Skip to content

fix(providers): rename deepseek-v4-flash to deepseek-flash - #12269

Open
michaelneale wants to merge 2 commits into
mainfrom
fix-deepseek-model-names
Open

michaelneale wants to merge 2 commits into
mainfrom
fix-deepseek-model-names

Conversation

@michaelneale

Copy link
Copy Markdown
Collaborator

Closes #12218

DeepSeek retired the deepseek-v4-flash model name — requests to it are now served by DeepSeek-V4.1-Flash, and the current name is deepseek-flash (per DeepSeek's deprecation notice quoted in the issue). Since the bundled deepseek.json definition is compiled into the binary, the desktop model picker keeps showing the stale name until a release ships this update.

Changes

  • crates/goose-providers/src/declarative/definitions/deepseek.json: deepseek-v4-flashdeepseek-flash.

Out of scope: gateway/aggregator definitions (fireworks, empiriolabs, opencode, etc.) expose their own model IDs, and DeepSeek still accepts the legacy names, so nothing breaks there.

Verification

  • cargo test -p goose-providers passes (definitions are validated at load).
  • Legacy name remains accepted by DeepSeek's API for existing configs.

DeepSeek retired the deepseek-v4-flash name; requests are now served by
DeepSeek-V4.1-Flash under the deepseek-flash model name. Update the
bundled provider definition so the desktop model list and default show
the current name.

Closes #12218
@michaelneale
michaelneale requested a review from a team as a code owner September 21, 2026 02:21

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 6c646d822c

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

Comment thread crates/goose-providers/src/declarative/definitions/deepseek.json Outdated
@michaelneale

Copy link
Copy Markdown
Collaborator Author

🤖 Addressed the codex feedback: dropped the stale 128K context_limit on deepseek-flash so the canonical metadata (1M context) applies via ContextLimitResolver instead of being overridden.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Update deepseek v4 flash model

1 participant