Skip to content

fix: enable reasoning visibility for DeepSeek V3.1 and GLM-4.5 models #7372

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

Closed
wants to merge 1 commit into from
Closed

fix: enable reasoning visibility for DeepSeek V3.1 and GLM-4.5 models #7372

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
2 changes: 2 additions & 0 deletions packages/types/src/providers/deepseek.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@ export const deepSeekModels = {
contextWindow: 128_000,
supportsImages: false,
supportsPromptCache: true,
supportsReasoningEffort: true, // DeepSeek V3.1 supports thinking/reasoning
Copy link
Contributor Author

@roomote roomote bot Aug 24, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The implementation looks correct and follows the existing pattern. However, would it be helpful to add test coverage for the supportsReasoningEffort flag? The existing tests in src/api/providers/__tests__/deepseek.spec.ts could verify this capability is properly set.

inputPrice: 0.27, // $0.27 per million tokens (cache miss)
outputPrice: 1.1, // $1.10 per million tokens
cacheWritesPrice: 0.27, // $0.27 per million tokens (cache miss)
Expand All @@ -22,6 +23,7 @@ export const deepSeekModels = {
contextWindow: 128_000,
supportsImages: false,
supportsPromptCache: true,
supportsReasoningEffort: true, // DeepSeek-R1 supports thinking/reasoning
Copy link
Contributor Author

@roomote roomote bot Aug 24, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Minor suggestion: For consistency, consider standardizing the comment wording across all models. Currently using "DeepSeek V3.1 supports thinking/reasoning" here and "DeepSeek-R1 supports thinking/reasoning" below. Perhaps just "supports reasoning" would be cleaner?

inputPrice: 0.55, // $0.55 per million tokens (cache miss)
outputPrice: 2.19, // $2.19 per million tokens
cacheWritesPrice: 0.55, // $0.55 per million tokens (cache miss)
Expand Down
4 changes: 4 additions & 0 deletions packages/types/src/providers/zai.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,7 @@ export const internationalZAiModels = {
contextWindow: 131_072,
supportsImages: false,
supportsPromptCache: true,
supportsReasoningEffort: true, // GLM-4.5 supports thinking/reasoning
Copy link
Contributor Author

@roomote roomote bot Aug 24, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good to see both international and mainland configurations updated consistently. Similar to the DeepSeek models, consider adding test assertions in src/api/providers/__tests__/zai.spec.ts to verify the supportsReasoningEffort flag is properly set for these models.

inputPrice: 0.6,
outputPrice: 2.2,
cacheWritesPrice: 0,
Expand All @@ -24,6 +25,7 @@ export const internationalZAiModels = {
contextWindow: 131_072,
supportsImages: false,
supportsPromptCache: true,
supportsReasoningEffort: true, // GLM-4.5-Air supports thinking/reasoning
inputPrice: 0.2,
outputPrice: 1.1,
cacheWritesPrice: 0,
Expand All @@ -41,6 +43,7 @@ export const mainlandZAiModels = {
contextWindow: 131_072,
supportsImages: false,
supportsPromptCache: true,
supportsReasoningEffort: true, // GLM-4.5 supports thinking/reasoning
inputPrice: 0.29,
outputPrice: 1.14,
cacheWritesPrice: 0,
Expand Down Expand Up @@ -73,6 +76,7 @@ export const mainlandZAiModels = {
contextWindow: 131_072,
supportsImages: false,
supportsPromptCache: true,
supportsReasoningEffort: true, // GLM-4.5-Air supports thinking/reasoning
inputPrice: 0.1,
outputPrice: 0.6,
cacheWritesPrice: 0,
Expand Down
Loading