From 41e9cc491ed08a70b96ba1f77efb39de76680943 Mon Sep 17 00:00:00 2001 From: Mateo Wang <277851410+mateo-berri@users.noreply.github.com> Date: Thu, 9 Jul 2026 14:31:27 -0700 Subject: [PATCH] fix(bedrock-invoke): retain clear_tool_uses_20250919 context_management edits and emit context-management-2025-06-27 beta (LIT-3393) (#32658) * fix(bedrock-invoke): retain clear_tool_uses_20250919 context_management edits and emit context-management-2025-06-27 beta (LIT-3393) Copy of #29206 by oss-agent-shin, rebased onto litellm_internal_staging so CircleCI can run. Bedrock InvokeModel supports automatic tool-call clearing (clear_tool_uses_20250919) under the context-management-2025-06-27 beta, but LiteLLM stripped the edit and dropped the beta header, causing a Bedrock 400. This maps bedrock.context-management-2025-06-27 to itself in anthropic_beta_headers_config.json (bedrock_converse stays null) and rewrites _filter_context_management_for_bedrock_invoke around an allowlist of supported edit types that keeps each supported edit and adds its matching beta. * test(bedrock-invoke): restore beta-headers config cache with a shared fixture in LIT-3393 tests Greptile flagged that three of the four new tests reloaded the module-level beta-headers config into local mode without restoring it on teardown, leaking state into later tests in the same process. Move setup/teardown into a local_beta_headers_config fixture used by all four tests. --------- Co-authored-by: oss-agent-shin --- litellm/anthropic_beta_headers_config.json | 2 +- .../anthropic_claude3_transformation.py | 57 ++++-- .../test_anthropic_claude3_transformation.py | 171 ++++++++++++++++++ 3 files changed, 211 insertions(+), 19 deletions(-) diff --git a/litellm/anthropic_beta_headers_config.json b/litellm/anthropic_beta_headers_config.json index 11fdb26e42d..3f6817f6e35 100644 --- a/litellm/anthropic_beta_headers_config.json +++ b/litellm/anthropic_beta_headers_config.json @@ -102,7 +102,7 @@ "computer-use-2025-01-24": "computer-use-2025-01-24", "computer-use-2025-11-24": "computer-use-2025-11-24", "context-1m-2025-08-07": "context-1m-2025-08-07", - "context-management-2025-06-27": null, + "context-management-2025-06-27": "context-management-2025-06-27", "effort-2025-11-24": "effort-2025-11-24", "fast-mode-2026-02-01": null, "files-api-2025-04-14": null, diff --git a/litellm/llms/bedrock/messages/invoke_transformations/anthropic_claude3_transformation.py b/litellm/llms/bedrock/messages/invoke_transformations/anthropic_claude3_transformation.py index f5309d521a9..daee3369a3c 100644 --- a/litellm/llms/bedrock/messages/invoke_transformations/anthropic_claude3_transformation.py +++ b/litellm/llms/bedrock/messages/invoke_transformations/anthropic_claude3_transformation.py @@ -489,24 +489,43 @@ class AmazonAnthropicClaudeMessagesConfig( if self._supports_tool_search_on_bedrock(model): beta_set.add("tool-search-tool-2025-10-19") + # Bedrock-InvokeModel-supported ``context_management.edits`` types and the + # ``anthropic-beta`` header that each one requires. ``clear_thinking_20251015`` + # is intentionally absent — it is LiteLLM-internal, consumed via + # ``_ensure_thinking_for_clear_thinking_context_management``, and forwarding + # the raw edit trips Bedrock's + # ``"context_management: Extra inputs are not permitted"`` 400. + # + # Bedrock InvokeModel DOES support ``clear_tool_uses_20250919`` under the + # ``context-management-2025-06-27`` beta. AWS docs: + # https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-anthropic-claude-messages-tool-use.md + _BEDROCK_INVOKE_SUPPORTED_CONTEXT_MANAGEMENT_EDITS: Dict[str, str] = { + "compact_20260112": ANTHROPIC_BETA_HEADER_VALUES.COMPACT_2026_01_12.value, + "clear_tool_uses_20250919": ANTHROPIC_BETA_HEADER_VALUES.CONTEXT_MANAGEMENT_2025_06_27.value, + } + @staticmethod def _filter_context_management_for_bedrock_invoke( anthropic_messages_request: Dict, beta_set: set, ) -> None: """ - Bedrock InvokeModel accepts ``context_management`` only when it carries - ``compact_20260112`` edits paired with the ``compact-2026-01-12`` - anthropic-beta header. Other edit types (notably ``clear_thinking_20251015``, - which Claude Code sends on every request) are LiteLLM-internal and would - cause Bedrock to 400 with ``"context_management: Extra inputs are not - permitted"``. + Filter ``context_management.edits`` to the subset that Bedrock InvokeModel + accepts and add the matching ``anthropic-beta`` header for each surviving + edit type. - Filter the edits list to the supported subset, add the beta header when - compact edits remain, and drop ``context_management`` entirely when no - supported edits are left so the safety-net allowlist can pass it through. + - ``compact_20260112`` -> ``compact-2026-01-12`` + - ``clear_tool_uses_20250919`` -> ``context-management-2025-06-27`` - Ref: https://github.com/BerriAI/litellm/issues/27532 + Other edit types (notably ``clear_thinking_20251015``, which Claude Code + sends on every request) are LiteLLM-internal: thinking is injected + separately via ``_ensure_thinking_for_clear_thinking_context_management``, + and forwarding the raw edit would trip Bedrock's + ``"context_management: Extra inputs are not permitted"`` 400. + + Refs: + * https://github.com/BerriAI/litellm/issues/27532 + * https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-anthropic-claude-messages-tool-use.md """ cm = anthropic_messages_request.get("context_management") if not isinstance(cm, dict): @@ -516,15 +535,17 @@ class AmazonAnthropicClaudeMessagesConfig( anthropic_messages_request.pop("context_management", None) return - compact_edits = [e for e in edits if isinstance(e, dict) and e.get("type") == "compact_20260112"] - if compact_edits: - beta_set.add(ANTHROPIC_BETA_HEADER_VALUES.COMPACT_2026_01_12.value) - anthropic_messages_request["context_management"] = { - **cm, - "edits": compact_edits, - } - else: + supported = AmazonAnthropicClaudeMessagesConfig._BEDROCK_INVOKE_SUPPORTED_CONTEXT_MANAGEMENT_EDITS + retained_edits = [e for e in edits if isinstance(e, dict) and e.get("type") in supported] + if not retained_edits: anthropic_messages_request.pop("context_management", None) + return + + beta_set.update(supported[e["type"]] for e in retained_edits) + anthropic_messages_request["context_management"] = { + **cm, + "edits": retained_edits, + } def _get_bedrock_invoke_anthropic_beta_headers( self, diff --git a/tests/test_litellm/llms/bedrock/messages/invoke_transformations/test_anthropic_claude3_transformation.py b/tests/test_litellm/llms/bedrock/messages/invoke_transformations/test_anthropic_claude3_transformation.py index d7a62aae38b..532c6ff3598 100644 --- a/tests/test_litellm/llms/bedrock/messages/invoke_transformations/test_anthropic_claude3_transformation.py +++ b/tests/test_litellm/llms/bedrock/messages/invoke_transformations/test_anthropic_claude3_transformation.py @@ -2090,3 +2090,174 @@ def test_bedrock_clear_thinking_leaves_enabled_thinking_on_non_adaptive_model(): assert changed is False assert request["thinking"] == {"type": "enabled", "budget_tokens": 8000} assert "output_config" not in request + + +@pytest.fixture +def local_beta_headers_config(monkeypatch): + from litellm.anthropic_beta_headers_manager import reload_beta_headers_config + + monkeypatch.setenv("LITELLM_LOCAL_ANTHROPIC_BETA_HEADERS", "True") + reload_beta_headers_config() + yield + monkeypatch.delenv("LITELLM_LOCAL_ANTHROPIC_BETA_HEADERS", raising=False) + reload_beta_headers_config() + + +def test_bedrock_messages_preserves_clear_tool_uses_context_management_and_adds_beta( + local_beta_headers_config, +): + """ + LIT-3393: Bedrock InvokeModel supports automatic tool-call clearing via + ``clear_tool_uses_20250919`` under the ``context-management-2025-06-27`` + beta. Before the LIT-3393 fix, the transformation stripped this edit (only + ``compact_20260112`` survived) AND the beta was filtered out by + ``filter_and_transform_beta_headers`` for ``bedrock``, producing a Bedrock + 400 ``"context_management: Extra inputs are not permitted"``. + + Post-fix, the edit must reach the body and the beta must reach + ``anthropic_beta``. + + AWS docs ("Automatic tool call clearing (Beta)"): + https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-anthropic-claude-messages-tool-use.md + """ + from litellm.types.router import GenericLiteLLMParams + + cfg = AmazonAnthropicClaudeMessagesConfig() + messages = [{"role": "user", "content": [{"type": "text", "text": "Hi"}]}] + optional_params = { + "max_tokens": 4096, + "context_management": { + "edits": [{"type": "clear_tool_uses_20250919"}] + }, + } + + result = cfg.transform_anthropic_messages_request( + model="anthropic.claude-haiku-4-5-20251001-v1:0", + messages=messages, + anthropic_messages_optional_request_params=optional_params, + litellm_params=GenericLiteLLMParams(), + headers={}, + ) + + assert result.get("context_management") == { + "edits": [{"type": "clear_tool_uses_20250919"}] + }, "clear_tool_uses_20250919 edit must reach Bedrock InvokeModel body" + assert "context-management-2025-06-27" in result.get("anthropic_beta", []), ( + "context-management-2025-06-27 beta must reach the InvokeModel body so " + "the tool-call-clearing edit is accepted" + ) + + +def test_bedrock_messages_preserves_mixed_compact_and_clear_tool_uses_edits( + local_beta_headers_config, +): + """ + LIT-3393: a request mixing ``compact_20260112`` and + ``clear_tool_uses_20250919`` must keep BOTH edits and emit BOTH + anthropic-beta values (``compact-2026-01-12`` + ``context-management-2025-06-27``). + """ + from litellm.types.router import GenericLiteLLMParams + + cfg = AmazonAnthropicClaudeMessagesConfig() + messages = [{"role": "user", "content": [{"type": "text", "text": "Hi"}]}] + optional_params = { + "max_tokens": 4096, + "context_management": { + "edits": [ + {"type": "compact_20260112"}, + {"type": "clear_tool_uses_20250919"}, + ] + }, + } + + result = cfg.transform_anthropic_messages_request( + model="anthropic.claude-sonnet-4-6-20250929-v1:0", + messages=messages, + anthropic_messages_optional_request_params=optional_params, + litellm_params=GenericLiteLLMParams(), + headers={}, + ) + + cm = result.get("context_management") + assert cm is not None + edit_types = sorted(e.get("type") for e in cm["edits"]) + assert edit_types == ["clear_tool_uses_20250919", "compact_20260112"] + + betas = result.get("anthropic_beta", []) + assert "compact-2026-01-12" in betas + assert "context-management-2025-06-27" in betas + + +def test_bedrock_messages_filters_clear_thinking_keeps_clear_tool_uses( + local_beta_headers_config, +): + """ + LIT-3393: ``clear_thinking_20251015`` remains LiteLLM-internal (consumed via + thinking-injection) and MUST be stripped from the body, while + ``clear_tool_uses_20250919`` (officially supported on Bedrock InvokeModel) + survives in the same request. + """ + from litellm.types.router import GenericLiteLLMParams + + cfg = AmazonAnthropicClaudeMessagesConfig() + messages = [{"role": "user", "content": [{"type": "text", "text": "Hi"}]}] + optional_params = { + "max_tokens": 4096, + "context_management": { + "edits": [ + {"type": "clear_thinking_20251015", "keep": "all"}, + {"type": "clear_tool_uses_20250919"}, + ] + }, + } + + result = cfg.transform_anthropic_messages_request( + model="anthropic.claude-opus-4-7", + messages=messages, + anthropic_messages_optional_request_params=optional_params, + litellm_params=GenericLiteLLMParams(), + headers={}, + ) + + cm = result.get("context_management") + assert cm is not None + assert [e.get("type") for e in cm["edits"]] == [ + "clear_tool_uses_20250919" + ], "clear_thinking_20251015 must still be stripped (LiteLLM-internal)" + + betas = result.get("anthropic_beta", []) + assert "context-management-2025-06-27" in betas + # ``compact-2026-01-12`` was not requested. + assert "compact-2026-01-12" not in betas + + +def test_filter_and_transform_beta_headers_passes_context_management_for_bedrock( + local_beta_headers_config, +): + """ + LIT-3393: ``anthropic_beta_headers_config.json`` previously mapped + ``bedrock.context-management-2025-06-27`` to ``null``, so + ``filter_and_transform_beta_headers`` dropped the header even when the + transformation tried to set it. This regression guard locks the bundled + mapping in place. + + Pinned to the bundled local config via ``LITELLM_LOCAL_ANTHROPIC_BETA_HEADERS`` + so the assertion is not subject to whatever the upstream remote currently + serves or what previous tests left in the module cache. + """ + from litellm.anthropic_beta_headers_manager import filter_and_transform_beta_headers + + out = filter_and_transform_beta_headers( + ["context-management-2025-06-27"], + provider="bedrock", + ) + assert out == ["context-management-2025-06-27"] + + # Bedrock_converse genuinely lacks it per AWS docs; this guard prevents + # an accidental flip there. + out_converse = filter_and_transform_beta_headers( + ["context-management-2025-06-27"], + provider="bedrock_converse", + ) + assert out_converse == [] +