feat: migrate xAI provider to Responses API with reusable transform utils by carlesso · Pull Request #11962 · RooCodeInc/Roo-Code

carlesso · 2026-03-19T23:21:03Z

Related GitHub Issue

Closes: #11961

Roo Code Task Context (Optional)

N/A

Description

Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt.

New shared utilities (src/api/transform/):

responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider.
responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings.

xAI provider (src/api/providers/xai.ts):

Switch from client.chat.completions.create() to client.responses.create()
Use shared transform utilities for stream handling and input conversion
Enable reasoning traces via include: ["reasoning.encrypted_content"]
System prompt via instructions field, store: false for privacy
completePrompt() also migrated to Responses API

Model updates (packages/types/src/providers/xai.ts):

Add grok-4.20 as the new default model (2M context, $2/$6 pricing)
Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

Test Procedure

Tested locally

Pre-Submission Checklist

Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
Scope: My changes are focused on the linked issue (one major feature/fix per PR).
Self-Review: I have performed a thorough self-review of my code.
Testing: New and/or updated tests have been added to cover my changes (if applicable).
Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

N/A

Documentation Updates

Does this PR necessitate updates to user-facing documentation?

No documentation updates are required.
Yes, documentation updates are required. (Please describe what needs to be updated or link to a PR in the docs repository).

Additional Notes

N/A

Get in Touch

Discord: ecarlesso

Interactively review PR in Roo Code Cloud

…tilities Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt. New shared utilities (src/api/transform/): - responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider. - responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings. xAI provider (src/api/providers/xai.ts): - Switch from client.chat.completions.create() to client.responses.create() - Use shared transform utilities for stream handling and input conversion - Enable reasoning traces via include: ["reasoning.encrypted_content"] - System prompt via instructions field, store: false for privacy - completePrompt() also migrated to Responses API Model updates (packages/types/src/providers/xai.ts): - Add grok-4.20 as the new default model (2M context, $2/$6 pricing) - Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

hannesrudolph · 2026-03-19T23:54:23Z

@roomote do a full review please and leave inline comments.

ghost · 2026-03-19T23:55:58Z

Fixed the reported issues. All local checks (tests, lint, type-check) passed.

_{View commit | Revert commit}

ghost

Good migration to the Responses API overall. The shared transform utilities are a solid pattern. I found several issues that need addressing -- mostly around tool schema handling, missing metadata pass-through, and a bug in the input converter for assistant string content.

Issues found:

mapResponseTools bypasses base provider schema hardening (src/api/providers/xai.ts:52-64) - Doesn't use convertToolSchemaForOpenAI(), doesn't handle MCP tools, sets strict: false for all tools.
metadata?.tool_choice is ignored (src/api/providers/xai.ts:89) - Always uses "auto" instead of respecting caller's tool_choice. Also missing parallel_tool_calls.
Assistant string content uses input_text instead of output_text (src/api/transform/responses-api-input.ts:20-26) - For assistant role with string content, wraps as input_text but should use output_text / message format.
modelInfo parameter unused (src/api/transform/responses-api-stream.ts:113) - createUsageNormalizer accepts modelInfo but never references it.
Missing cacheWriteTokens extraction (src/api/transform/responses-api-stream.ts:133) - Normalizer doesn't extract cache_creation_input_tokens.
Pervasive any types - Both shared utilities use any extensively. Not blocking but worth noting.
Inconsistent test function (src/api/providers/__tests__/xai.spec.ts:85) - Uses test() while all others use it().

Will fix all of these.

hannesrudolph · 2026-03-20T00:03:33Z

@roomote fix it all

- Use base provider convertToolSchemaForOpenAI() for tool schema hardening - Handle MCP tools with isMcpTool() (strict: false for MCP, true otherwise) - Respect metadata.tool_choice instead of always using "auto" - Add parallel_tool_calls pass-through from metadata - Fix assistant string content to use output_text format (not input_text) - Remove unused modelInfo param from createUsageNormalizer - Add cacheWriteTokens extraction to usage normalizer - Fix test() -> it() inconsistency in xai.spec.ts - Update tests to match all changes

ghost · 2026-03-20T00:05:18Z

Fixed the reported issues. Restored getModelParams for proper reasoning effort/temperature/maxTokens handling, simplified completePrompt to use output_text, fixed duplicate cacheWriteTokens property in usage normalizer, and added tests for reasoning effort with mini models. All local checks passed.

_{View commit | Revert commit}

…eWriteTokens, simplify completePrompt

carlesso · 2026-03-30T16:19:27Z

Hey @hannesrudolph could you kindly give this another go?

hannesrudolph · 2026-04-07T18:35:12Z

@roomote please re-review this

ghost · 2026-04-07T18:37:01Z

Fixed three issues found during re-review: removed double schema processing in mapResponseTools (was calling convertToolSchemaForOpenAI twice), removed redundant stream: true in the request body, and added URL image source support in convertToResponsesApiInput. All tests and type checks pass.

_{View commit | Revert commit}

hannesrudolph · 2026-04-08T14:46:09Z

@roomote are there any outstanding review issues?

ghost · 2026-04-08T14:47:58Z

No outstanding review issues. The original review raised 7 items (schema hardening, tool_choice pass-through, assistant string format, unused param, missing cacheWriteTokens, any types, test consistency) -- all were addressed in the follow-up commits. CI is green, all 55 tests pass, and TypeScript compiles cleanly.

carlesso · 2026-04-08T15:25:12Z

Manually re-tested with the latest changes, all looks great!

done

carlesso · 2026-03-19T23:21:03Z

Related GitHub Issue

Closes: #11961

Roo Code Task Context (Optional)

N/A

Description

Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt.

