From 4b498d1bba768bd81e4a434a14ea7ae2870dd01f Mon Sep 17 00:00:00 2001
From: "devin-ai-integration[bot]"
 <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Date: Wed, 11 Nov 2026 01:08:48 +0000
Subject: [PATCH] chore(databricks): add claude-fable-5-2, date glm-5-2
 retirement and fix llama-3-1-8b context from the supported-models page
 (#51076)

Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---
 ...odel_prices_and_context_window_backup.json | 59 ++++++++++++++++++-
 model_prices_and_context_window.json          | 59 ++++++++++++++++++-
 2 files changed, 112 insertions(+), 6 deletions(-)

diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json
index 0ee57858b17..092987ce69e 100644
--- a/litellm/model_prices_and_context_window_backup.json
+++ b/litellm/model_prices_and_context_window_backup.json
@@ -21878,7 +21878,7 @@
         "thinking_always_on": true
     },
     "databricks/claude-fable-latest": {
-        "alias_of": "databricks/databricks-claude-fable-5-1"
+        "alias_of": "databricks/databricks-claude-fable-5-2"
     },
     "databricks/databricks-claude-haiku-4-5": {
         "cache_creation_input_token_cost": 1.24999e-06,
@@ -22804,7 +22804,10 @@
         "input_cost_per_token_priority": 2.45e-06,
         "output_cost_per_token_priority": 7.7e-06,
         "cache_read_input_token_cost_priority": 4.55e-07,
-        "cache_creation_input_token_cost_priority": 2.45e-06
+        "cache_creation_input_token_cost_priority": 2.45e-06,
+        "deprecation_date": "2027-02-04",
+        "deprecation_source": "https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/supported-models",
+        "successor_model": "databricks/databricks-glm-5-3"
     },
     "databricks/databricks-glm-5-3": {
         "cache_creation_input_token_cost": 1.4e-06,
@@ -23471,7 +23474,7 @@
         "input_cost_per_token": 1.5001e-07,
         "input_dbu_cost_per_token": 2.1429999999999996e-06,
         "litellm_provider": "databricks",
-        "max_input_tokens": 200000,
+        "max_input_tokens": 128000,
         "max_output_tokens": 128000,
         "metadata": {
             "notes": "Input/output cost per token is dbu cost * $0.070. Number provided for reference, '*_dbu_cost_per_token' used in actual calculation."
@@ -99977,5 +99980,55 @@
         "supported_output_modalities": [
             "video"
         ]
+    },
+    "databricks/databricks-claude-fable-5-2": {
+        "cache_creation_input_token_cost": 8e-06,
+        "cache_read_input_token_cost": 8e-06,
+        "input_cost_per_token": 8e-06,
+        "litellm_provider": "databricks",
+        "max_input_tokens": 1000000,
+        "max_output_tokens": 128000,
+        "metadata": {"notes":"Claude Fable 5.2 on Databricks is available only through Unity Gateway with the model service name system.ai.claude-fable-5-2, and Databricks hosts it through Amazon Bedrock. The Databricks pricing page lists no row for it yet, so input and output are inherited from the catalog Bedrock row anthropic.claude-fable-5-2, no '*_dbu_cost_per_token' fields are published, and cache tokens bill at the input rate until Databricks publishes cache DBU rates."},
+        "mode": "chat",
+        "output_cost_per_token": 4e-05,
+        "prompt_cache_min_tokens": 512.0,
+        "source": "https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/supported-models",
+        "supports_adaptive_thinking": true,
+        "supports_anthropic_thinking_payload": true,
+        "supports_assistant_prefill": false,
+        "supports_forced_tool_use": false,
+        "supports_function_calling": true,
+        "supports_mid_conversation_system": true,
+        "supports_prompt_caching": true,
+        "supports_reasoning": true,
+        "supports_sampling_params": false,
+        "supports_tool_choice": true,
+        "supports_vision": true,
+        "thinking_always_on": true
+    },
+    "databricks/system.ai.claude-fable-5-2": {
+        "cache_creation_input_token_cost": 8e-06,
+        "cache_read_input_token_cost": 8e-06,
+        "input_cost_per_token": 8e-06,
+        "litellm_provider": "databricks",
+        "max_input_tokens": 1000000,
+        "max_output_tokens": 128000,
+        "metadata": {"notes":"Claude Fable 5.2 on Databricks is available only through Unity Gateway with the model service name system.ai.claude-fable-5-2, and Databricks hosts it through Amazon Bedrock. The Databricks pricing page lists no row for it yet, so input and output are inherited from the catalog Bedrock row anthropic.claude-fable-5-2, no '*_dbu_cost_per_token' fields are published, and cache tokens bill at the input rate until Databricks publishes cache DBU rates."},
+        "mode": "chat",
+        "output_cost_per_token": 4e-05,
+        "prompt_cache_min_tokens": 512.0,
+        "source": "https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/supported-models",
+        "supports_adaptive_thinking": true,
+        "supports_anthropic_thinking_payload": true,
+        "supports_assistant_prefill": false,
+        "supports_forced_tool_use": false,
