From 85d338da8bc9bc1f9ea8fa7cc4ccb0f1940e4407 Mon Sep 17 00:00:00 2001 From: Yujong Lee Date: Fri, 18 Sep 2026 04:25:44 +0000 Subject: [PATCH 1/9] feat(dashscope): add qwen3.8-omni-flash and qwen3.8-flash to the model cost map Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- ...odel_prices_and_context_window_backup.json | 74 +++++++++++++++++++ model_prices_and_context_window.json | 74 +++++++++++++++++++ 2 files changed, 148 insertions(+) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 355e2b90e96..3409d9b4e46 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -16690,6 +16690,43 @@ "supports_tool_choice": true, "supports_vision": true }, + "dashscope/qwen3.8-flash": { + "cache_read_input_token_cost": 1.6e-08, + "input_cost_per_token": 1.5e-07, + "litellm_provider": "dashscope", + "max_input_tokens": 991808, + "max_output_tokens": 131072, + "max_tokens": 131072, + "mode": "chat", + "output_cost_per_token": 4.7e-07, + "source": "https://docs.modelstudio.console.alibabacloud.com/en/model-studio/model-pricing", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_web_search": true + }, + "dashscope/qwen3.8-omni-flash": { + "cache_read_input_token_cost": 1.6e-08, + "input_cost_per_token": 1.5e-07, + "litellm_provider": "dashscope", + "max_input_tokens": 991808, + "max_output_tokens": 131072, + "max_tokens": 131072, + "mode": "chat", + "output_cost_per_token": 4.7e-07, + "source": "https://docs.modelstudio.console.alibabacloud.com/en/model-studio/model-pricing", + "supports_audio_input": true, + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_video_input": true, + "supports_vision": true, + "supports_web_search": true + }, "dashscope/qwq-plus": { "input_cost_per_token": 8e-07, "litellm_provider": "dashscope", @@ -18594,6 +18631,43 @@ "supports_tool_choice": true, "supports_vision": true }, + "qwen_ai_platform/qwen3.8-flash": { + "cache_read_input_token_cost": 1.6e-08, + "input_cost_per_token": 1.5e-07, + "litellm_provider": "qwen_ai_platform", + "max_input_tokens": 991808, + "max_output_tokens": 131072, + "max_tokens": 131072, + "mode": "chat", + "output_cost_per_token": 4.7e-07, + "source": "https://docs.modelstudio.console.alibabacloud.com/en/model-studio/model-pricing", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_web_search": true + }, + "qwen_ai_platform/qwen3.8-omni-flash": { + "cache_read_input_token_cost": 1.6e-08, + "input_cost_per_token": 1.5e-07, + "litellm_provider": "qwen_ai_platform", + "max_input_tokens": 991808, + "max_output_tokens": 131072, + "max_tokens": 131072, + "mode": "chat", + "output_cost_per_token": 4.7e-07, + "source": "https://docs.modelstudio.console.alibabacloud.com/en/model-studio/model-pricing", + "supports_audio_input": true, + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_video_input": true, + "supports_vision": true, + "supports_web_search": true + }, "qwen_ai_platform/qwq-plus": { "input_cost_per_token": 8e-07, "litellm_provider": "qwen_ai_platform", diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 355e2b90e96..3409d9b4e46 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -16690,6 +16690,43 @@ "supports_tool_choice": true, "supports_vision": true }, + "dashscope/qwen3.8-flash": { + "cache_read_input_token_cost": 1.6e-08, + "input_cost_per_token": 1.5e-07, + "litellm_provider": "dashscope", + "max_input_tokens": 991808, + "max_output_tokens": 131072, + "max_tokens": 131072, + "mode": "chat", + "output_cost_per_token": 4.7e-07, + "source": "https://docs.modelstudio.console.alibabacloud.com/en/model-studio/model-pricing", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_web_search": true + }, + "dashscope/qwen3.8-omni-flash": { + "cache_read_input_token_cost": 1.6e-08, + "input_cost_per_token": 1.5e-07, + "litellm_provider": "dashscope", + "max_input_tokens": 991808, + "max_output_tokens": 131072, + "max_tokens": 