New shared utilities (src/api/transform/):

responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider.
responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings.

xAI provider (src/api/providers/xai.ts):

Switch from client.chat.completions.create() to client.responses.create()
Use shared transform utilities for stream handling and input conversion
Enable reasoning traces via include: ["reasoning.encrypted_content"]
System prompt via instructions field, store: false for privacy
completePrompt() also migrated to Responses API

Model updates (packages/types/src/providers/xai.ts):

Add grok-4.20 as the new default model (2M context, $2/$6 pricing)
Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

Test Procedure

Tested locally

Pre-Submission Checklist

Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
Scope: My changes are focused on the linked issue (one major feature/fix per PR).
Self-Review: I have performed a thorough self-review of my code.
Testing: New and/or updated tests have been added to cover my changes (if applicable).
Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

N/A

Documentation Updates

Does this PR necessitate updates to user-facing documentation?

No documentation updates are required.
Yes, documentation updates are required. (Please describe what needs to be updated or link to a PR in the docs repository).

Additional Notes

N/A

Get in Touch

Discord: ecarlesso

Interactively review PR in Roo Code Cloud

…tilities Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt. New shared utilities (src/api/transform/): - responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider. - responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings. xAI provider (src/api/providers/xai.ts): - Switch from client.chat.completions.create() to client.responses.create() - Use shared transform utilities for stream handling and input conversion - Enable reasoning traces via include: ["reasoning.encrypted_content"] - System prompt via instructions field, store: false for privacy - completePrompt() also migrated to Responses API Model updates (packages/types/src/providers/xai.ts): - Add grok-4.20 as the new default model (2M context, $2/$6 pricing) - Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

hannesrudolph · 2026-03-19T23:54:23Z

@roomote do a full review please and leave inline comments.

ghost · 2026-03-19T23:55:58Z

Fixed the reported issues. All local checks (tests, lint, type-check) passed.

_{View commit | Revert commit}

ghost

Good migration to the Responses API overall. The shared transform utilities are a solid pattern. I found several issues that need addressing -- mostly around tool schema handling, missing metadata pass-through, and a bug in the input converter for assistant string content.

Issues found:

mapResponseTools bypasses base provider schema hardening (src/api/providers/xai.ts:52-64) - Doesn't use convertToolSchemaForOpenAI(), doesn't handle MCP tools, sets strict: false for all tools.
metadata?.tool_choice is ignored (src/api/providers/xai.ts:89) - Always uses "auto" instead of respecting caller's tool_choice. Also missing parallel_tool_calls.
Assistant string content uses input_text instead of output_text (src/api/transform/responses-api-input.ts:20-26) - For assistant role with string content, wraps as input_text but should use output_text / message format.
modelInfo parameter unused (src/api/transform/responses-api-stream.ts:113) - createUsageNormalizer accepts modelInfo but never references it.
Missing cacheWriteTokens extraction (src/api/transform/responses-api-stream.ts:133) - Normalizer doesn't extract cache_creation_input_tokens.
Pervasive any types - Both shared utilities use any extensively. Not blocking but worth noting.
Inconsistent test function (src/api/providers/__tests__/xai.spec.ts:85) - Uses test() while all others use it().

Will fix all of these.

hannesrudolph · 2026-03-20T00:03:33Z

@roomote fix it all

- Use base provider convertToolSchemaForOpenAI() for tool schema hardening - Handle MCP tools with isMcpTool() (strict: false for MCP, true otherwise) - Respect metadata.tool_choice instead of always using "auto" - Add parallel_tool_calls pass-through from metadata - Fix assistant string content to use output_text format (not input_text) - Remove unused modelInfo param from createUsageNormalizer - Add cacheWriteTokens extraction to usage normalizer - Fix test() -> it() inconsistency in xai.spec.ts - Update tests to match all changes

ghost · 2026-03-20T00:05:18Z

Fixed the reported issues. Restored getModelParams for proper reasoning effort/temperature/maxTokens handling, simplified completePrompt to use output_text, fixed duplicate cacheWriteTokens property in usage normalizer, and added tests for reasoning effort with mini models. All local checks passed.

_{View commit | Revert commit}

…eWriteTokens, simplify completePrompt

carlesso · 2026-03-30T16:19:27Z

Hey @hannesrudolph could you kindly give this another go?

hannesrudolph · 2026-04-07T18:35:12Z

@roomote please re-review this

ghost · 2026-04-07T18:37:01Z

Fixed three issues found during re-review: removed double schema processing in mapResponseTools (was calling convertToolSchemaForOpenAI twice), removed redundant stream: true in the request body, and added URL image source support in convertToResponsesApiInput. All tests and type checks pass.

_{View commit | Revert commit}

hannesrudolph · 2026-04-08T14:46:09Z

@roomote are there any outstanding review issues?

ghost · 2026-04-08T14:47:58Z

No outstanding review issues. The original review raised 7 items (schema hardening, tool_choice pass-through, assistant string format, unused param, missing cacheWriteTokens, any types, test consistency) -- all were addressed in the follow-up commits. CI is green, all 55 tests pass, and TypeScript compiles cleanly.

carlesso · 2026-04-08T15:25:12Z

Manually re-tested with the latest changes, all looks great!

done

carlesso · 2026-03-19T23:21:03Z

Related GitHub Issue

Closes: #11961

Roo Code Task Context (Optional)

N/A

Description

Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt.