+        "supports_function_calling": true,
+        "supports_mid_conversation_system": true,
+        "supports_prompt_caching": true,
+        "supports_reasoning": true,
+        "supports_sampling_params": false,
+        "supports_tool_choice": true,
+        "supports_vision": true,
+        "thinking_always_on": true
     }
 }
diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json
index 0ee57858b17..092987ce69e 100644
--- a/model_prices_and_context_window.json
+++ b/model_prices_and_context_window.json
@@ -21878,7 +21878,7 @@
         "thinking_always_on": true
     },
     "databricks/claude-fable-latest": {
-        "alias_of": "databricks/databricks-claude-fable-5-1"
+        "alias_of": "databricks/databricks-claude-fable-5-2"
     },
     "databricks/databricks-claude-haiku-4-5": {
         "cache_creation_input_token_cost": 1.24999e-06,
@@ -22804,7 +22804,10 @@
         "input_cost_per_token_priority": 2.45e-06,
         "output_cost_per_token_priority": 7.7e-06,
         "cache_read_input_token_cost_priority": 4.55e-07,
-        "cache_creation_input_token_cost_priority": 2.45e-06
+        "cache_creation_input_token_cost_priority": 2.45e-06,
+        "deprecation_date": "2027-02-04",
+        "deprecation_source": "https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/supported-models",
+        "successor_model": "databricks/databricks-glm-5-3"
     },
     "databricks/databricks-glm-5-3": {
         "cache_creation_input_token_cost": 1.4e-06,
@@ -23471,7 +23474,7 @@
         "input_cost_per_token": 1.5001e-07,
         "input_dbu_cost_per_token": 2.1429999999999996e-06,
         "litellm_provider": "databricks",
-        "max_input_tokens": 200000,
+        "max_input_tokens": 128000,
         "max_output_tokens": 128000,
         "metadata": {
             "notes": "Input/output cost per token is dbu cost * $0.070. Number provided for reference, '*_dbu_cost_per_token' used in actual calculation."
@@ -99977,5 +99980,55 @@
         "supported_output_modalities": [
             "video"
         ]
+    },
+    "databricks/databricks-claude-fable-5-2": {
+        "cache_creation_input_token_cost": 8e-06,
+        "cache_read_input_token_cost": 8e-06,
+        "input_cost_per_token": 8e-06,
+        "litellm_provider": "databricks",
+        "max_input_tokens": 1000000,
+        "max_output_tokens": 128000,
+        "metadata": {"notes":"Claude Fable 5.2 on Databricks is available only through Unity Gateway with the model service name system.ai.claude-fable-5-2, and Databricks hosts it through Amazon Bedrock. The Databricks pricing page lists no row for it yet, so input and output are inherited from the catalog Bedrock row anthropic.claude-fable-5-2, no '*_dbu_cost_per_token' fields are published, and cache tokens bill at the input rate until Databricks publishes cache DBU rates."},
+        "mode": "chat",
+        "output_cost_per_token": 4e-05,
+        "prompt_cache_min_tokens": 512.0,
+        "source": "https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/supported-models",
+        "supports_adaptive_thinking": true,
+        "supports_anthropic_thinking_payload": true,
+        "supports_assistant_prefill": false,
+        "supports_forced_tool_use": false,
+        "supports_function_calling": true,
+        "supports_mid_conversation_system": true,
+        "supports_prompt_caching": true,
+        "supports_reasoning": true,
+        "supports_sampling_params": false,
+        "supports_tool_choice": true,
+        "supports_vision": true,
+        "thinking_always_on": true
+    },
+    "databricks/system.ai.claude-fable-5-2": {
+        "cache_creation_input_token_cost": 8e-06,
+        "cache_read_input_token_cost": 8e-06,
+        "input_cost_per_token": 8e-06,
+        "litellm_provider": "databricks",
+        "max_input_tokens": 1000000,
+        "max_output_tokens": 128000,
+        "metadata": {"notes":"Claude Fable 5.2 on Databricks is available only through Unity Gateway with the model service name system.ai.claude-fable-5-2, and Databricks hosts it through Amazon Bedrock. The Databricks pricing page lists no row for it yet, so input and output are inherited from the catalog Bedrock row anthropic.claude-fable-5-2, no '*_dbu_cost_per_token' fields are published, and cache tokens bill at the input rate until Databricks publishes cache DBU rates."},
+        "mode": "chat",
+        "output_cost_per_token": 4e-05,
+        "prompt_cache_min_tokens": 512.0,
+        "source": "https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/supported-models",
+        "supports_adaptive_thinking": true,
+        "supports_anthropic_thinking_payload": true,
+        "supports_assistant_prefill": false,
+        "supports_forced_tool_use": false,
+        "supports_function_calling": true,
+        "supports_mid_conversation_system": true,
+        "supports_prompt_caching": true,
+        "supports_reasoning": true,
+        "supports_sampling_params": false,
+        "supports_tool_choice": true,
+        "supports_vision": true,
+        "thinking_always_on": true
     }
 }