diff --git a/providers/stackit/models/Qwen/Qwen3-VL-235B-A22B-Instruct-FP8.toml b/providers/stackit/models/Qwen/Qwen3-VL-235B-A22B-Instruct-FP8.toml index 4df523123..80144afa5 100644 --- a/providers/stackit/models/Qwen/Qwen3-VL-235B-A22B-Instruct-FP8.toml +++ b/providers/stackit/models/Qwen/Qwen3-VL-235B-A22B-Instruct-FP8.toml @@ -10,12 +10,12 @@ structured_output = false open_weights = true [cost] -input = 1.64 -output = 1.91 +input = 1.76 +output = 2.05 [limit] context = 218_000 -output = 8_192 +output = 16_384 [modalities] input = ["text", "image"] diff --git a/providers/stackit/models/Qwen/Qwen3.6-27B.toml b/providers/stackit/models/Qwen/Qwen3.6-27B.toml new file mode 100644 index 000000000..0be7b70d2 --- /dev/null +++ b/providers/stackit/models/Qwen/Qwen3.6-27B.toml @@ -0,0 +1,22 @@ +name = "Qwen3.6-27B" +family = "qwen" +release_date = "2026-04-22" +last_updated = "2026-04-22" +attachment = true +reasoning = false +temperature = true +tool_call = true +structured_output = true +open_weights = true + +[cost] +input = 0.53 +output = 0.76 + +[limit] +context = 262_144 +output = 16_384 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/stackit/models/cortecs/Llama-3.3-70B-Instruct-FP8-Dynamic.toml b/providers/stackit/models/cortecs/Llama-3.3-70B-Instruct-FP8-Dynamic.toml index 468f3da31..cd31de457 100644 --- a/providers/stackit/models/cortecs/Llama-3.3-70B-Instruct-FP8-Dynamic.toml +++ b/providers/stackit/models/cortecs/Llama-3.3-70B-Instruct-FP8-Dynamic.toml @@ -10,12 +10,12 @@ structured_output = false open_weights = true [cost] -input = 0.49 -output = 0.71 +input = 0.53 +output = 0.76 [limit] context = 128_000 -output = 8_192 +output = 4_096 [modalities] input = ["text"] diff --git a/providers/stackit/models/google/gemma-3-27b-it.toml b/providers/stackit/models/google/gemma-3-27b-it.toml index c25df2ab6..48c83c263 100644 --- a/providers/stackit/models/google/gemma-3-27b-it.toml +++ b/providers/stackit/models/google/gemma-3-27b-it.toml @@ -10,12 +10,12 @@ structured_output = false open_weights = true [cost] -input = 0.49 -output = 0.71 +input = 0.53 +output = 0.76 [limit] context = 37_000 -output = 8_192 +output = 4_096 [modalities] input = ["text", "image"] diff --git a/providers/stackit/models/neuralmagic/Meta-Llama-3.1-8B-Instruct-FP8.toml b/providers/stackit/models/neuralmagic/Meta-Llama-3.1-8B-Instruct-FP8.toml deleted file mode 100644 index bdfbda001..000000000 --- a/providers/stackit/models/neuralmagic/Meta-Llama-3.1-8B-Instruct-FP8.toml +++ /dev/null @@ -1,22 +0,0 @@ -name = "Llama 3.1 8B" -family = "llama" -release_date = "2024-07-23" -last_updated = "2024-07-23" -attachment = false -reasoning = false -temperature = true -tool_call = true -structured_output = true -open_weights = true - -[cost] -input = 0.16 -output = 0.27 - -[limit] -context = 128_000 -output = 8_192 - -[modalities] -input = ["text"] -output = ["text"] diff --git a/providers/stackit/models/openai/gpt-oss-120b.toml b/providers/stackit/models/openai/gpt-oss-120b.toml index 33acefc69..f7da5ffbe 100644 --- a/providers/stackit/models/openai/gpt-oss-120b.toml +++ b/providers/stackit/models/openai/gpt-oss-120b.toml @@ -10,8 +10,8 @@ structured_output = false open_weights = true [cost] -input = 0.49 -output = 0.71 +input = 0.53 +output = 0.76 [limit] context = 131_000 diff --git a/providers/stackit/models/neuralmagic/Mistral-Nemo-Instruct-2407-FP8.toml b/providers/stackit/models/openai/gpt-oss-20b.toml similarity index 53% rename from providers/stackit/models/neuralmagic/Mistral-Nemo-Instruct-2407-FP8.toml rename to providers/stackit/models/openai/gpt-oss-20b.toml index 2f35759aa..ea765b84d 100644 --- a/providers/stackit/models/neuralmagic/Mistral-Nemo-Instruct-2407-FP8.toml +++ b/providers/stackit/models/openai/gpt-oss-20b.toml @@ -1,20 +1,20 @@ -name = "Mistral Nemo" -family = "mistral" -release_date = "2024-07-01" -last_updated = "2024-07-01" +name = "GPT-OSS 20B" +family = "gpt" +release_date = "2025-08-05" +last_updated = "2025-08-05" attachment = false -reasoning = false +reasoning = true temperature = true tool_call = true structured_output = false open_weights = true [cost] -input = 0.49 -output = 0.71 +input = 0.18 +output = 0.29 [limit] -context = 128_000 +context = 131_000 output = 8_192 [modalities]