From 60129bebc6b393d6081b8f3f62fe1edfcaf15a8e Mon Sep 17 00:00:00 2001 From: "github-actions[bot]" <41898282+github-actions[bot]@users.noreply.github.com> Date: Tue, 21 Jul 2026 14:57:07 +0000 Subject: [PATCH] chore(sync): update OpenRouter model catalog --- .../models/deepseek/deepseek-v4-flash.toml | 6 ++--- .../models/google/gemma-4-31b-it.toml | 7 +++-- .../models/moonshotai/kimi-k2.7-code.toml | 6 ++--- .../nvidia/nemotron-3-super-120b-a12b.toml | 4 +-- .../qwen/qwen3-next-80b-a3b-instruct.toml | 6 ++--- providers/openrouter/models/tencent/hy3.toml | 8 +++--- .../openrouter/models/tencent/hy3:free.toml | 27 ------------------- providers/openrouter/models/z-ai/glm-5.2.toml | 6 ++--- 8 files changed, 20 insertions(+), 50 deletions(-) delete mode 100644 providers/openrouter/models/tencent/hy3:free.toml diff --git a/providers/openrouter/models/deepseek/deepseek-v4-flash.toml b/providers/openrouter/models/deepseek/deepseek-v4-flash.toml index 33210a3fde..194406bf4c 100644 --- a/providers/openrouter/models/deepseek/deepseek-v4-flash.toml +++ b/providers/openrouter/models/deepseek/deepseek-v4-flash.toml @@ -8,9 +8,9 @@ type = "effort" values = ["high", "xhigh"] [cost] -input = 0.098 -output = 0.196 -cache_read = 0.0196 +input = 0.0938 +output = 0.1876 +cache_read = 0.01876 [limit] context = 1_048_576 diff --git a/providers/openrouter/models/google/gemma-4-31b-it.toml b/providers/openrouter/models/google/gemma-4-31b-it.toml index 4966c13144..8032b4a8f0 100644 --- a/providers/openrouter/models/google/gemma-4-31b-it.toml +++ b/providers/openrouter/models/google/gemma-4-31b-it.toml @@ -4,12 +4,11 @@ base_model = "google/gemma-4-31b-it" type = "toggle" [cost] -input = 0.22 -output = 0.55 -cache_read = 0.12 +input = 0.12 +output = 0.37 [limit] -output = 262_144 +output = 16_384 [modalities] input = ["image", "text", "video"] diff --git a/providers/openrouter/models/moonshotai/kimi-k2.7-code.toml b/providers/openrouter/models/moonshotai/kimi-k2.7-code.toml index c9fdd6fc98..715fa3d8a2 100644 --- a/providers/openrouter/models/moonshotai/kimi-k2.7-code.toml +++ b/providers/openrouter/models/moonshotai/kimi-k2.7-code.toml @@ -3,9 +3,9 @@ temperature = true reasoning_options = [] [cost] -input = 0.85 -output = 3.8 -cache_read = 0.17 +input = 0.82 +output = 3.75 +cache_read = 0.16 [modalities] input = ["text", "image"] diff --git a/providers/openrouter/models/nvidia/nemotron-3-super-120b-a12b.toml b/providers/openrouter/models/nvidia/nemotron-3-super-120b-a12b.toml index 3ae06cbadf..4eec59d79a 100644 --- a/providers/openrouter/models/nvidia/nemotron-3-super-120b-a12b.toml +++ b/providers/openrouter/models/nvidia/nemotron-3-super-120b-a12b.toml @@ -9,8 +9,8 @@ values = ["low", "medium"] type = "budget_tokens" [cost] -input = 0.085 -output = 0.4 +input = 0.08 +output = 0.45 [limit] context = 1_000_000 diff --git a/providers/openrouter/models/qwen/qwen3-next-80b-a3b-instruct.toml b/providers/openrouter/models/qwen/qwen3-next-80b-a3b-instruct.toml index 6d8a40b129..d622551a9a 100644 --- a/providers/openrouter/models/qwen/qwen3-next-80b-a3b-instruct.toml +++ b/providers/openrouter/models/qwen/qwen3-next-80b-a3b-instruct.toml @@ -2,10 +2,8 @@ base_model = "alibaba/qwen3-next-80b-a3b-instruct" structured_output = true [cost] -input = 0.1 -output = 1.1 -cache_read = 0.07 +input = 0.0975 +output = 0.78 [limit] context = 262_144 -output = 262_144 diff --git a/providers/openrouter/models/tencent/hy3.toml b/providers/openrouter/models/tencent/hy3.toml index 1bd6a030ae..fba8f06a0a 100644 --- a/providers/openrouter/models/tencent/hy3.toml +++ b/providers/openrouter/models/tencent/hy3.toml @@ -15,13 +15,13 @@ type = "effort" values = ["none", "low", "high"] [cost] -input = 0.2 -output = 0.8 -cache_read = 0.05 +input = 0.14 +output = 0.58 +cache_read = 0.035 [limit] context = 262_144 -output = 131_072 +output = 262_144 [modalities] input = ["text"] diff --git a/providers/openrouter/models/tencent/hy3:free.toml b/providers/openrouter/models/tencent/hy3:free.toml deleted file mode 100644 index 055296d72e..0000000000 --- a/providers/openrouter/models/tencent/hy3:free.toml +++ /dev/null @@ -1,27 +0,0 @@ -name = "Hy3 (free)" -description = "Tencent Hy reasoning model for coding, instruction following, and agent tasks" -family = "hy3" -release_date = "2026-07-06" -last_updated = "2026-07-06" -attachment = false -reasoning = true -temperature = true -tool_call = true -structured_output = true -open_weights = true - -[[reasoning_options]] -type = "effort" -values = ["none", "low", "high"] - -[cost] -input = 0 -output = 0 - -[limit] -context = 262_144 -output = 262_144 - -[modalities] -input = ["text"] -output = ["text"] diff --git a/providers/openrouter/models/z-ai/glm-5.2.toml b/providers/openrouter/models/z-ai/glm-5.2.toml index 84fff5dbdc..cabe568ad7 100644 --- a/providers/openrouter/models/z-ai/glm-5.2.toml +++ b/providers/openrouter/models/z-ai/glm-5.2.toml @@ -8,9 +8,9 @@ type = "effort" values = ["high", "xhigh"] [cost] -input = 0.9688 -output = 3.0448 -cache_read = 0.17992 +input = 0.8148 +output = 2.5608 +cache_read = 0.15132 [limit] context = 1_048_576