From 5a0a8ffafe6b8af4c73cc28b5cbb17376a835350 Mon Sep 17 00:00:00 2001 From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Date: Wed, 19 Aug 2026 13:17:15 +0000 Subject: [PATCH 1/2] fix(model_prices): correct gemini and deepseek pricing and add deprecation dates Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- ...odel_prices_and_context_window_backup.json | 54 ++++++++++--------- model_prices_and_context_window.json | 54 ++++++++++--------- 2 files changed, 60 insertions(+), 48 deletions(-) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 07f9027313b..1a1652451a1 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -10004,6 +10004,7 @@ "supports_vision": true }, "babbage-002": { + "deprecation_date": "2026-09-28", "input_cost_per_token": 4e-07, "litellm_provider": "text-completion-openai", "max_input_tokens": 16384, @@ -15042,6 +15043,7 @@ "mode": "search" }, "davinci-002": { + "deprecation_date": "2026-09-28", "input_cost_per_token": 2e-06, "litellm_provider": "text-completion-openai", "max_input_tokens": 16384, @@ -20562,8 +20564,8 @@ "web_search_billing_unit": "per_query" }, "gemini/gemini-3.1-flash-image": { - "input_cost_per_token": 2.5e-07, - "input_cost_per_token_batches": 1.25e-07, + "input_cost_per_token": 5e-07, + "input_cost_per_token_batches": 2.5e-07, "litellm_provider": "gemini", "max_input_tokens": 65536, "max_output_tokens": 32768, @@ -20571,8 +20573,8 @@ "mode": "image_generation", "output_cost_per_image": 0.045, "output_cost_per_image_token": 6e-05, - "output_cost_per_token": 1.5e-06, - "output_cost_per_token_batches": 7.5e-07, + "output_cost_per_token": 3e-06, + "output_cost_per_token_batches": 1.5e-06, "rpm": 1000, "tpm": 4000000, "source": "https://ai.google.dev/gemini-api/docs/pricing#gemini-3.1-flash-image", @@ -20605,8 +20607,8 @@ }, "gemini/gemini-3.1-flash-image-preview": { "deprecation_date": "2026-06-25", - "input_cost_per_token": 2.5e-07, - "input_cost_per_token_batches": 1.25e-07, + "input_cost_per_token": 5e-07, + "input_cost_per_token_batches": 2.5e-07, "litellm_provider": "gemini", "max_input_tokens": 65536, "max_output_tokens": 32768, @@ -20614,8 +20616,8 @@ "mode": "image_generation", "output_cost_per_image": 0.045, "output_cost_per_image_token": 6e-05, - "output_cost_per_token": 1.5e-06, - "output_cost_per_token_batches": 7.5e-07, + "output_cost_per_token": 3e-06, + "output_cost_per_token_batches": 1.5e-06, "rpm": 1000, "tpm": 4000000, "source": "https://ai.google.dev/gemini-api/docs/pricing#gemini-3.1-flash-image-preview", @@ -23245,6 +23247,7 @@ "supports_tool_choice": true }, "gpt-3.5-turbo-instruct": { + "deprecation_date": "2026-09-28", "input_cost_per_token": 1.5e-06, "litellm_provider": "text-completion-openai", "max_input_tokens": 8192, @@ -36161,6 +36164,7 @@ "source": "https://cloud.google.com/vertex-ai/generative-ai/pricing" }, "text-moderation-007": { + "deprecation_date": "2025-10-27", "input_cost_per_token": 0.0, "litellm_provider": "openai", "max_input_tokens": 32768, @@ -36170,6 +36174,7 @@ "output_cost_per_token": 0.0 }, "text-moderation-latest": { + "deprecation_date": "2025-10-27", "input_cost_per_token": 0.0, "litellm_provider": "openai", "max_input_tokens": 32768, @@ -36179,6 +36184,7 @@ "output_cost_per_token": 0.0 }, "text-moderation-stable": { + "deprecation_date": "2025-10-27", "input_cost_per_token": 0.0, "litellm_provider": "openai", "max_input_tokens": 32768, @@ -47975,15 +47981,15 @@ }, "deepseek-v4-flash": { "cache_creation_input_token_cost": 0.0, - "cache_read_input_token_cost": 2.8e-09, - "input_cost_per_token": 1.4e-07, - "input_cost_per_token_cache_hit": 2.8e-09, + "cache_read_input_token_cost": 