feat(zai): update GLM-5 model specs with correct pricing and token limits#11479
feat(zai): update GLM-5 model specs with correct pricing and token limits#11479roomote[bot] wants to merge 1 commit intomainfrom
Conversation
Review complete -- no issues found. International GLM-5 pricing and token limits match the official docs. Mainland China pricing is proportionally scaled from GLM-4.7 ratios, which is consistent with the existing pattern. All 33 tests pass. Mention @roomote in a comment to request specific changes to this pull request or fix all unresolved issues. |
|
something is wrong with this bot roomote. 2 days for such an easy adjustment and still not promoted. |
|
@roomote please release this |
|
@roomote Approved! Release it now! |
No code issues to fix -- reviewer confirmed the PR is clean. Marked the PR as ready for review. |
roomote
bot
requested review from
cte,
daniel-lxs,
hannesrudolph,
jr and
mrubens
as code owners
dosubot
bot
added
the
size:M
Review complete -- no issues found. International GLM-5 pricing ($1.00/$3.20/$0.20 input/output/cached) and token limits (200k context, 128k max output) match the official docs. Mainland China pricing is proportionally scaled from GLM-4.7 ratios, consistent with the existing pattern. Changeset is correct. Mention @roomote in a comment to request specific changes to this pull request or fix all unresolved issues. |
|
could you please approve and merge. |
|
@roomote: request reviews and merge |
No code changes needed -- the review found no issues. Merging and requesting reviews from maintainers is outside my scope as a fixer agent. A maintainer will need to approve and merge this PR. |
4 participants
Related GitHub Issue
Closes: #11438
Description
This PR attempts to address Issue #11438 by updating the GLM-5 model entries in the Z.ai provider with correct specs from the official docs, as provided by contributors in the issue comments.
The previous PR #11443 had merge conflicts, so this is a fresh implementation on current main.
International (z.ai) GLM-5 changes:
maxTokens: 16,384 -> 128,000 (128k max output per docs)contextWindow: 202,752 -> 200,000 (200k per docs)inputPrice: $0.60 -> $1.00 per 1M tokensoutputPrice: $2.20 -> $3.20 per 1M tokenscacheReadsPrice: $0.11 -> $0.20 per 1M tokensMainland China (bigmodel.cn) GLM-5 changes:
maxTokens: 16,384 -> 128,000contextWindow: 202,752 -> 200,000Features (already correctly configured, unchanged):
Feedback and guidance are welcome.
Test Procedure
npx vitest run api/providers/__tests__/zai.spec.tsfromsrc/directory -- all 33 tests passPre-Submission Checklist
Documentation Updates
Additional Notes
The GLM-5 model entry already existed in the codebase but with placeholder values copied from GLM-4.7. This PR corrects those values using the specs provided by @damianar1984 in the issue comments, sourced from the official z.ai documentation. This replaces the conflicting PR #11443 with a clean implementation on current main.