From 26f14204d3e0ecbc4bcb85302f4a2b5edd6b801c Mon Sep 17 00:00:00 2001 From: respectus Date: Fri, 11 Sep 2026 21:05:01 -0700 Subject: [PATCH] Limit GitGot entries to open-weight models Narrows this to the seven models whose licences permit third-party serving: gpt-oss (Apache-2.0), DeepSeek (MIT), Kimi (modified MIT) and Llama 3.3. The proprietary entries are dropped, and the default model placeholder moves to gitgot/openai/gpt-oss-120b. 10 tests still pass. --- ...odel_prices_and_context_window_backup.json | 105 ++++++++++++++++++ .../provider_create_fields.json | 2 +- model_prices_and_context_window.json | 105 ++++++++++++++++++ .../llms/openai_like/test_gitgot_provider.py | 12 +- .../src/components/provider_info_helpers.tsx | 2 +- 5 files changed, 218 insertions(+), 8 deletions(-) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 84920f9e4e1..aa8fe3f1249 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -65419,5 +65419,110 @@ "supports_tool_choice": true, "supports_vision": false, "supports_reasoning": false + }, + "gitgot/deepseek-ai/DeepSeek-V4-Flash": { + "max_tokens": 32768, + "max_input_tokens": 163840, + "max_output_tokens": 32768, + "input_cost_per_token": 0.0000000882, + "output_cost_per_token": 0.0000001764, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false + }, + "gitgot/deepseek-ai/DeepSeek-V4-Flash-0731": { + "max_tokens": 32768, + "max_input_tokens": 163840, + "max_output_tokens": 32768, + "input_cost_per_token": 0.0000000588, + "output_cost_per_token": 0.0000001764, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false + }, + "gitgot/deepseek-ai/DeepSeek-V4-Pro": { + "max_tokens": 32768, + "max_input_tokens": 163840, + "max_output_tokens": 32768, + "input_cost_per_token": 0.000001274, + "output_cost_per_token": 0.000002548, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": true + }, + "gitgot/meta-llama/Llama-3.3-70B-Instruct": { + "max_tokens": 32768, + "max_input_tokens": 131072, + "max_output_tokens": 32768, + "input_cost_per_token": 0.0000001323, + "output_cost_per_token": 0.000000392, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false + }, + "gitgot/moonshotai/Kimi-K2.6": { + "max_tokens": 64000, + "max_input_tokens": 262144, + "max_output_tokens": 64000, + "input_cost_per_token": 0.000000735, + "output_cost_per_token": 0.00000343, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false + }, + "gitgot/moonshotai/Kimi-K2.7-Code": { + "max_tokens": 64000, + "max_input_tokens": 262144, + "max_output_tokens": 64000, + "input_cost_per_token": 0.0000006664, + "output_cost_per_token": 0.000003332, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false + }, + "gitgot/openai/gpt-oss-120b": { + "max_tokens": 32768, + "max_input_tokens": 131072, + "max_output_tokens": 32768, + "input_cost_per_token": 0.00000003626, + "output_cost_per_token": 0.0000001666, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false } } diff --git a/litellm/proxy/public_endpoints/provider_create_fields.json b/litellm/proxy/public_endpoints/provider_create_fields.json index e9d0d667ad3..07a4219858c 100644 --- a/litellm/proxy/public_endpoints/provider_create_fields.json +++ b/litellm/proxy/public_endpoints/provider_create_fields.json @@ -2877,7 +2877,7 @@ "default_value": null } ], - "default_model_placeholder": "gitgot/openai/gpt-5-nano" + "default_model_placeholder": "gitgot/openai/gpt-oss-120b" }, { "provider": "SCX_AI", diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 84920f9e4e1..aa8fe3f1249 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -65419,5 +65419,110 @@ "supports_tool_choice": true, "supports_vision": false, "supports_reasoning": false + }, + "gitgot/deepseek-ai/DeepSeek-V4-Flash": { + "max_tokens": 32768, + "max_input_tokens": 163840, + "max_output_tokens": 32768, + "input_cost_per_token": 0.0000000882, + "output_cost_per_token": 0.0000001764, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false + }, + "gitgot/deepseek-ai/DeepSeek-V4-Flash-0731": { + "max_tokens": 32768, + "max_input_tokens": 163840, + "max_output_tokens": 32768, + "input_cost_per_token": 0.0000000588, + "output_cost_per_token": 0.0000001764, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false + }, + "gitgot/deepseek-ai/DeepSeek-V4-Pro": { + "max_tokens": 32768, + "max_input_tokens": 