131072, + "mode": "chat", + "output_cost_per_token": 4.7e-07, + "source": "https://docs.modelstudio.console.alibabacloud.com/en/model-studio/model-pricing", + "supports_audio_input": true, + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_video_input": true, + "supports_vision": true, + "supports_web_search": true + }, "dashscope/qwq-plus": { "input_cost_per_token": 8e-07, "litellm_provider": "dashscope", @@ -18594,6 +18631,43 @@ "supports_tool_choice": true, "supports_vision": true }, + "qwen_ai_platform/qwen3.8-flash": { + "cache_read_input_token_cost": 1.6e-08, + "input_cost_per_token": 1.5e-07, + "litellm_provider": "qwen_ai_platform", + "max_input_tokens": 991808, + "max_output_tokens": 131072, + "max_tokens": 131072, + "mode": "chat", + "output_cost_per_token": 4.7e-07, + "source": "https://docs.modelstudio.console.alibabacloud.com/en/model-studio/model-pricing", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_web_search": true + }, + "qwen_ai_platform/qwen3.8-omni-flash": { + "cache_read_input_token_cost": 1.6e-08, + "input_cost_per_token": 1.5e-07, + "litellm_provider": "qwen_ai_platform", + "max_input_tokens": 991808, + "max_output_tokens": 131072, + "max_tokens": 131072, + "mode": "chat", + "output_cost_per_token": 4.7e-07, + "source": "https://docs.modelstudio.console.alibabacloud.com/en/model-studio/model-pricing", + "supports_audio_input": true, + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_video_input": true, + "supports_vision": true, + "supports_web_search": true + }, "qwen_ai_platform/qwq-plus": { "input_cost_per_token": 8e-07, "litellm_provider": "qwen_ai_platform", From 27b566590123e494107d1b39bab5a411a8d045b0 Mon Sep 17 00:00:00 2001 From: Yujong Lee Date: Fri, 18 Sep 2026 04:45:28 +0000 Subject: [PATCH 2/9] fix(dashscope): mark qwen3.8-flash as vision and video capable with explicit cache creation cost Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- litellm/model_prices_and_context_window_backup.json | 6 ++++++ model_prices_and_context_window.json | 6 ++++++ 2 files changed, 12 insertions(+) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 3409d9b4e46..e903b239881 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -16691,6 +16691,7 @@ "supports_vision": true }, "dashscope/qwen3.8-flash": { + "cache_creation_input_token_cost": 2e-07, "cache_read_input_token_cost": 1.6e-08, "input_cost_per_token": 1.5e-07, "litellm_provider": "dashscope", @@ -16705,6 +16706,8 @@ "supports_reasoning": true, "supports_response_schema": true, "supports_tool_choice": true, + "supports_video_input": true, + "supports_vision": true, "supports_web_search": true }, "dashscope/qwen3.8-omni-flash": { @@ -18632,6 +18635,7 @@ "supports_vision": true }, "qwen_ai_platform/qwen3.8-flash": { + "cache_creation_input_token_cost": 2e-07, "cache_read_input_token_cost": 1.6e-08, "input_cost_per_token": 1.5e-07, "litellm_provider": "qwen_ai_platform", @@ -18646,6 +18650,8 @@ "supports_reasoning": true, "supports_response_schema": true, "supports_tool_choice": true, + "supports_video_input": true, + "supports_vision": true, "supports_web_search": true }, "qwen_ai_platform/qwen3.8-omni-flash": { diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 3409d9b4e46..e903b239881 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -16691,6 +16691,7 @@ "supports_vision": true }, "dashscope/qwen3.8-flash": { + "cache_creation_input_token_cost": 2e-07, "cache_read_input_token_cost": 1.6e-08, "input_cost_per_token": 1.5e-07, "litellm_provider": "dashscope", @@ -16705,6 +16706,8 @@ "supports_reasoning": true, "supports_response_schema": true, "supports_tool_choice": true, + "supports_video_input": true, + "supports_vision": true, "supports_web_search": true }, "dashscope/qwen3.8-omni-flash": { @@ -18632,6 +18635,7 @@ "supports_vision": true }, "qwen_ai_platform/qwen3.8-flash": { + "cache_creation_input_token_cost": 2e-07, "cache_read_input_token_cost": 