Skip to content

feat(cline-pass): add glm-5.3-flash - #6053

Open
munim wants to merge 4 commits into
anomalyco:devfrom
munim:add-clinepass-glm-5.3-flash
Open

feat(cline-pass): add glm-5.3-flash#6053
munim wants to merge 4 commits into
anomalyco:devfrom
munim:add-clinepass-glm-5.3-flash

Conversation

@munim

@munim munim commented Sep 1, 2026

Copy link
Copy Markdown
Contributor

Closes #6052.

glm-5.3-flash is available in ClinePass but missing from provider data. Adds providers/cline-pass/models/cline-pass/glm-5.3-flash.toml:

  • base_model = "zhipuai/glm-5.3-flash" (lab entry already exists)
  • reasoning effort low|high|max + interleaved reasoning_content, matching glm-5.3
  • reference pricing matches glm-5.3: $1.4 input / $4.4 output / $0.26 cache read per 1M tokens

bun validate passes.

glm-5.3-flash is available in ClinePass but missing from the provider
data. Adds providers/cline-pass/models/cline-pass/glm-5.3-flash.toml
with base_model pointing at the existing zhipuai lab entry and
reference pricing matching glm-5.3.
@github-actions

github-actions Bot commented Sep 1, 2026

Copy link
Copy Markdown
Contributor

No actionable findings.

@github-actions github-actions Bot added the reviewer: ready Automated review found no actionable items label Sep 1, 2026
$0.075 input / $0.25 output per 1M tokens; cache read $0.015.
@munim

munim commented Sep 1, 2026

Copy link
Copy Markdown
Contributor Author

Updated pricing: $0.075 input / $0.25 output per 1M tokens (cache read $0.015). ee4068e

@github-actions github-actions Bot removed the reviewer: ready Automated review found no actionable items label Sep 1, 2026
@github-actions

github-actions Bot commented Sep 1, 2026

Copy link
Copy Markdown
Contributor

No actionable findings.

@github-actions github-actions Bot added the reviewer: ready Automated review found no actionable items label Sep 1, 2026
$0.15 input / $0.50 output / $0.03 cache read per 1M tokens.
@munim

munim commented Sep 1, 2026

Copy link
Copy Markdown
Contributor Author

Corrected to actual pricing: $0.15 input / $0.50 output / $0.03 cache read per 1M tokens.

@github-actions github-actions Bot removed the reviewer: ready Automated review found no actionable items label Sep 1, 2026
@github-actions

github-actions Bot commented Sep 1, 2026

Copy link
Copy Markdown
Contributor

No actionable findings.

@github-actions github-actions Bot added the reviewer: ready Automated review found no actionable items label Sep 1, 2026
No upstream lab mapping on ClinePass for this slug. Inline the full
model metadata (name/description from the ClinePass catalog, specs
matching Z.ai's GLM-5.3-Flash) instead of base_model.
@munim

munim commented Sep 1, 2026

Copy link
Copy Markdown
Contributor Author

Switched from base_model to a full inline entry: ClinePass lists this slug with name cline-pass/glm-5.3-flash and description "Latest natively multimodal model in the GLM-5 series" and declares no upstream lab mapping, so the provider file now carries the complete metadata (specs match Z.ai's GLM-5.3-Flash: 1M context, 131K output, text/image/video/pdf input, effort low|high|max, always-on reasoning via reasoning_content).

@github-actions github-actions Bot removed the reviewer: ready Automated review found no actionable items label Sep 1, 2026
@github-actions

github-actions Bot commented Sep 1, 2026

Copy link
Copy Markdown
Contributor

Action items

  • [high] [violation] providers/cline-pass/models/cline-pass/glm-5.3-flash.toml:6 - Check: Non-lab hosts must use base_model with override-only fields when the lab model is nameable. Why: ClinePass is a multi-model relay, not the lab. models/zhipuai/glm-5.3-flash.toml already exists, and peer providers/cline-pass/models/cline-pass/glm-5.3.toml correctly uses base_model = "zhipuai/glm-5.3". Patch 4 dropped base_model and inlined full lab metadata (name, description, family, dates, capabilities, [limit], [modalities]) because ClinePass “does not declare an upstream lab mapping.” That is not a valid exception—lack of host-side mapping does not make the model unique-to-host. Action: Restore base_model = "zhipuai/glm-5.3-flash" and keep only provider-specific fields/real deltas (cost, reasoning_options, interleaved, and any true overrides). Remove the restated identical lab fields.
  • [low] [possible mistake] providers/cline-pass/models/cline-pass/glm-5.3-flash.toml:32 - Check: Material cost changes should cite direct provider pricing evidence mapped to the claim. Why: Cost flipped across commits (1.4/4.4/0.260.075/0.25/0.0150.15/0.5/0.03) while the only citation is the general ClinePass page. First-party Zhipuai currently records the promo half-rate (0.075/0.25/0.015) with list 0.15/0.50/0.03, so the final numbers need a clear ClinePass-specific source. Action: Confirm the ClinePass reference rates for this model ID and note in the leading comment (or PR body) which doc/section supports $0.15 / $0.50 / $0.03 per 1M tokens.

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.

cline-pass: glm-5.3-flash missing from provider data

1 participant