Skip to content

Fix Bedrock HTTP 400 for gpt-5.6-sol by using auto reasoning summary #111

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
wch merged 2 commits into from
Sep 10, 2026
Merged

Fix Bedrock HTTP 400 for gpt-5.6-sol by using auto reasoning summary #111

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a791bbf
Use "auto" reasoning summary on the Bedrock Mantle Responses route
wch Sep 10, 2026
f9d686e
Update tests
wch Sep 10, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion packages/ai-provider-bridge/src/model-clients/BedrockClient.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -175,7 +175,11 @@ export class BedrockClient implements ModelClient {
store: false,
forceReasoning: true,
reasoningEffort: mantleReasoningEffort,
reasoningSummary: "detailed",
// Use "auto" (the OpenAI default) rather than an explicit level:
// Bedrock rejects `reasoning.summary: "concise"/"detailed"` for
// some models (e.g. gpt-5.6-sol), and it doesn't return summary
// text either way, so the explicit level bought nothing.
reasoningSummary: "auto",
...(usesExplicitPromptCaching
? {
promptCacheOptions: EXPLICIT_PROMPT_CACHE_OPTIONS,
Expand Down
2 changes: 1 addition & 1 deletion packages/ai-provider-bridge/src/model-clients/__tests__/bedrock-mantle-routing.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,7 +98,7 @@ describe("Bedrock Mantle protocol routing", () => {
store: false,
forceReasoning: true,
reasoningEffort: "none",
reasoningSummary: "detailed",
reasoningSummary: "auto",
},
},
}),
Expand Down
4 changes: 2 additions & 2 deletions packages/ai-provider-bridge/src/model-clients/__tests__/bedrock-mantle-wire.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -147,7 +147,7 @@ describe("Bedrock Mantle wire requests", () => {
expect(requestBody).toMatchObject({
model: "openai.gpt-5.5",
store: false,
reasoning: { effort: "none", summary: "detailed" },
reasoning: { effort: "none", summary: "auto" },
include: ["reasoning.encrypted_content"],
});
expect(requestBody).not.toHaveProperty("max_output_tokens");
Expand Down Expand Up @@ -190,7 +190,7 @@ describe("Bedrock Mantle wire requests", () => {
prompt_cache_key: "bedrock-conversation-1",
prompt_cache_options: { mode: "explicit", ttl: "30m" },
store: false,
reasoning: { effort: "high", summary: "detailed" },
reasoning: { effort: "high", summary: "auto" },
});
expect(breakpointPaths(requestBody)).toEqual([
"input[0].content[0].prompt_cache_breakpoint",
Expand Down