New shared utilities (src/api/transform/):

responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider.
responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings.

xAI provider (src/api/providers/xai.ts):

Switch from client.chat.completions.create() to client.responses.create()
Use shared transform utilities for stream handling and input conversion
Enable reasoning traces via include: ["reasoning.encrypted_content"]
System prompt via instructions field, store: false for privacy
completePrompt() also migrated to Responses API

Model updates (packages/types/src/providers/xai.ts):

Add grok-4.20 as the new default model (2M context, $2/$6 pricing)
Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

Test Procedure

Tested locally

Pre-Submission Checklist

Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
Scope: My changes are focused on the linked issue (one major feature/fix per PR).
Self-Review: I have performed a thorough self-review of my code.
Testing: New and/or updated tests have been added to cover my changes (if applicable).
Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

N/A

Documentation Updates

Does this PR necessitate updates to user-facing documentation?

No documentation updates are required.
Yes, documentation updates are required. (Please describe what needs to be updated or link to a PR in the docs repository).

Additional Notes

N/A

Get in Touch

Discord: ecarlesso

Interactively review PR in Roo Code Cloud

…tilities Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt. New shared utilities (src/api/transform/): - responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider. - responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings. xAI provider (src/api/providers/xai.ts): - Switch from client.chat.completions.create() to client.responses.create() - Use shared transform utilities for stream handling and input conversion - Enable reasoning traces via include: ["reasoning.encrypted_content"] - System prompt via instructions field, store: false for privacy - completePrompt() also migrated to Responses API Model updates (packages/types/src/providers/xai.ts): - Add grok-4.20 as the new default model (2M context, $2/$6 pricing) - Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

hannesrudolph · 2026-03-19T23:54:23Z

@roomote do a full review please and leave inline comments.

ghost · 2026-03-19T23:55:58Z

Fixed the reported issues. All local checks (tests, lint, type-check) passed.

_{View commit | Revert commit}

ghost

Good migration to the Responses API overall. The shared transform utilities are a solid pattern. I found several issues that need addressing -- mostly around tool schema handling, missing metadata pass-through, and a bug in the input converter for assistant string content.

Issues found:

mapResponseTools bypasses base provider schema hardening (src/api/providers/xai.ts:52-64) - Doesn't use convertToolSchemaForOpenAI(), doesn't handle MCP tools, sets strict: false for all tools.
metadata?.tool_choice is ignored (src/api/providers/xai.ts:89) - Always uses "auto" instead of respecting caller's tool_choice. Also missing parallel_tool_calls.
Assistant string content uses input_text instead of output_text (src/api/transform/responses-api-input.ts:20-26) - For assistant role with string content, wraps as input_text but should use output_text / message format.
modelInfo parameter unused (src/api/transform/responses-api-stream.ts:113) - createUsageNormalizer accepts modelInfo but never references it.
Missing cacheWriteTokens extraction (src/api/transform/responses-api-stream.ts:133) - Normalizer doesn't extract cache_creation_input_tokens.
Pervasive any types - Both shared utilities use any extensively. Not blocking but worth noting.
Inconsistent test function (src/api/providers/__tests__/xai.spec.ts:85) - Uses test() while all others use it().

Will fix all of these.

hannesrudolph · 2026-03-20T00:03:33Z

@roomote fix it all

- Use base provider convertToolSchemaForOpenAI() for tool schema hardening - Handle MCP tools with isMcpTool() (strict: false for MCP, true otherwise) - Respect metadata.tool_choice instead of always using "auto" - Add parallel_tool_calls pass-through from metadata - Fix assistant string content to use output_text format (not input_text) - Remove unused modelInfo param from createUsageNormalizer - Add cacheWriteTokens extraction to usage normalizer - Fix test() -> it() inconsistency in xai.spec.ts - Update tests to match all changes

ghost · 2026-03-20T00:05:18Z

Fixed the reported issues. Restored getModelParams for proper reasoning effort/temperature/maxTokens handling, simplified completePrompt to use output_text, fixed duplicate cacheWriteTokens property in usage normalizer, and added tests for reasoning effort with mini models. All local checks passed.

_{View commit | Revert commit}

…eWriteTokens, simplify completePrompt

carlesso · 2026-03-30T16:19:27Z

Hey @hannesrudolph could you kindly give this another go?

hannesrudolph · 2026-04-07T18:35:12Z

@roomote please re-review this

ghost · 2026-04-07T18:37:01Z

Fixed three issues found during re-review: removed double schema processing in mapResponseTools (was calling convertToolSchemaForOpenAI twice), removed redundant stream: true in the request body, and added URL image source support in convertToResponsesApiInput. All tests and type checks pass.

_{View commit | Revert commit}

hannesrudolph · 2026-04-08T14:46:09Z

@roomote are there any outstanding review issues?

ghost · 2026-04-08T14:47:58Z

No outstanding review issues. The original review raised 7 items (schema hardening, tool_choice pass-through, assistant string format, unused param, missing cacheWriteTokens, any types, test consistency) -- all were addressed in the follow-up commits. CI is green, all 55 tests pass, and TypeScript compiles cleanly.

carlesso · 2026-04-08T15:25:12Z

Manually re-tested with the latest changes, all looks great!

done

carlesso · 2026-03-19T23:21:03Z

Related GitHub Issue

Closes: #11961

Roo Code Task Context (Optional)

N/A

Description

Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt.

New shared utilities (src/api/transform/):

responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider.
responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings.

xAI provider (src/api/providers/xai.ts):

Switch from client.chat.completions.create() to client.responses.create()
Use shared transform utilities for stream handling and input conversion
Enable reasoning traces via include: ["reasoning.encrypted_content"]
System prompt via instructions field, store: false for privacy
completePrompt() also migrated to Responses API

Model updates (packages/types/src/providers/xai.ts):

Add grok-4.20 as the new default model (2M context, $2/$6 pricing)
Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

Test Procedure

Tested locally

Pre-Submission Checklist

Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
Scope: My changes are focused on the linked issue (one major feature/fix per PR).
Self-Review: I have performed a thorough self-review of my code.
Testing: New and/or updated tests have been added to cover my changes (if applicable).
Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

N/A

Documentation Updates

Does this PR necessitate updates to user-facing documentation?