1.6e-08, "input_cost_per_token": 1.5e-07, "litellm_provider": "qwen_ai_platform", @@ -18646,6 +18650,8 @@ "supports_reasoning": true, "supports_response_schema": true, "supports_tool_choice": true, + "supports_video_input": true, + "supports_vision": true, "supports_web_search": true }, "qwen_ai_platform/qwen3.8-omni-flash": { From ca405879a2ba9915b4fec8a35efc48360a719d1e Mon Sep 17 00:00:00 2001 From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Date: Fri, 18 Sep 2026 13:22:57 +0000 Subject: [PATCH 3/9] fix(cohere): set embed v3 context length to 512 tokens Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- .../model_prices_and_context_window_backup.json | 16 ++++++++-------- model_prices_and_context_window.json | 16 ++++++++-------- 2 files changed, 16 insertions(+), 16 deletions(-) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 92df776adf4..60f48ec9679 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -22170,8 +22170,8 @@ "embed-english-light-v3.0": { "input_cost_per_token": 1e-07, "litellm_provider": "cohere", - "max_input_tokens": 1024, - "max_tokens": 1024, + "max_input_tokens": 512, + "max_tokens": 512, "mode": "embedding", "output_cost_per_token": 0.0 }, @@ -22188,8 +22188,8 @@ "input_cost_per_image": 0.0001, "input_cost_per_token": 1e-07, "litellm_provider": "cohere", - "max_input_tokens": 1024, - "max_tokens": 1024, + "max_input_tokens": 512, + "max_tokens": 512, "metadata": { "notes": "'supports_image_input' is a deprecated field. Use 'supports_embedding_image_input' instead." }, @@ -22210,8 +22210,8 @@ "embed-multilingual-v3.0": { "input_cost_per_token": 1e-07, "litellm_provider": "cohere", - "max_input_tokens": 1024, - "max_tokens": 1024, + "max_input_tokens": 512, + "max_tokens": 512, "mode": "embedding", "output_cost_per_token": 0.0, "supports_embedding_image_input": true @@ -22219,8 +22219,8 @@ "embed-multilingual-light-v3.0": { "input_cost_per_token": 0.0001, "litellm_provider": "cohere", - "max_input_tokens": 1024, - "max_tokens": 1024, + "max_input_tokens": 512, + "max_tokens": 512, "mode": "embedding", "output_cost_per_token": 0.0, "supports_embedding_image_input": true diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 92df776adf4..60f48ec9679 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -22170,8 +22170,8 @@ "embed-english-light-v3.0": { "input_cost_per_token": 1e-07, "litellm_provider": "cohere", - "max_input_tokens": 1024, - "max_tokens": 1024, + "max_input_tokens": 512, + "max_tokens": 512, "mode": "embedding", "output_cost_per_token": 0.0 }, @@ -22188,8 +22188,8 @@ "input_cost_per_image": 0.0001, "input_cost_per_token": 1e-07, "litellm_provider": "cohere", - "max_input_tokens": 1024, - "max_tokens": 1024, + "max_input_tokens": 512, + "max_tokens": 512, "metadata": { "notes": "'supports_image_input' is a deprecated field. Use 'supports_embedding_image_input' instead." }, @@ -22210,8 +22210,8 @@ "embed-multilingual-v3.0": { "input_cost_per_token": 1e-07, "litellm_provider": "cohere", - "max_input_tokens": 1024, - "max_tokens": 1024, + "max_input_tokens": 512, + "max_tokens": 512, "mode": "embedding", "output_cost_per_token": 0.0, "supports_embedding_image_input": true @@ -22219,8 +22219,8 @@ "embed-multilingual-light-v3.0": { "input_cost_per_token": 0.0001, "litellm_provider": "cohere", - "max_input_tokens": 1024, - "max_tokens": 1024, + "max_input_tokens": 512, + "max_tokens": 512, "mode": "embedding", "output_cost_per_token": 0.0, "supports_embedding_image_input": true From 2299846b48e4c29868bdbd7a5e145493e4de067a Mon Sep 17 00:00:00 2001 From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Date: Fri, 18 Sep 2026 19:18:02 +0000 Subject: [PATCH 4/9] fix(bedrock_mantle): align gpt-5.6-sol pricing with the AWS in-region model card Co-authored-by: kusumakarb Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- .../model_prices_and_context_window_backup.json | 16 ++++++++-------- model_prices_and_context_window.json | 16 ++++++++-------- ...st_bedrock_mantle_responses_transformation.py | 16 ++++++++++++++++ 3 files changed, 32 insertions(+), 16 deletions(-) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 94e20fa0213..b91c7274a1d 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -57990,14 +57990,14 @@ "supports_tool_choice": true }, "bedrock_mantle/openai.gpt-5.6-sol": { - "input_cost_per_token": 5.5e-06, - "input_cost_per_token_above_272k_tokens": 1.1e-05, - "cache_creation_input_token_cost": 6.875e-06, - "cache_creation_input_token_cost_above_272k_tokens": 1.375e-05, - "cache_read_input_token_cost": 5.5e-07, - "cache_read_input_token_cost_above_272k_tokens": 1.1e-06, - "output_cost_per_token": 3.3e-05, - "output_cost_per_token_above_272k_tokens": 4.95e-05, + "input_cost_per_token": 4.4e-06, + "input_cost_per_token_above_272k_tokens": 8.8e-06, + "cache_creation_input_token_cost": 5.5e-06, + "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05, + "cache_read_input_token_cost": 4.4e-07, + "cache_read_input_token_cost_above_272k_tokens": 8.8e-07, + "output_cost_per_token": 2.2e-05, + "output_cost_per_token_above_272k_tokens": 3.3e-05, "search_context_cost_per_query": { "search_context_size_high": 0.012, "search_context_size_low": 0.012, diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 94e20fa0213..b91c7274a1d 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -57990,14 +57990,14 @@ "supports_tool_choice": true }, "bedrock_mantle/openai.gpt-5.6-sol": { - "input_cost_per_token": 5.5e-06, - "input_cost_per_token_above_272k_tokens": 1.1e-05, - "cache_creation_input_token_cost": 6.875e-06, - "cache_creation_input_token_cost_above_272k_tokens": 1.375e-05, - "cache_read_input_token_cost": 5.5e-07, - "cache_read_input_token_cost_above_272k_tokens": 1.1e-06, - "output_cost_per_token": 3.3e-05, - "output_cost_per_token_above_272k_tokens": 4.95e-05, + "input_cost_per_token": 4.4e-06, + "input_cost_per_token_above_272k_tokens": 8.8e-06, + "cache_creation_input_token_cost": 5.5e-06, + "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05, + "cache_read_input_token_cost": 4.4e-07, + "cache_read_input_token_cost_above_272k_tokens": 8.8e-07, + "output_cost_per_token": 2.2e-05, + "output_cost_per_token_above_272k_tokens": 3.3e-05, "search_context_cost_per_query": { "search_context_size_high": 0.012, "search_context_size_low": 0.012, diff --git a/tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py b/tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py index 901c005f5a3..81c39f736a5 100644 --- a/tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py +++ b/tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py @@ -1839,6 +1839,22 @@ class TestBedrockMantleResponsesSigV4: class TestBedrockMantleResponsesPricing: + @pytest.mark.parametrize( + "model", + ["openai.gpt-5.6-sol", "openai.gpt-5.6-terra", "openai.gpt-5.6-luna"], + ) + def test_mantle_matches_in_region_converse_pricing(self, local_cost_map, model): + mantle = litellm.model_cost[f"bedrock_mantle/{model}"] + converse = litellm.model_cost[f"us.{model}"] + + cost_fields = [k for k in converse if "cost" in k and k != "search_context_cost_per_query"] + assert cost_fields, "expected cost fields on the converse entry" + for field in cost_fields: + assert mantle.get(field) == pytest.approx(converse[field]), ( + f"{model}: {field} is {mantle.get(field)} on bedrock_mantle " + f"but {converse[field]} on us. (bedrock_converse)" + ) + def test_models_registered(self, local_cost_map): assert "bedrock_mantle/openai.gpt-5.5" in litellm.bedrock_mantle_models assert "bedrock_mantle/openai.gpt-5.4" in litellm.bedrock_mantle_models From 31f1c5d33107d6cd5179c3ec8c6074a8d5742f87 Mon Sep 17 00:00:00 2001 From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Date: Fri, 18 Sep 2026 19:18:14 +0000 Subject: [PATCH 5/9] fix(openrouter): refresh glm-5.3, glm-latest, deepseek-flash-latest and hy3 pricing Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- ...odel_prices_and_context_window_backup.json | 22 +++++++++---------- model_prices_and_context_window.json | 22 +++++++++---------- 2 files changed, 22 insertions(+), 22 deletions(-) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index b91c7274a1d..f8a537fb922 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -65896,9 +65896,9 @@ "supports_web_search": false }, "openrouter/z-ai/glm-5.3": { - "input_cost_per_token": 1.4e-06, - "output_cost_per_token": 4.4e-06, - "cache_read_input_token_cost": 2.6e-07, + "input_cost_per_token": 9.1e-07, + "output_cost_per_token": 2.86e-06, + "cache_read_input_token_cost": 1.69e-07, "litellm_provider": "openrouter", "max_input_tokens": 1310720, "max_output_tokens": 943717, @@ -70619,14 +70619,14 @@ "supports_web_search": false }, "openrouter/~deepseek/deepseek-flash-latest": { - "cache_read_input_token_cost": 1.5e-08, - "input_cost_per_token": 1.5e-07, + "cache_read_input_token_cost": 4.2e-09, + "input_cost_per_token": 1.4e-07, "litellm_provider": "openrouter", "max_input_tokens": 1048576, "max_output_tokens": 393216, "max_tokens": 393216, "mode": "chat", - "output_cost_per_token": 6e-07, + "output_cost_per_token": 4.2e-07, "source": "https://openrouter.ai/api/v1/models", "supports_audio_input": false, "supports_function_calling": true, @@ -70911,14 +70911,14 @@ "supports_web_search": false }, "openrouter/~z-ai/glm-latest": { - "cache_read_input_token_cost": 1.5e-07, + "cache_read_input_token_cost": 1.46625e-07, "input_cost_per_token": 9e-07, "litellm_provider": "openrouter", "max_input_tokens": 1310720, "max_output_tokens": 235929, "max_tokens": 235929, "mode": "chat", - "output_cost_per_token": 3e-06, + "output_cost_per_token": 2.805e-06, "source": "https://openrouter.ai/api/v1/models", "supports_audio_input": false, "supports_function_calling": true, @@ -74062,14 +74062,14 @@ "supports_web_search": false }, "openrouter/tencent/hy3": { - "cache_read_input_token_cost": 3.3e-08, - "input_cost_per_token": 1.32e-07, + "cache_read_input_token_cost": 2.0625e-08, + "input_cost_per_token": 8.25e-08, "litellm_provider": "openrouter", "max_input_tokens": 262144, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", - "output_cost_per_token": 5.28e-07, + "output_cost_per_token": 3.3e-07, "source": "https://openrouter.ai/api/v1/models", "supports_audio_input": false, "supports_function_calling": true, diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index b91c7274a1d..f8a537fb922 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -65896,9 +65896,9 @@ "supports_web_search": false }, "openrouter/z-ai/glm-5.3": { - "input_cost_per_token": 1.4e-06, - "output_cost_per_token": 4.4e-06, - "cache_read_input_token_cost": 2.6e-07, + "input_cost_per_token": 9.1e-07, + "output_cost_per_token": 2.86e-06, + "cache_read_input_token_cost": 1.69e-07, "litellm_provider": "openrouter", "max_input_tokens": 1310720, "max_output_tokens": 943717, @@ -70619,14 +70619,14 @@ "supports_web_search": false }, "openrouter/~deepseek/deepseek-flash-latest": { - "cache_read_input_token_cost": 1.5e-08, - "input_cost_per_token": 1.5e-07, + "cache_read_input_token_cost": 4.2e-09, + "input_cost_per_token": 1.4e-07, "litellm_provider": "openrouter", "max_input_tokens": 1048576, "max_output_tokens": 393216, "max_tokens": 393216, "mode": "chat", - "output_cost_per_token": 6e-07, + "output_cost_per_token": 4.2e-07, "source": "https://openrouter.ai/api/v1/models", "supports_audio_input": false, "supports_function_calling": true, @@ -70911,14 +70911,14 @@ "supports_web_search": false }, "openrouter/~z-ai/glm-latest": { - "cache_read_input_token_cost": 1.5e-07, + "cache_read_input_token_cost": 1.46625e-07, "input_cost_per_token": 9e-07, "litellm_provider": "openrouter", "max_input_tokens": 