163840, + "max_output_tokens": 32768, + "input_cost_per_token": 0.000001274, + "output_cost_per_token": 0.000002548, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": true + }, + "gitgot/meta-llama/Llama-3.3-70B-Instruct": { + "max_tokens": 32768, + "max_input_tokens": 131072, + "max_output_tokens": 32768, + "input_cost_per_token": 0.0000001323, + "output_cost_per_token": 0.000000392, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false + }, + "gitgot/moonshotai/Kimi-K2.6": { + "max_tokens": 64000, + "max_input_tokens": 262144, + "max_output_tokens": 64000, + "input_cost_per_token": 0.000000735, + "output_cost_per_token": 0.00000343, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false + }, + "gitgot/moonshotai/Kimi-K2.7-Code": { + "max_tokens": 64000, + "max_input_tokens": 262144, + "max_output_tokens": 64000, + "input_cost_per_token": 0.0000006664, + "output_cost_per_token": 0.000003332, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false + }, + "gitgot/openai/gpt-oss-120b": { + "max_tokens": 32768, + "max_input_tokens": 131072, + "max_output_tokens": 32768, + "input_cost_per_token": 0.00000003626, + "output_cost_per_token": 0.0000001666, + "litellm_provider": "gitgot", + "mode": "chat", + "source": "https://inference.gitgot.ai/v1/pricing", + "supports_function_calling": true, + "supports_response_schema": true, + "supports_tool_choice": true, + "supports_vision": false, + "supports_reasoning": false } } diff --git a/tests/test_litellm/llms/openai_like/test_gitgot_provider.py b/tests/test_litellm/llms/openai_like/test_gitgot_provider.py index 25aae9de538..16ccb802f7f 100644 --- a/tests/test_litellm/llms/openai_like/test_gitgot_provider.py +++ b/tests/test_litellm/llms/openai_like/test_gitgot_provider.py @@ -38,7 +38,7 @@ class TestGitGotProviderConfig: from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider model, provider, api_key, api_base = get_llm_provider( - model="gitgot/openai/gpt-5-nano", + model="gitgot/openai/gpt-oss-120b", custom_llm_provider=None, api_base=None, api_key=None, @@ -46,7 +46,7 @@ class TestGitGotProviderConfig: # GitGot model ids carry a vendor prefix, so only the leading "gitgot/" # is stripped and the rest of the path is preserved. - assert model == "openai/gpt-5-nano" + assert model == "openai/gpt-oss-120b" assert provider == "gitgot" assert api_base == "https://inference.gitgot.ai/v1" @@ -54,7 +54,7 @@ class TestGitGotProviderConfig: from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider model, provider, api_key, api_base = get_llm_provider( - model="gitgot/openai/gpt-5-nano", + model="gitgot/openai/gpt-oss-120b", custom_llm_provider=None, api_base="https://custom.gitgot.ai/v1", api_key="gg-test", @@ -67,7 +67,7 @@ class TestGitGotProviderConfig: def test_gitgot_model_prices_registered(self, local_model_cost_map): from litellm import model_cost - entry = model_cost["gitgot/openai/gpt-5-nano"] + entry = model_cost["gitgot/openai/gpt-oss-120b"] assert entry["litellm_provider"] == "gitgot" assert entry["mode"] == "chat" # Costs in this file are per token, not per million. A per-million @@ -91,7 +91,7 @@ class TestGitGotProviderConfig: ) ], created=1234567890, - model="gitgot/openai/gpt-5-nano", + model="gitgot/openai/gpt-oss-120b", object="chat.completion", usage=Usage( prompt_tokens=1_000_000, @@ -112,7 +112,7 @@ class TestGitGotProviderConfig: { "model_name": "gitgot-chat", "litellm_params": { - "model": "gitgot/openai/gpt-5-nano", + "model": "gitgot/openai/gpt-oss-120b", "api_key": "test-key", }, } diff --git a/ui/litellm-dashboard/src/components/provider_info_helpers.tsx b/ui/litellm-dashboard/src/components/provider_info_helpers.tsx index 7eb5d886c7d..61b1969d2b1 100644 --- a/ui/litellm-dashboard/src/components/provider_info_helpers.tsx +++ b/ui/litellm-dashboard/src/components/provider_info_helpers.tsx @@ -444,7 +444,7 @@ const providerPlaceholderMap: Partial> = { [Providers.RunwayML]: "runwayml/gen4_turbo", [Providers.SageMaker]: "sagemaker/jumpstart-dft-meta-textgeneration-llama-2-7b", [Providers.SCX_AI]: "scx-ai/GLM-5.2", - [Providers.GitGot]: "gitgot/openai/gpt-5-nano", + [Providers.GitGot]: "gitgot/openai/gpt-oss-120b", [Providers.Snowflake]: "snowflake/mistral-7b", [Providers.Vertex_AI]: "gemini-pro", [Providers.VolcEngine]: "volcengine/",