No documentation updates are required.
Yes, documentation updates are required. (Please describe what needs to be updated or link to a PR in the docs repository).

Additional Notes

N/A

Get in Touch

Discord: ecarlesso

Interactively review PR in Roo Code Cloud

…tilities Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt. New shared utilities (src/api/transform/): - responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider. - responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings. xAI provider (src/api/providers/xai.ts): - Switch from client.chat.completions.create() to client.responses.create() - Use shared transform utilities for stream handling and input conversion - Enable reasoning traces via include: ["reasoning.encrypted_content"] - System prompt via instructions field, store: false for privacy - completePrompt() also migrated to Responses API Model updates (packages/types/src/providers/xai.ts): - Add grok-4.20 as the new default model (2M context, $2/$6 pricing) - Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

hannesrudolph · 2026-03-19T23:54:23Z

@roomote do a full review please and leave inline comments.

ghost · 2026-03-19T23:55:58Z

Fixed the reported issues. All local checks (tests, lint, type-check) passed.

_{View commit | Revert commit}

ghost

Good migration to the Responses API overall. The shared transform utilities are a solid pattern. I found several issues that need addressing -- mostly around tool schema handling, missing metadata pass-through, and a bug in the input converter for assistant string content.

Issues found:

mapResponseTools bypasses base provider schema hardening (src/api/providers/xai.ts:52-64) - Doesn't use convertToolSchemaForOpenAI(), doesn't handle MCP tools, sets strict: false for all tools.
metadata?.tool_choice is ignored (src/api/providers/xai.ts:89) - Always uses "auto" instead of respecting caller's tool_choice. Also missing parallel_tool_calls.
Assistant string content uses input_text instead of output_text (src/api/transform/responses-api-input.ts:20-26) - For assistant role with string content, wraps as input_text but should use output_text / message format.
modelInfo parameter unused (src/api/transform/responses-api-stream.ts:113) - createUsageNormalizer accepts modelInfo but never references it.
Missing cacheWriteTokens extraction (src/api/transform/responses-api-stream.ts:133) - Normalizer doesn't extract cache_creation_input_tokens.
Pervasive any types - Both shared utilities use any extensively. Not blocking but worth noting.
Inconsistent test function (src/api/providers/__tests__/xai.spec.ts:85) - Uses test() while all others use it().

Will fix all of these.

hannesrudolph · 2026-03-20T00:03:33Z

@roomote fix it all

- Use base provider convertToolSchemaForOpenAI() for tool schema hardening - Handle MCP tools with isMcpTool() (strict: false for MCP, true otherwise) - Respect metadata.tool_choice instead of always using "auto" - Add parallel_tool_calls pass-through from metadata - Fix assistant string content to use output_text format (not input_text) - Remove unused modelInfo param from createUsageNormalizer - Add cacheWriteTokens extraction to usage normalizer - Fix test() -> it() inconsistency in xai.spec.ts - Update tests to match all changes

ghost · 2026-03-20T00:05:18Z

Fixed the reported issues. Restored getModelParams for proper reasoning effort/temperature/maxTokens handling, simplified completePrompt to use output_text, fixed duplicate cacheWriteTokens property in usage normalizer, and added tests for reasoning effort with mini models. All local checks passed.

_{View commit | Revert commit}

…eWriteTokens, simplify completePrompt

carlesso · 2026-03-30T16:19:27Z

Hey @hannesrudolph could you kindly give this another go?

hannesrudolph · 2026-04-07T18:35:12Z

@roomote please re-review this

ghost · 2026-04-07T18:37:01Z

Fixed three issues found during re-review: removed double schema processing in mapResponseTools (was calling convertToolSchemaForOpenAI twice), removed redundant stream: true in the request body, and added URL image source support in convertToResponsesApiInput. All tests and type checks pass.

_{View commit | Revert commit}

hannesrudolph · 2026-04-08T14:46:09Z

@roomote are there any outstanding review issues?

ghost · 2026-04-08T14:47:58Z

No outstanding review issues. The original review raised 7 items (schema hardening, tool_choice pass-through, assistant string format, unused param, missing cacheWriteTokens, any types, test consistency) -- all were addressed in the follow-up commits. CI is green, all 55 tests pass, and TypeScript compiles cleanly.

carlesso · 2026-04-08T15:25:12Z

Manually re-tested with the latest changes, all looks great!

done

carlesso · 2026-03-19T23:21:03Z

Related GitHub Issue

Closes: #11961

Roo Code Task Context (Optional)

N/A

Description

Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt.

New shared utilities (src/api/transform/):

responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider.
responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings.

xAI provider (src/api/providers/xai.ts):

Switch from client.chat.completions.create() to client.responses.create()
Use shared transform utilities for stream handling and input conversion
Enable reasoning traces via include: ["reasoning.encrypted_content"]
System prompt via instructions field, store: false for privacy
completePrompt() also migrated to Responses API

Model updates (packages/types/src/providers/xai.ts):

Add grok-4.20 as the new default model (2M context, $2/$6 pricing)
Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

Test Procedure

Tested locally

Pre-Submission Checklist

Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
Scope: My changes are focused on the linked issue (one major feature/fix per PR).
Self-Review: I have performed a thorough self-review of my code.
Testing: New and/or updated tests have been added to cover my changes (if applicable).
Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

N/A

Documentation Updates

Does this PR necessitate updates to user-facing documentation?