1310720, "max_output_tokens": 235929, "max_tokens": 235929, "mode": "chat", - "output_cost_per_token": 3e-06, + "output_cost_per_token": 2.805e-06, "source": "https://openrouter.ai/api/v1/models", "supports_audio_input": false, "supports_function_calling": true, @@ -74062,14 +74062,14 @@ "supports_web_search": false }, "openrouter/tencent/hy3": { - "cache_read_input_token_cost": 3.3e-08, - "input_cost_per_token": 1.32e-07, + "cache_read_input_token_cost": 2.0625e-08, + "input_cost_per_token": 8.25e-08, "litellm_provider": "openrouter", "max_input_tokens": 262144, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", - "output_cost_per_token": 5.28e-07, + "output_cost_per_token": 3.3e-07, "source": "https://openrouter.ai/api/v1/models", "supports_audio_input": false, "supports_function_calling": true, From a774e71175177641c2914dce33ac5d8d106ca131 Mon Sep 17 00:00:00 2001 From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Date: Fri, 18 Sep 2026 19:18:45 +0000 Subject: [PATCH 6/9] test(bedrock_mantle): document mantle in-region pricing invariant Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- .../test_bedrock_mantle_responses_transformation.py | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py b/tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py index 81c39f736a5..951911ac066 100644 --- a/tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py +++ b/tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py @@ -1844,6 +1844,11 @@ class TestBedrockMantleResponsesPricing: ["openai.gpt-5.6-sol", "openai.gpt-5.6-terra", "openai.gpt-5.6-luna"], ) def test_mantle_matches_in_region_converse_pricing(self, local_cost_map, model): + """bedrock-mantle serves these models In-Region only, and the AWS model + cards price In-Region and Geo CRIS identically -- so every cost field on + the mantle key must equal the `us.` converse key. A price change applied + to one namespace but not the other shows up here. + """ mantle = litellm.model_cost[f"bedrock_mantle/{model}"] converse = litellm.model_cost[f"us.{model}"] From 4f904f69b6235497f784839be5078f1f77d5255d Mon Sep 17 00:00:00 2001 From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Date: Fri, 18 Sep 2026 19:44:18 +0000 Subject: [PATCH 7/9] test(proxy): restore app routes after pass-through reload tests The two pass-through reload tests register real FastAPI routes on the shared proxy app and only clean up the internal registry, so any test running after them in the same worker sees stray /v1/kept-* and /v1/deleted-* routes. test_component_allowlists counts those as uncovered and fails. Snapshot app.routes and the registry up front and restore both in a finally block. Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- tests/test_litellm/proxy/test_proxy_server.py | 52 +++++++++++++------ 1 file changed, 35 insertions(+), 17 deletions(-) diff --git a/tests/test_litellm/proxy/test_proxy_server.py b/tests/test_litellm/proxy/test_proxy_server.py index ce809eda847..19822032d1f 100644 --- a/tests/test_litellm/proxy/test_proxy_server.py +++ b/tests/test_litellm/proxy/test_proxy_server.py @@ -7536,25 +7536,35 @@ async def test_deleting_the_stored_pass_through_row_takes_the_route_out_of_servi deleted. The proxy's own registry of live pass-through routes is what decides whether a request is routed upstream or falls through to the auth error, so it has to lose the entry on the reload rather than at the next process restart.""" - from litellm.proxy.pass_through_endpoints.pass_through_endpoints import InitPassThroughEndpointHelpers - from litellm.proxy.proxy_server import ProxyConfig + from litellm.proxy.pass_through_endpoints.pass_through_endpoints import ( + InitPassThroughEndpointHelpers, + _registered_pass_through_routes, + ) + from litellm.proxy.proxy_server import ProxyConfig, app path: Final = f"/v1/deleted-{uuid.uuid4().hex[:8]}" db_endpoint: Final = {"id": "db-1", "path": path, "target": "https://example.com/post"} + prior_routes: Final = list(app.routes) + prior_registry: Final = dict(_registered_pass_through_routes) def live_routes() -> set[str]: return {route for route in InitPassThroughEndpointHelpers.get_all_registered_pass_through_routes() if path in route} settings: Final = patch("litellm.proxy.proxy_server.general_settings", {}) # test-quality-ok: the method reads this module global; no injection seam yaml_endpoints: Final = patch("litellm.proxy.proxy_server.config_passthrough_endpoints", None) # test-quality-ok: module global holding the YAML endpoints; this case has none - with settings, yaml_endpoints: - pc = ProxyConfig() - await pc._update_general_settings(db_general_settings={"pass_through_endpoints": [db_endpoint]}) - assert live_routes(), "the stored endpoint should be serving before the row is deleted" + try: + with settings, yaml_endpoints: + pc = ProxyConfig() + await pc._update_general_settings(db_general_settings={"pass_through_endpoints": [db_endpoint]}) + assert live_routes(), "the stored endpoint should be serving before the row is deleted" - await pc._update_general_settings(db_general_settings={}) + await pc._update_general_settings(db_general_settings={}) - assert live_routes() == set() + assert live_routes() == set() + finally: + app.routes[:] = prior_routes + _registered_pass_through_routes.clear() + _registered_pass_through_routes.update(prior_registry) @pytest.mark.asyncio @@ -7564,15 +7574,18 @@ async def test_a_stored_pass_through_row_never_disturbs_the_config_declared_rout serving untouched. The stored entry never gets a route of its own.""" from litellm.proxy.pass_through_endpoints.pass_through_endpoints import ( InitPassThroughEndpointHelpers, + _registered_pass_through_routes, initialize_pass_through_endpoints, ) - from litellm.proxy.proxy_server import ProxyConfig + from litellm.proxy.proxy_server import ProxyConfig, app marker: Final = uuid.uuid4().hex[:8] config_path: Final = f"/v1/kept-{marker}" db_path: Final = f"/v1/ignored-{marker}" config_endpoint: Final = {"id": f"cfg-{marker}", "path": config_path, "target": "https://example.com/post"} db_endpoint: Final = {"id": f"db-{marker}", "path": db_path, "target": "https://example.com/post"} + prior_routes: Final = list(app.routes) + prior_registry: Final = dict(_registered_pass_through_routes) def live_paths() -> set[str]: registered: Final = InitPassThroughEndpointHelpers.get_all_registered_pass_through_routes() @@ -7580,17 +7593,22 @@ async def test_a_stored_pass_through_row_never_disturbs_the_config_declared_rout settings: Final = patch("litellm.proxy.proxy_server.general_settings", {"pass_through_endpoints": [config_endpoint]}) # test-quality-ok: the method reads this module global; no injection seam yaml_endpoints: Final = patch("litellm.proxy.proxy_server.config_passthrough_endpoints", [config_endpoint]) # test-quality-ok: module global holding the YAML endpoints the reload merges in - with settings, yaml_endpoints: - await initialize_pass_through_endpoints(pass_through_endpoints=[config_endpoint]) - assert live_paths() == {config_path} + try: + with settings, yaml_endpoints: + await initialize_pass_through_endpoints(pass_through_endpoints=[config_endpoint]) + assert live_paths() == {config_path} - pc = ProxyConfig() - await pc._update_general_settings(db_general_settings={"pass_through_endpoints": [db_endpoint]}) - assert live_paths() == {config_path} + pc = ProxyConfig() + await pc._update_general_settings(db_general_settings={"pass_through_endpoints": [db_endpoint]}) + assert live_paths() == {config_path} - await pc._update_general_settings(db_general_settings={}) + await pc._update_general_settings(db_general_settings={}) - assert live_paths() == {config_path} + assert live_paths() == {config_path} + finally: + app.routes[:] = prior_routes + _registered_pass_through_routes.clear() + _registered_pass_through_routes.update(prior_registry) def _fill_user_api_key_cache(cache: DualCache, count: int) -> None: From 0743c97bef983d54d7881254fb7c0d5292b9b743 Mon Sep 17 00:00:00 2001 From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Date: Fri, 18 Sep 2026 19:44:18 +0000 Subject: [PATCH 8/9] chore(deps): bump anyio to 4.14.2 in uv.lock Addresses GHSA-5p39-cfhj-2xmp and GHSA-82r6-8w77-94w6 flagged by the osv-scan job. Hand-edited the single anyio block so the lockfile stays byte-stable for CI's older uv version; dependency set is unchanged. Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- uv.lock | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/uv.lock b/uv.lock index a5e60c68515..e30274c1f6a 100644 --- a/uv.lock +++ b/uv.lock @@ -315,16 +315,16 @@ vertex = [ [[package]] name = "anyio" -version = "4.13.0" +version = "4.14.2" source = { registry = "https://pypi.org/simple" } dependencies = [ { name = "exceptiongroup", marker = "python_full_version < '3.11'" }, { name = "idna" }, { name = "typing-extensions", marker = "python_full_version < '3.13'" }, ] -sdist = { url = "https://files.pythonhosted.org/packages/19/14/2c5dd9f512b66549ae92767a9c7b330ae88e1932ca57876909410251fe13/anyio-4.13.0.tar.gz", hash = "sha256:334b70e641fd2221c1505b3890c69882fe4a2df910cba14d97019b90b24439dc", size = 231622, upload-time = "2026-03-24T12:59:09.671Z" } +sdist = { url = "https://files.pythonhosted.org/packages/61/cc/a381afa6efea9f496eff839d4a6a1aed3bfafc7b3ab4b0d1b243a12573dd/anyio-4.14.2.tar.gz", hash = "sha256:cfa139f3ed1a23ee8f88a145ddb5ac7605b8bbfd8592baacd7ce3d8bb4313c7f", size = 260176, upload-time = "2026-07-12T20:29:07.082Z" } wheels = [ - { url = "https://files.pythonhosted.org/packages/da/42/e921fccf5015463e32a3cf6ee7f980a6ed0f395ceeaa45060b61d86486c2/anyio-4.13.0-py3-none-any.whl", hash = "sha256:08b310f9e24a9594186fd75b4f73f4a4152069e3853f1ed8bfbf58369f4ad708", size = 114353, upload-time = "2026-03-24T12:59:08.246Z" }, + { url = "https://files.pythonhosted.org/packages/da/35/f2287558c17e29fafc8ef3daf819bb9834061cfa43bff8014f7df7f63bdc/anyio-4.14.2-py3-none-any.whl", hash = "sha256:9f505dda5ac9f0c8309b5e8bd445a8c2bf7246f3ce950121e45ea15bc41d1494", size = 125813, upload-time = "2026-07-12T20:29:05.763Z" }, ] [[package]] From 04193745a65cc18a36820be8ef1d901d6cbe19db Mon Sep 17 00:00:00 2001 From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Date: Fri, 18 Sep 2026 20:10:20 +0000 Subject: [PATCH 9/9] fix(docs-test): restore line-anchored table regex in router settings check A merge on this branch dropped the ^ anchor and MULTILINE flag from the doc_key_pattern, so the unanchored match started swallowing key names into captured fields whenever a row's description cell itself contains a pipe (Literal unions and similar). The check then reported 27 long-documented keys as undocumented. Restores the exact pattern used on main, verified against a live litellm-docs checkout: all 58 Router init params resolve as documented. Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- tests/documentation_tests/test_router_settings.py | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/tests/documentation_tests/test_router_settings.py b/tests/documentation_tests/test_router_settings.py index 75032f80dfa..7e3d0c07459 100644 --- a/tests/documentation_tests/test_router_settings.py +++ b/tests/documentation_tests/test_router_settings.py @@ -51,9 +51,7 @@ try: if general_settings_section: # Extract the table rows, which contain the documented keys table_content = general_settings_section.group(1) - doc_key_pattern = re.compile( - r"\|\s*([^\|]+?)\s*\|" - ) # Capture the key from each row of the table + doc_key_pattern = re.compile(r"^\|\s*([^\|]+?)\s*\|", re.MULTILINE) documented_keys.update(doc_key_pattern.findall(table_content)) except Exception as e: raise Exception(