1.4e-08, + "input_cost_per_token": 4.4e-07, + "input_cost_per_token_cache_hit": 1.4e-08, "litellm_provider": "deepseek", "max_input_tokens": 1000000, "max_output_tokens": 393216, "max_tokens": 393216, "mode": "chat", - "output_cost_per_token": 2.8e-07, + "output_cost_per_token": 1.32e-06, "source": "https://api-docs.deepseek.com/quick_start/pricing", "supported_endpoints": [ "/v1/chat/completions" @@ -48001,15 +48007,15 @@ }, "deepseek-v4-pro": { "cache_creation_input_token_cost": 0.0, - "cache_read_input_token_cost": 3.625e-09, - "input_cost_per_token": 4.35e-07, - "input_cost_per_token_cache_hit": 3.625e-09, + "cache_read_input_token_cost": 4.4e-08, + "input_cost_per_token": 1.32e-06, + "input_cost_per_token_cache_hit": 4.4e-08, "litellm_provider": "deepseek", "max_input_tokens": 1000000, "max_output_tokens": 393216, "max_tokens": 393216, "mode": "chat", - "output_cost_per_token": 8.7e-07, + "output_cost_per_token": 3.96e-06, "source": "https://api-docs.deepseek.com/quick_start/pricing", "supported_endpoints": [ "/v1/chat/completions" @@ -48027,15 +48033,15 @@ }, "deepseek/deepseek-v4-flash": { "cache_creation_input_token_cost": 0.0, - "cache_read_input_token_cost": 2.8e-09, - "input_cost_per_token": 1.4e-07, - "input_cost_per_token_cache_hit": 2.8e-09, + "cache_read_input_token_cost": 1.4e-08, + "input_cost_per_token": 4.4e-07, + "input_cost_per_token_cache_hit": 1.4e-08, "litellm_provider": "deepseek", "max_input_tokens": 1000000, "max_output_tokens": 393216, "max_tokens": 393216, "mode": "chat", - "output_cost_per_token": 2.8e-07, + "output_cost_per_token": 1.32e-06, "source": "https://api-docs.deepseek.com/quick_start/pricing", "supported_endpoints": [ "/v1/chat/completions" @@ -48053,15 +48059,15 @@ }, "deepseek/deepseek-v4-pro": { "cache_creation_input_token_cost": 0.0, - "cache_read_input_token_cost": 3.625e-09, - "input_cost_per_token": 4.35e-07, - "input_cost_per_token_cache_hit": 3.625e-09, + "cache_read_input_token_cost": 4.4e-08, + "input_cost_per_token": 1.32e-06, + "input_cost_per_token_cache_hit": 4.4e-08, "litellm_provider": "deepseek", "max_input_tokens": 1000000, "max_output_tokens": 393216, "max_tokens": 393216, "mode": "chat", - "output_cost_per_token": 8.7e-07, + "output_cost_per_token": 3.96e-06, "source": "https://api-docs.deepseek.com/quick_start/pricing", "supported_endpoints": [ "/v1/chat/completions" diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 07f9027313b..1a1652451a1 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -10004,6 +10004,7 @@ "supports_vision": true }, "babbage-002": { + "deprecation_date": "2026-09-28", "input_cost_per_token": 4e-07, "litellm_provider": "text-completion-openai", "max_input_tokens": 16384, @@ -15042,6 +15043,7 @@ "mode": "search" }, "davinci-002": { + "deprecation_date": "2026-09-28", "input_cost_per_token": 2e-06, "litellm_provider": "text-completion-openai", "max_input_tokens": 16384, @@ -20562,8 +20564,8 @@ "web_search_billing_unit": "per_query" }, "gemini/gemini-3.1-flash-image": { - "input_cost_per_token": 2.5e-07, - "input_cost_per_token_batches": 1.25e-07, + "input_cost_per_token": 5e-07, + "input_cost_per_token_batches": 2.5e-07, "litellm_provider": "gemini", "max_input_tokens": 65536, "max_output_tokens": 32768, @@ -20571,8 +20573,8 @@ "mode": "image_generation", "output_cost_per_image": 0.045, "output_cost_per_image_token": 6e-05, - "output_cost_per_token": 1.5e-06, - "output_cost_per_token_batches": 7.5e-07, + "output_cost_per_token": 3e-06, + "output_cost_per_token_batches": 1.5e-06, "rpm": 1000, "tpm": 4000000, "source": "https://ai.google.dev/gemini-api/docs/pricing#gemini-3.1-flash-image", @@ -20605,8 +20607,8 @@ }, "gemini/gemini-3.1-flash-image-preview": { "deprecation_date": "2026-06-25", - "input_cost_per_token": 2.5e-07, - "input_cost_per_token_batches": 1.25e-07, + "input_cost_per_token": 5e-07, + "input_cost_per_token_batches": 2.5e-07, "litellm_provider": "gemini", "max_input_tokens": 65536, "max_output_tokens": 32768, @@ -20614,8 +20616,8 @@ "mode": "image_generation", "output_cost_per_image": 0.045, "output_cost_per_image_token": 6e-05, - "output_cost_per_token": 1.5e-06, - "output_cost_per_token_batches": 7.5e-07, + "output_cost_per_token": 3e-06, + "output_cost_per_token_batches": 1.5e-06, "rpm": 1000, "tpm": 4000000, "source": "https://ai.google.dev/gemini-api/docs/pricing#gemini-3.1-flash-image-preview", @@ -23245,6 +23247,7 @@ "supports_tool_choice": true }, "gpt-3.5-turbo-instruct": { + "deprecation_date": "2026-09-28", "input_cost_per_token": 1.5e-06, "litellm_provider": "text-completion-openai", "max_input_tokens": 8192, @@ -36161,6 +36164,7 @@ "source": "https://cloud.google.com/vertex-ai/generative-ai/pricing" }, "text-moderation-007": { + "deprecation_date": "2025-10-27", "input_cost_per_token": 0.0, "litellm_provider": "openai", "max_input_tokens": 32768, @@ -36170,6 +36174,7 @@ "output_cost_per_token": 0.0 }, "text-moderation-latest": { + "deprecation_date": "2025-10-27", "input_cost_per_token": 0.0, "litellm_provider": "openai", "max_input_tokens": 32768, @@ -36179,6 +36184,7 @@ "output_cost_per_token": 0.0 }, "text-moderation-stable": { + "deprecation_date": "2025-10-27", "input_cost_per_token": 0.0, "litellm_provider": "openai", "max_input_tokens": 32768, @@ -47975,15 +47981,15 @@ }, "deepseek-v4-flash": { "cache_creation_input_token_cost": 0.0, - "cache_read_input_token_cost": 2.8e-09, - "input_cost_per_token": 1.4e-07, - "input_cost_per_token_cache_hit": 2.8e-09, + "cache_read_input_token_cost": 1.4e-08, + "input_cost_per_token": 4.4e-07, + "input_cost_per_token_cache_hit": 1.4e-08, "litellm_provider": "deepseek", "max_input_tokens": 1000000, "max_output_tokens": 393216, "max_tokens": 393216, "mode": "chat", - "output_cost_per_token": 2.8e-07, + "output_cost_per_token": 1.32e-06, "source": "https://api-docs.deepseek.com/quick_start/pricing", "supported_endpoints": [ "/v1/chat/completions" @@ -48001,15 +48007,15 @@ }, "deepseek-v4-pro": { "cache_creation_input_token_cost": 0.0, - "cache_read_input_token_cost": 3.625e-09, - "input_cost_per_token": 4.35e-07, - "input_cost_per_token_cache_hit": 3.625e-09, + "cache_read_input_token_cost": 4.4e-08, + "input_cost_per_token": 1.32e-06, + "input_cost_per_token_cache_hit": 4.4e-08, "litellm_provider": "deepseek", "max_input_tokens": 1000000, "max_output_tokens": 393216, "max_tokens": 393216, "mode": "chat", - "output_cost_per_token": 8.7e-07, + "output_cost_per_token": 3.96e-06, "source": "https://api-docs.deepseek.com/quick_start/pricing", "supported_endpoints": [ "/v1/chat/completions" @@ -48027,15 +48033,15 @@ }, "deepseek/deepseek-v4-flash": { "cache_creation_input_token_cost": 0.0, - "cache_read_input_token_cost": 2.8e-09, - "input_cost_per_token": 1.4e-07, - "input_cost_per_token_cache_hit": 2.8e-09, + "cache_read_input_token_cost": 1.4e-08, + "input_cost_per_token": 4.4e-07, + "input_cost_per_token_cache_hit": 1.4e-08, "litellm_provider": "deepseek", "max_input_tokens": 1000000, "max_output_tokens": 393216, "max_tokens": 393216, "mode": "chat", - "output_cost_per_token": 2.8e-07, + "output_cost_per_token": 1.32e-06, "source": "https://api-docs.deepseek.com/quick_start/pricing", "supported_endpoints": [ "/v1/chat/completions" @@ -48053,15 +48059,15 @@ }, "deepseek/deepseek-v4-pro": { "cache_creation_input_token_cost": 0.0, - "cache_read_input_token_cost": 3.625e-09, - "input_cost_per_token": 4.35e-07, - "input_cost_per_token_cache_hit": 3.625e-09, + "cache_read_input_token_cost": 4.4e-08, + "input_cost_per_token": 1.32e-06, + "input_cost_per_token_cache_hit": 4.4e-08, "litellm_provider": "deepseek", "max_input_tokens": 1000000, "max_output_tokens": 393216, "max_tokens": 393216, "mode": "chat", - "output_cost_per_token": 8.7e-07, + "output_cost_per_token": 3.96e-06, "source": "https://api-docs.deepseek.com/quick_start/pricing", "supported_endpoints": [ "/v1/chat/completions" From dcd8bb3f38561c0030499a0fa074cddee1aed4e8 Mon Sep 17 00:00:00 2001 From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Date: Wed, 19 Aug 2026 13:47:44 +0000 Subject: [PATCH 2/2] test(model_prices): update DeepSeek V4 pricing expectations Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- tests/test_litellm/test_utils.py | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/tests/test_litellm/test_utils.py b/tests/test_litellm/test_utils.py index efccdc4a986..0218f8f6219 100644 --- a/tests/test_litellm/test_utils.py +++ b/tests/test_litellm/test_utils.py @@ -3787,8 +3787,8 @@ def test_deepseek_v4_models_in_cost_map(): configured in model_prices_and_context_window.json. Prices sourced from https://api-docs.deepseek.com/quick_start/pricing: - - deepseek-v4-flash: $0.14/M input, $0.28/M output - - deepseek-v4-pro: $0.435/M input, $0.87/M output (75% discounted active price) + - deepseek-v4-flash: $0.44/M input, $1.32/M output + - deepseek-v4-pro: $1.32/M input, $3.96/M output Closes https://github.com/BerriAI/litellm/issues/26709 """ @@ -3801,8 +3801,8 @@ def test_deepseek_v4_models_in_cost_map(): # --- bare model names --- for key, expected_input, expected_output, expected_cache in [ - ("deepseek-v4-flash", 1.4e-07, 2.8e-07, 2.8e-09), - ("deepseek-v4-pro", 4.35e-07, 8.7e-07, 3.625e-09), + ("deepseek-v4-flash", 4.4e-07, 1.32e-06, 1.4e-08), + ("deepseek-v4-pro", 1.32e-06, 3.96e-06, 4.4e-08), ]: info = model_cost.get(key) assert info is not None, f"{key} missing from model_prices_and_context_window.json" @@ -3817,8 +3817,8 @@ def test_deepseek_v4_models_in_cost_map(): # --- provider-prefixed names --- for key, expected_input, expected_output, expected_cache in [ - ("deepseek/deepseek-v4-flash", 1.4e-07, 2.8e-07, 2.8e-09), - ("deepseek/deepseek-v4-pro", 4.35e-07, 8.7e-07, 3.625e-09), + ("deepseek/deepseek-v4-flash", 4.4e-07, 1.32e-06, 1.4e-08), + ("deepseek/deepseek-v4-pro", 1.32e-06, 3.96e-06, 4.4e-08), ]: info = model_cost.get(key) assert info is not None, f"{key} missing from model_prices_and_context_window.json" @@ -3845,8 +3845,8 @@ def test_deepseek_v4_models_in_backup_cost_map(): # --- bare model names --- for key, expected_input, expected_output, expected_cache in [ - ("deepseek-v4-flash", 1.4e-07, 2.8e-07, 2.8e-09), - ("deepseek-v4-pro", 4.35e-07, 8.7e-07, 3.625e-09), + ("deepseek-v4-flash", 4.4e-07, 1.32e-06, 1.4e-08), + ("deepseek-v4-pro", 1.32e-06, 3.96e-06, 4.4e-08), ]: info = model_cost.get(key) assert info is not None, f"{key} missing from backup JSON" @@ -3859,8 +3859,8 @@ def test_deepseek_v4_models_in_backup_cost_map(): # --- provider-prefixed names --- for key, expected_input, expected_output, expected_cache in [ - ("deepseek/deepseek-v4-flash", 1.4e-07, 2.8e-07, 2.8e-09), - ("deepseek/deepseek-v4-pro", 4.35e-07, 8.7e-07, 3.625e-09), + ("deepseek/deepseek-v4-flash", 4.4e-07, 1.32e-06, 1.4e-08), + ("deepseek/deepseek-v4-pro", 1.32e-06, 3.96e-06, 4.4e-08), ]: info = model_cost.get(key) assert info is not None, f"{key} missing from backup JSON"