No documentation updates are required.
Yes, documentation updates are required. (Please describe what needs to be updated or link to a PR in the docs repository).

Additional Notes

N/A

Get in Touch

Discord: ecarlesso

Interactively review PR in Roo Code Cloud

…tilities Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt. New shared utilities (src/api/transform/): - responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider. - responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings. xAI provider (src/api/providers/xai.ts): - Switch from client.chat.completions.create() to client.responses.create() - Use shared transform utilities for stream handling and input conversion - Enable reasoning traces via include: ["reasoning.encrypted_content"] - System prompt via instructions field, store: false for privacy - completePrompt() also migrated to Responses API Model updates (packages/types/src/providers/xai.ts): - Add grok-4.20 as the new default model (2M context, $2/$6 pricing) - Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

hannesrudolph · 2026-03-19T23:54:23Z

@roomote do a full review please and leave inline comments.

ghost · 2026-03-19T23:55:58Z

Fixed the reported issues. All local checks (tests, lint, type-check) passed.

_{View commit | Revert commit}

ghost

Good migration to the Responses API overall. The shared transform utilities are a solid pattern. I found several issues that need addressing -- mostly around tool schema handling, missing metadata pass-through, and a bug in the input converter for assistant string content.

Issues found:

mapResponseTools bypasses base provider schema hardening (src/api/providers/xai.ts:52-64) - Doesn't use convertToolSchemaForOpenAI(), doesn't handle MCP tools, sets strict: false for all tools.
metadata?.tool_choice is ignored (src/api/providers/xai.ts:89) - Always uses "auto" instead of respecting caller's tool_choice. Also missing parallel_tool_calls.
Assistant string content uses input_text instead of output_text (src/api/transform/responses-api-input.ts:20-26) - For assistant role with string content, wraps as input_text but should use output_text / message format.
modelInfo parameter unused (src/api/transform/responses-api-stream.ts:113) - createUsageNormalizer accepts modelInfo but never references it.
Missing cacheWriteTokens extraction (src/api/transform/responses-api-stream.ts:133) - Normalizer doesn't extract cache_creation_input_tokens.
Pervasive any types - Both shared utilities use any extensively. Not blocking but worth noting.
Inconsistent test function (src/api/providers/__tests__/xai.spec.ts:85) - Uses test() while all others use it().

Will fix all of these.

hannesrudolph · 2026-03-20T00:03:33Z

@roomote fix it all

- Use base provider convertToolSchemaForOpenAI() for tool schema hardening - Handle MCP tools with isMcpTool() (strict: false for MCP, true otherwise) - Respect metadata.tool_choice instead of always using "auto" - Add parallel_tool_calls pass-through from metadata - Fix assistant string content to use output_text format (not input_text) - Remove unused modelInfo param from createUsageNormalizer - Add cacheWriteTokens extraction to usage normalizer - Fix test() -> it() inconsistency in xai.spec.ts - Update tests to match all changes

ghost · 2026-03-20T00:05:18Z

Fixed the reported issues. Restored getModelParams for proper reasoning effort/temperature/maxTokens handling, simplified completePrompt to use output_text, fixed duplicate cacheWriteTokens property in usage normalizer, and added tests for reasoning effort with mini models. All local checks passed.

_{View commit | Revert commit}

…eWriteTokens, simplify completePrompt

carlesso · 2026-03-30T16:19:27Z

Hey @hannesrudolph could you kindly give this another go?

hannesrudolph · 2026-04-07T18:35:12Z

@roomote please re-review this

ghost · 2026-04-07T18:37:01Z

Fixed three issues found during re-review: removed double schema processing in mapResponseTools (was calling convertToolSchemaForOpenAI twice), removed redundant stream: true in the request body, and added URL image source support in convertToResponsesApiInput. All tests and type checks pass.

_{View commit | Revert commit}

hannesrudolph · 2026-04-08T14:46:09Z

@roomote are there any outstanding review issues?

ghost · 2026-04-08T14:47:58Z

No outstanding review issues. The original review raised 7 items (schema hardening, tool_choice pass-through, assistant string format, unused param, missing cacheWriteTokens, any types, test consistency) -- all were addressed in the follow-up commits. CI is green, all 55 tests pass, and TypeScript compiles cleanly.

carlesso · 2026-04-08T15:25:12Z

Manually re-tested with the latest changes, all looks great!

done

carlesso · 2026-03-19T23:21:03Z

Related GitHub Issue

Closes: #11961

Roo Code Task Context (Optional)

N/A

Description

Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt.

New shared utilities (src/api/transform/):

responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider.
responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings.

xAI provider (src/api/providers/xai.ts):

Switch from client.chat.completions.create() to client.responses.create()
Use shared transform utilities for stream handling and input conversion
Enable reasoning traces via include: ["reasoning.encrypted_content"]
System prompt via instructions field, store: false for privacy
completePrompt() also migrated to Responses API

Model updates (packages/types/src/providers/xai.ts):

Add grok-4.20 as the new default model (2M context, $2/$6 pricing)
Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

Test Procedure

Tested locally

Pre-Submission Checklist

Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
Scope: My changes are focused on the linked issue (one major feature/fix per PR).
Self-Review: I have performed a thorough self-review of my code.
Testing: New and/or updated tests have been added to cover my changes (if applicable).
Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

N/A

Documentation Updates

Does this PR necessitate updates to user-facing documentation?

No documentation updates are required.
Yes, documentation updates are required. (Please describe what needs to be updated or link to a PR in the docs repository).

