.20,
-            "cost_per_1m_out": 0.60,
+            "cost_per_1m_in": 0.2,
+            "cost_per_1m_out": 0.6,
             "context_window": 32768,
             "default_max_tokens": 4000,
             "can_reason": true,
@@ -70,10 +60,8 @@
         {
             "id": "qwen-3-235b-a22b-instruct-2507",
             "name": "Qwen 3 235B Instruct",
-            "parameters": "235B",
-            "speed_tokens_per_s": 1400,
-            "cost_per_1m_in": 0.60,
-            "cost_per_1m_out": 1.20,
+            "cost_per_1m_in": 0.6,
+            "cost_per_1m_out": 1.2,
             "context_window": 131072,
             "default_max_tokens": 16384,
             "can_reason": true,
@@ -82,10 +70,8 @@
         {
             "id": "qwen-3-235b-a22b-thinking-2507",
             "name": "Qwen 3 235B Thinking",
-            "parameters": "235B",
-            "speed_tokens_per_s": 1700,
-            "cost_per_1m_in": 0.60,
-            "cost_per_1m_out": 1.20,
+            "cost_per_1m_in": 0.6,
+            "cost_per_1m_out": 1.2,
             "context_window": 128000,
             "default_max_tokens": 32768,
             "can_reason": true,
@@ -94,17 +80,12 @@
         {
             "id": "qwen-3-coder-480b",
             "name": "Qwen 3 480B Coder",
-            "parameters": "480B",
-            "speed_tokens_per_s": 2000,
-            "cost_per_1m_in": 2.00,
-            "cost_per_1m_out": 2.00,
+            "cost_per_1m_in": 2.0,
+            "cost_per_1m_out": 2.0,
             "context_window": 131072,
             "default_max_tokens": 65536,
             "can_reason": true,
             "supports_attachments": false
         }
-    ],
-    "default_headers": {
-        "User-Agent": "Crush-Client/1.0"
-    }
-}
+    ]
+}