diff --git a/providers/cortecs/models/codestral-2508.toml b/providers/cortecs/models/codestral-2508.toml index f9547b3dd4e..844a12bf1a9 100644 --- a/providers/cortecs/models/codestral-2508.toml +++ b/providers/cortecs/models/codestral-2508.toml @@ -12,9 +12,9 @@ knowledge = "2025-03" open_weights = true [cost] -input = 0.334 -output = 1.003 -cache_read = 0.033 +input = 0.368 +output = 1.103 +cache_read = 0.037 [limit] context = 256_000 diff --git a/providers/cortecs/models/deepseek-v4.1-flash.toml b/providers/cortecs/models/deepseek-v4.1-flash.toml new file mode 100644 index 00000000000..1a21e3f6a14 --- /dev/null +++ b/providers/cortecs/models/deepseek-v4.1-flash.toml @@ -0,0 +1,13 @@ +# Effort: reasoning_effort = low|medium|high +# https://docs.cortecs.ai/routing/reasoning +base_model = "deepseek/deepseek-v4.1-flash" +reasoning_options = [{ type = "effort", values = ["low", "medium", "high"] }] + +[cost] +input = 0.5 +output = 1.499 +cache_read = 0.13 + +[limit] +context = 1_048_576 +output = 64_000 diff --git a/providers/cortecs/models/ministral-14b-2512.toml b/providers/cortecs/models/ministral-14b-2512.toml index e57017e1d77..6b79c2b3c20 100644 --- a/providers/cortecs/models/ministral-14b-2512.toml +++ b/providers/cortecs/models/ministral-14b-2512.toml @@ -10,8 +10,8 @@ structured_output = true open_weights = false [cost] -input = 0.223 -output = 0.223 +input = 0.24 +output = 0.24 cache_read = 0.022 [limit] diff --git a/providers/cortecs/models/ministral-3b-2512.toml b/providers/cortecs/models/ministral-3b-2512.toml index 4e686e35de2..036dc387677 100644 --- a/providers/cortecs/models/ministral-3b-2512.toml +++ b/providers/cortecs/models/ministral-3b-2512.toml @@ -10,9 +10,9 @@ structured_output = true open_weights = false [cost] -input = 0.111 -output = 0.111 -cache_read = 0.011 +input = 0.123 +output = 0.123 +cache_read = 0.012 [limit] context = 256_000 diff --git a/providers/cortecs/models/ministral-8b-2512.toml b/providers/cortecs/models/ministral-8b-2512.toml index ddbd9624484..429f3dd2215 100644 --- a/providers/cortecs/models/ministral-8b-2512.toml +++ b/providers/cortecs/models/ministral-8b-2512.toml @@ -10,8 +10,8 @@ structured_output = true open_weights = false [cost] -input = 0.167 -output = 0.167 +input = 0.179 +output = 0.179 cache_read = 0.017 [limit] diff --git a/providers/cortecs/models/mistral-large-2512.toml b/providers/cortecs/models/mistral-large-2512.toml index 52b4e581a25..cbca6a6506d 100644 --- a/providers/cortecs/models/mistral-large-2512.toml +++ b/providers/cortecs/models/mistral-large-2512.toml @@ -2,9 +2,9 @@ base_model = "mistral/mistral-large-2512" structured_output = true [cost] -input = 0.557 -output = 1.671 -cache_read = 0.056 +input = 0.613 +output = 1.838 +cache_read = 0.061 [limit] context = 256_000 diff --git a/providers/cortecs/models/mistral-medium-3.5.toml b/providers/cortecs/models/mistral-medium-3.5.toml index e814aa7640a..7c463c40224 100644 --- a/providers/cortecs/models/mistral-medium-3.5.toml +++ b/providers/cortecs/models/mistral-medium-3.5.toml @@ -11,9 +11,9 @@ open_weights = false reasoning_options = [] [cost] -input = 1.393 -output = 7.13 -cache_read = 0.139 +input = 1.532 +output = 7.843 +cache_read = 0.154 [limit] context = 256_000 diff --git a/providers/cortecs/models/mistral-small-2603.toml b/providers/cortecs/models/mistral-small-2603.toml index 89e30734258..050ea77050d 100644 --- a/providers/cortecs/models/mistral-small-2603.toml +++ b/providers/cortecs/models/mistral-small-2603.toml @@ -3,9 +3,9 @@ structured_output = true reasoning_options = [] [cost] -input = 0.143 -output = 0.568 -cache_read = 0.014 +input = 0.156 +output = 0.625 +cache_read = 0.016 [limit] context = 262_144 diff --git a/providers/cortecs/models/voxtral-small-2507.toml b/providers/cortecs/models/voxtral-small-2507.toml index f8ce5732ff8..040a8cd7951 100644 --- a/providers/cortecs/models/voxtral-small-2507.toml +++ b/providers/cortecs/models/voxtral-small-2507.toml @@ -10,9 +10,9 @@ structured_output = true open_weights = false [cost] -input = 0.111 -output = 0.334 -cache_read = 0.011 +input = 0.123 +output = 0.368 +cache_read = 0.012 [limit] context = 32_000