Additional Notes

N/A

Get in Touch

Discord: ecarlesso

Interactively review PR in Roo Code Cloud

…tilities Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt. New shared utilities (src/api/transform/): - responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider. - responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings. xAI provider (src/api/providers/xai.ts): - Switch from client.chat.completions.create() to client.responses.create() - Use shared transform utilities for stream handling and input conversion - Enable reasoning traces via include: ["reasoning.encrypted_content"] - System prompt via instructions field, store: false for privacy - completePrompt() also migrated to Responses API Model updates (packages/types/src/providers/xai.ts): - Add grok-4.20 as the new default model (2M context, $2/$6 pricing) - Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

hannesrudolph · 2026-03-19T23:54:23Z

@roomote do a full review please and leave inline comments.

ghost · 2026-03-19T23:55:58Z

Fixed the reported issues. All local checks (tests, lint, type-check) passed.

_{View commit | Revert commit}

ghost

Good migration to the Responses API overall. The shared transform utilities are a solid pattern. I found several issues that need addressing -- mostly around tool schema handling, missing metadata pass-through, and a bug in the input converter for assistant string content.

Issues found:

mapResponseTools bypasses base provider schema hardening (src/api/providers/xai.ts:52-64) - Doesn't use convertToolSchemaForOpenAI(), doesn't handle MCP tools, sets strict: false for all tools.
metadata?.tool_choice is ignored (src/api/providers/xai.ts:89) - Always uses "auto" instead of respecting caller's tool_choice. Also missing parallel_tool_calls.
Assistant string content uses input_text instead of output_text (src/api/transform/responses-api-input.ts:20-26) - For assistant role with string content, wraps as input_text but should use output_text / message format.
modelInfo parameter unused (src/api/transform/responses-api-stream.ts:113) - createUsageNormalizer accepts modelInfo but never references it.
Missing cacheWriteTokens extraction (src/api/transform/responses-api-stream.ts:133) - Normalizer doesn't extract cache_creation_input_tokens.
Pervasive any types - Both shared utilities use any extensively. Not blocking but worth noting.
Inconsistent test function (src/api/providers/__tests__/xai.spec.ts:85) - Uses test() while all others use it().

Will fix all of these.

hannesrudolph · 2026-03-20T00:03:33Z

@roomote fix it all

- Use base provider convertToolSchemaForOpenAI() for tool schema hardening - Handle MCP tools with isMcpTool() (strict: false for MCP, true otherwise) - Respect metadata.tool_choice instead of always using "auto" - Add parallel_tool_calls pass-through from metadata - Fix assistant string content to use output_text format (not input_text) - Remove unused modelInfo param from createUsageNormalizer - Add cacheWriteTokens extraction to usage normalizer - Fix test() -> it() inconsistency in xai.spec.ts - Update tests to match all changes

ghost · 2026-03-20T00:05:18Z

Fixed the reported issues. Restored getModelParams for proper reasoning effort/temperature/maxTokens handling, simplified completePrompt to use output_text, fixed duplicate cacheWriteTokens property in usage normalizer, and added tests for reasoning effort with mini models. All local checks passed.

_{View commit | Revert commit}

…eWriteTokens, simplify completePrompt

carlesso · 2026-03-30T16:19:27Z

Hey @hannesrudolph could you kindly give this another go?

hannesrudolph · 2026-04-07T18:35:12Z

@roomote please re-review this

ghost · 2026-04-07T18:37:01Z

Fixed three issues found during re-review: removed double schema processing in mapResponseTools (was calling convertToolSchemaForOpenAI twice), removed redundant stream: true in the request body, and added URL image source support in convertToResponsesApiInput. All tests and type checks pass.

_{View commit | Revert commit}

hannesrudolph · 2026-04-08T14:46:09Z

@roomote are there any outstanding review issues?

ghost · 2026-04-08T14:47:58Z

No outstanding review issues. The original review raised 7 items (schema hardening, tool_choice pass-through, assistant string format, unused param, missing cacheWriteTokens, any types, test consistency) -- all were addressed in the follow-up commits. CI is green, all 55 tests pass, and TypeScript compiles cleanly.

carlesso · 2026-04-08T15:25:12Z

Manually re-tested with the latest changes, all looks great!

done

carlesso · 2026-03-19T23:21:03Z

Related GitHub Issue

Closes: #11961

Roo Code Task Context (Optional)

N/A

Description

Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt.

New shared utilities (src/api/transform/):

responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider.
responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings.

xAI provider (src/api/providers/xai.ts):

Switch from client.chat.completions.create() to client.responses.create()
Use shared transform utilities for stream handling and input conversion
Enable reasoning traces via include: ["reasoning.encrypted_content"]
System prompt via instructions field, store: false for privacy
completePrompt() also migrated to Responses API

Model updates (packages/types/src/providers/xai.ts):

Add grok-4.20 as the new default model (2M context, $2/$6 pricing)
Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

Test Procedure

Tested locally

Pre-Submission Checklist

Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
Scope: My changes are focused on the linked issue (one major feature/fix per PR).
Self-Review: I have performed a thorough self-review of my code.
Testing: New and/or updated tests have been added to cover my changes (if applicable).
Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

N/A

Documentation Updates

Does this PR necessitate updates to user-facing documentation?

No documentation updates are required.
Yes, documentation updates are required. (Please describe what needs to be updated or link to a PR in the docs repository).

