chore(sync): update OpenRouter model catalog

providers/openrouter/models/deepseek/deepseek-r1-0528.toml

@@ -1,26 +1,8 @@
-name = "R1 0528"
+base_model = "deepseek/deepseek-r1-0528"
 description = "DeepSeek reasoning model for multi-step analysis, math, coding, and tools"
-family = "deepseek"
-release_date = "2025-05-28"
-last_updated = "2025-05-28"
-attachment = false
-reasoning = true
-temperature = true
-tool_call = true
-structured_output = true
-knowledge = "2025-03-31"
-open_weights = true
 reasoning_options = []
 
 [cost]
 input = 0.5
 output = 2.15
 cache_read = 0.35
-
-[limit]
-context = 163_840
-output = 32_768
-
-[modalities]
-input = ["text"]
-output = ["text"]

providers/openrouter/models/deepseek/deepseek-v4-flash-0731.toml

@@ -10,9 +10,9 @@ type = "effort"
 values = ["low", "high", "max"]
 
 [cost]
-input = 0.0079
+input = 0.0064
 output = 1.28
-cache_read = 0.0079
+cache_read = 0.0064
 
 [limit]
 context = 1_048_576

providers/openrouter/models/deepseek/deepseek-v4-flash.toml

@@ -13,9 +13,9 @@ type = "effort"
 values = ["high", "xhigh"]
 
 [cost]
-input = 0.03
+input = 0.0131
 output = 1.28
-cache_read = 0.03
+cache_read = 0.0131
 
 [limit]
 context = 1_048_576

providers/openrouter/models/deepseek/deepseek-v4-pro.toml

@@ -13,9 +13,9 @@ type = "effort"
 values = ["high", "xhigh"]
 
 [cost]
-input = 0.2871
-output = 0.5742
-cache_read = 0.023925
+input = 0.2088
+output = 0.4176
+cache_read = 0.0174
 
 [limit]
 context = 1_048_576

providers/openrouter/models/meta-llama/llama-3.2-1b-instruct.toml

@@ -1,15 +1,7 @@
-name = "Llama 3.2 1B Instruct"
+base_model = "meta/llama-3.2-1b-instruct"
 description = "Open Llama instruction model for multilingual chat, reasoning, and coding"
-family = "llama"
-release_date = "2024-09-25"
-last_updated = "2024-09-25"
-attachment = false
-reasoning = false
-temperature = true
 tool_call = false
 structured_output = false
-knowledge = "2023-12-31"
-open_weights = true
 
 [cost]
 input = 0.027
@@ -18,7 +10,3 @@ output = 0.201
 [limit]
 context = 60_000
 output = 54_000
-
-[modalities]
-input = ["text"]
-output = ["text"]

providers/openrouter/models/moonshotai/kimi-k2-0905.toml

@@ -1,24 +1,7 @@
-name = "Kimi K2 0905"
+base_model = "moonshotai/kimi-k2-0905"
 description = "Kimi model for long-context chat, coding, and agentic reasoning"
-family = "kimi-k2"
-release_date = "2025-09-04"
-last_updated = "2025-09-04"
-attachment = false
-reasoning = false
-temperature = true
-tool_call = true
 structured_output = true
-knowledge = "2024-12-31"
-open_weights = true
 
 [cost]
 input = 0.6
 output = 2.5
-
-[limit]
-context = 262_144
-output = 98_304
-
-[modalities]
-input = ["text"]
-output = ["text"]

providers/openrouter/models/moonshotai/kimi-k2-thinking.toml

@@ -8,6 +8,7 @@ field = "reasoning_details"
 [cost]
 input = 0.6
 output = 2.5
+cache_read = 0.15
 
 [limit]
-output = 235_929
+output = 98_304

providers/openrouter/models/moonshotai/kimi-k2.6.toml

@@ -9,9 +9,9 @@ field = "reasoning_details"
 type = "toggle"
 
 [cost]
-input = 0.4348
+input = 0.4344
 output = 2.45
-cache_read = 0.1202
+cache_read = 0.1077
 
 [limit]
 output = 235_929

providers/openrouter/models/moonshotai/kimi-k3.toml

@@ -12,9 +12,9 @@ type = "effort"
 values = ["low", "high", "max"]
 
 [cost]
-input = 0.5
-output = 12
-cache_read = 0.33
+input = 0.9
+output = 14
+cache_read = 0.6
 
 [limit]
 output = 943_718

providers/openrouter/models/qwen/qwen3-14b.toml

@@ -21,7 +21,7 @@ input = 0.12
 output = 0.24
 
 [limit]
-context = 131_072
+context = 40_960
 output = 16_384
 
 [modalities]

providers/openrouter/models/qwen/qwen3-235b-a22b-thinking-2507.toml

@@ -13,12 +13,12 @@ open_weights = true
 reasoning_options = []
 
 [cost]
-input = 0.23
-output = 2.3
+input = 0.45
+output = 3.5
 
 [limit]
-context = 131_072
-output = 117_964
+context = 128_000
+output = 16_384
 
 [modalities]
 input = ["text"]

providers/openrouter/models/qwen/qwen3-30b-a3b.toml

@@ -10,3 +10,6 @@ type = "toggle"
 [cost]
 input = 0.12
 output = 0.5
+
+[limit]
+context = 40_960

providers/openrouter/models/qwen/qwen3-next-80b-a3b-thinking.toml

@@ -8,3 +8,4 @@ output = 1.2
 
 [limit]
 context = 262_144
+output = 235_929