Additional Notes

N/A

Get in Touch

Discord: ecarlesso

Interactively review PR in Roo Code Cloud

…tilities Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt. New shared utilities (src/api/transform/): - responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider. - responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings. xAI provider (src/api/providers/xai.ts): - Switch from client.chat.completions.create() to client.responses.create() - Use shared transform utilities for stream handling and input conversion - Enable reasoning traces via include: ["reasoning.encrypted_content"] - System prompt via instructions field, store: false for privacy - completePrompt() also migrated to Responses API Model updates (packages/types/src/providers/xai.ts): - Add grok-4.20 as the new default model (2M context, $2/$6 pricing) - Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

hannesrudolph · 2026-03-19T23:54:23Z

@roomote do a full review please and leave inline comments.

ghost · 2026-03-19T23:55:58Z

Fixed the reported issues. All local checks (tests, lint, type-check) passed.

_{View commit | Revert commit}

ghost

Good migration to the Responses API overall. The shared transform utilities are a solid pattern. I found several issues that need addressing -- mostly around tool schema handling, missing metadata pass-through, and a bug in the input converter for assistant string content.

Issues found:

mapResponseTools bypasses base provider schema hardening (src/api/providers/xai.ts:52-64) - Doesn't use convertToolSchemaForOpenAI(), doesn't handle MCP tools, sets strict: false for all tools.
metadata?.tool_choice is ignored (src/api/providers/xai.ts:89) - Always uses "auto" instead of respecting caller's tool_choice. Also missing parallel_tool_calls.
Assistant string content uses input_text instead of output_text (src/api/transform/responses-api-input.ts:20-26) - For assistant role with string content, wraps as input_text but should use output_text / message format.
modelInfo parameter unused (src/api/transform/responses-api-stream.ts:113) - createUsageNormalizer accepts modelInfo but never references it.
Missing cacheWriteTokens extraction (src/api/transform/responses-api-stream.ts:133) - Normalizer doesn't extract cache_creation_input_tokens.
Pervasive any types - Both shared utilities use any extensively. Not blocking but worth noting.
Inconsistent test function (src/api/providers/__tests__/xai.spec.ts:85) - Uses test() while all others use it().

Will fix all of these.

hannesrudolph · 2026-03-20T00:03:33Z

@roomote fix it all

- Use base provider convertToolSchemaForOpenAI() for tool schema hardening - Handle MCP tools with isMcpTool() (strict: false for MCP, true otherwise) - Respect metadata.tool_choice instead of always using "auto" - Add parallel_tool_calls pass-through from metadata - Fix assistant string content to use output_text format (not input_text) - Remove unused modelInfo param from createUsageNormalizer - Add cacheWriteTokens extraction to usage normalizer - Fix test() -> it() inconsistency in xai.spec.ts - Update tests to match all changes

ghost · 2026-03-20T00:05:18Z

Fixed the reported issues. Restored getModelParams for proper reasoning effort/temperature/maxTokens handling, simplified completePrompt to use output_text, fixed duplicate cacheWriteTokens property in usage normalizer, and added tests for reasoning effort with mini models. All local checks passed.

_{View commit | Revert commit}

…eWriteTokens, simplify completePrompt

carlesso · 2026-03-30T16:19:27Z

Hey @hannesrudolph could you kindly give this another go?

hannesrudolph · 2026-04-07T18:35:12Z

@roomote please re-review this

ghost · 2026-04-07T18:37:01Z

Fixed three issues found during re-review: removed double schema processing in mapResponseTools (was calling convertToolSchemaForOpenAI twice), removed redundant stream: true in the request body, and added URL image source support in convertToResponsesApiInput. All tests and type checks pass.

_{View commit | Revert commit}

hannesrudolph · 2026-04-08T14:46:09Z

@roomote are there any outstanding review issues?

ghost · 2026-04-08T14:47:58Z

No outstanding review issues. The original review raised 7 items (schema hardening, tool_choice pass-through, assistant string format, unused param, missing cacheWriteTokens, any types, test consistency) -- all were addressed in the follow-up commits. CI is green, all 55 tests pass, and TypeScript compiles cleanly.

carlesso · 2026-04-08T15:25:12Z

Manually re-tested with the latest changes, all looks great!

done

carlesso · 2026-03-19T23:21:03Z

Related GitHub Issue

Closes: #11961

Roo Code Task Context (Optional)

N/A

Description

Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt.

New shared utilities (src/api/transform/):

responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider.
responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings.

xAI provider (src/api/providers/xai.ts):

Switch from client.chat.completions.create() to client.responses.create()
Use shared transform utilities for stream handling and input conversion
Enable reasoning traces via include: ["reasoning.encrypted_content"]
System prompt via instructions field, store: false for privacy
completePrompt() also migrated to Responses API

Model updates (packages/types/src/providers/xai.ts):

Add grok-4.20 as the new default model (2M context, $2/$6 pricing)
Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

Test Procedure

Tested locally

Pre-Submission Checklist

Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
Scope: My changes are focused on the linked issue (one major feature/fix per PR).
Self-Review: I have performed a thorough self-review of my code.
Testing: New and/or updated tests have been added to cover my changes (if applicable).
Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

N/A

Documentation Updates

Does this PR necessitate updates to user-facing documentation?

No documentation updates are required.
Yes, documentation updates are required. (Please describe what needs to be updated or link to a PR in the docs repository).

Additional Notes

N/A

Get in Touch

Discord: ecarlesso

Interactively review PR in Roo Code Cloud

…tilities Migrate the xAI provider from the deprecated Chat Completions API to the Responses API, add grok-4.20 as the new default model, and introduce shared transform utilities for Responses API that other providers can adopt. New shared utilities (src/api/transform/): - responses-api-stream.ts: processResponsesApiStream() handles core Responses API stream events (text, reasoning, tool calls, usage) and createUsageNormalizer() provides configurable token/cost extraction. Designed for reuse by openai-native, openai-codex, or any future Responses API provider. - responses-api-input.ts: convertToResponsesApiInput() converts directly from Anthropic message format to Responses API input format, avoiding the intermediate Chat Completions conversion step. Handles input_text, input_image, function_call, and function_call_output mappings. xAI provider (src/api/providers/xai.ts): - Switch from client.chat.completions.create() to client.responses.create() - Use shared transform utilities for stream handling and input conversion - Enable reasoning traces via include: ["reasoning.encrypted_content"] - System prompt via instructions field, store: false for privacy - completePrompt() also migrated to Responses API Model updates (packages/types/src/providers/xai.ts): - Add grok-4.20 as the new default model (2M context, $2/$6 pricing) - Remove grok-4.20-beta-0309-reasoning and grok-4.20-beta-0309-non-reasoning

hannesrudolph · 2026-03-19T23:54:23Z

@roomote do a full review please and leave inline comments.

ghost · 2026-03-19T23:55:58Z

Fixed the reported issues. All local checks (tests, lint, type-check) passed.

_{View commit | Revert commit}

ghost

Good migration to the Responses API overall. The shared transform utilities are a solid pattern. I found several issues that need addressing -- mostly around tool schema handling, missing metadata pass-through, and a bug in the input converter for assistant string content.

Issues found:

mapResponseTools bypasses base provider schema hardening (src/api/providers/xai.ts:52-64) - Doesn't use convertToolSchemaForOpenAI(), doesn't handle MCP tools, sets strict: false for all tools.
metadata?.tool_choice is ignored (src/api/providers/xai.ts:89) - Always uses "auto" instead of respecting caller's tool_choice. Also missing parallel_tool_calls.
Assistant string content uses input_text instead of output_text (src/api/transform/responses-api-input.ts:20-26) - For assistant role with string content, wraps as input_text but should use output_text / message format.
modelInfo parameter unused (src/api/transform/responses-api-stream.ts:113) - createUsageNormalizer accepts modelInfo but never references it.
Missing cacheWriteTokens extraction (src/api/transform/responses-api-stream.ts:133) - Normalizer doesn't extract cache_creation_input_tokens.
Pervasive any types - Both shared utilities use any extensively. Not blocking but worth noting.
Inconsistent test function (src/api/providers/__tests__/xai.spec.ts:85) - Uses test() while all others use it().

Will fix all of these.

hannesrudolph · 2026-03-20T00:03:33Z

@roomote fix it all

- Use base provider convertToolSchemaForOpenAI() for tool schema hardening - Handle MCP tools with isMcpTool() (strict: false for MCP, true otherwise) - Respect metadata.tool_choice instead of always using "auto" - Add parallel_tool_calls pass-through from metadata - Fix assistant string content to use output_text format (not input_text) - Remove unused modelInfo param from createUsageNormalizer - Add cacheWriteTokens extraction to usage normalizer - Fix test() -> it() inconsistency in xai.spec.ts - Update tests to match all changes

ghost · 2026-03-20T00:05:18Z

Fixed the reported issues. Restored getModelParams for proper reasoning effort/temperature/maxTokens handling, simplified completePrompt to use output_text, fixed duplicate cacheWriteTokens property in usage normalizer, and added tests for reasoning effort with mini models. All local checks passed.

_{View commit | Revert commit}

…eWriteTokens, simplify completePrompt

carlesso · 2026-03-30T16:19:27Z

Hey @hannesrudolph could you kindly give this another go?

hannesrudolph · 2026-04-07T18:35:12Z

@roomote please re-review this

ghost · 2026-04-07T18:37:01Z

Fixed three issues found during re-review: removed double schema processing in mapResponseTools (was calling convertToolSchemaForOpenAI twice), removed redundant stream: true in the request body, and added URL image source support in convertToResponsesApiInput. All tests and type checks pass.

_{View commit | Revert commit}

hannesrudolph · 2026-04-08T14:46:09Z

@roomote are there any outstanding review issues?

ghost · 2026-04-08T14:47:58Z

No outstanding review issues. The original review raised 7 items (schema hardening, tool_choice pass-through, assistant string format, unused param, missing cacheWriteTokens, any types, test consistency) -- all were addressed in the follow-up commits. CI is green, all 55 tests pass, and TypeScript compiles cleanly.

carlesso · 2026-04-08T15:25:12Z

Manually re-tested with the latest changes, all looks great!

done

carlesso requested review from cte, daniel-lxs, hannesrudolph, jr and mrubens as code owners March 19, 2026 23:21

github-project-automationBot added this to Roo Code Roadmap Mar 19, 2026

github-project-automationBot moved this to New in Roo Code Roadmap Mar 19, 2026

dosubotBot added size:XXL This PR changes 1000+ lines, ignoring generated files. Enhancement New feature or request labels Mar 19, 2026

ghost previously requested changes Mar 19, 2026
View reviewed changes

Comment threadsrc/api/providers/xai.ts Outdated
Comment threadsrc/api/providers/xai.ts Outdated

fix: restore getModelParams, use base class tool conversion, add cach…
6951133
…eWriteTokens, simplify completePrompt

carlesso force-pushed the feat/xai-responses-api branch 2 times, most recently from 3c1500c to 6951133Compare March 20, 2026 16:43

hannesrudolph approved these changes Apr 8, 2026
View reviewed changes

dosubotBot added the lgtm This PR has been approved by a maintainer label Apr 8, 2026

hannesrudolph merged commit 5432fa2 into RooCodeInc:mainApr 8, 2026
10 checks passed

github-project-automationBot moved this from New to Done in Roo Code Roadmap Apr 8, 2026

github-actionsBot mentioned this pull request Apr 8, 2026
Changeset version bump #12085
Merged