feat: DeepSeek V4 models, Z.AI/GLM provider, model tags in provider cards#1213
feat: DeepSeek V4 models, Z.AI/GLM provider, model tags in provider cards#1213jasonjcwu wants to merge 6 commits intonesquena:masterfrom
Conversation
Add deepseek-v4-flash and deepseek-v4-pro model entries to: - api/config.py (_MODEL_LIST and _PROVIDER_MODELS) - static/index.html (model dropdown) - static/ui.js (static label map) These are the latest DeepSeek models with 1M context window, replacing the legacy deepseek-chat/deepseek-reasoner (deprecated 2026-07-24).
- Add zai (Z.AI / GLM / 智谱) to onboarding _SUPPORTED_PROVIDER_SETUPS with default model glm-5.1 - Add GLM models (glm-5.1, glm-5, glm-5-turbo, glm-4.x) to _MODEL_LIST for display in model dropdowns - Update DeepSeek default_model from deepseek-chat-v3-0324 to deepseek-v4-flash - Update DeepSeek default_base_url from /v1 to bare domain (API docs change)
- Remove deepseek-chat-v3-0324 (DeepSeek V3) and deepseek-reasoner (R1) from _MODEL_LIST, _PROVIDER_MODELS, static/index.html, and static/ui.js - Keep only deepseek-v4-flash and deepseek-v4-pro - These old model IDs are deprecated since 2026-07-24
Provider card improvements:
- Show model name tags when a provider card is expanded (panels.js)
- Add .provider-card-model-tag styling (style.css)
Custom providers in providers panel:
- Scan config.yaml custom_providers (e.g. glmcode, timicc) and list
them as providers with their configured models (api/providers.py)
- Detect API key status from env var references (${ENV_VAR})
…dels
- Test custom_providers entries (glmcode, deepseek) appear in get_providers()
- Test env var reference detection (${VAR_NAME} pattern)
- Test bare API key, missing key, empty/malformed entries
- Assert DeepSeek V4 models present, V3 deprecated models removed
- Assert GLM model series in _PROVIDER_MODELS and onboarding setup
|
Thanks for the PR, @jasonjcwu! This covers a good chunk of model list maintenance. A few notes before merge: Deprecation date concern — V3 model removal may be premature The commit message states the V3/R1 models are "deprecated since 2026-07-24." Today is 2026-04-28, so that date is still ~3 months away. If the models remain usable until July, removing them now would break any user who has configured Bundled scope — consider splitting This PR touches four independent concerns:
Items 1 and 2 are additive and low-risk. Items 3 and 4 are larger in blast radius. If you'd like faster review, splitting into a "model additions only" PR and a separate "V3 cleanup + card UI" PR would help. Z.AI / GLM details to verify
Unit tests The new test coverage for Happy to help iterate if you want to split this into smaller pieces! |
- Restore deepseek-chat-v3-0324 and deepseek-reasoner with '(legacy)' labels; these are deprecated 2026-07-24 but still live until then - Fix zai (Z.AI/GLM) default_base_url: use /api/paas/v4 instead of /api/coding/paas/v4; the coding plan path is for the glmcode custom provider, not the general API - Update test assertions to match
|
Good catches — fixed: V3 removal: Restored with (legacy) labels. They stay until 2026-07-24. Z.AI base_url: Good catch. The /coding/ path was wrong — that's for the coding-plan endpoint (glmcode custom provider), not the general API. Fixed to https://open.bigmodel.cn/api/paas/v4. glm-5.1: Yes, confirmed — it's the current flagship model ID in upstream hermes-agent's model_metadata.py and _PROVIDER_MODELS['zai']. Tests: Updated assertions + 10 tests all passing. |
|
Thanks for the quick fixes, @jasonjcwu! The review feedback has been addressed cleanly:
Remaining items before merge:
This is shaping up well — appreciate the responsiveness on the review feedback! |
jasonjcwu
changed the title
|
Full test suite — 68 passed, 0 failed |
|
Thanks for sharing the full test run, @jasonjcwu! 68 passed, 0 failed across all five test modules is a clean result. The coverage across |
|
Merged in v0.50.237 via #1243. Thank you @jasonjcwu! 🎉 |
3 participants
4 commits:
Add deepseek-v4-flash and deepseek-v4-pro model entries to _MODEL_LIST, _PROVIDER_MODELS, static dropdown, and label map.
Add zai provider to onboarding _SUPPORTED_PROVIDER_SETUPS with default model glm-5.1 and base URL https://open.bigmodel.cn/api/coding/paas/v4. Add GLM model series (glm-5.1, glm-5, glm-5-turbo, glm-4.7, glm-4.5, glm-4.5-flash) to _MODEL_LIST. Update DeepSeek default model to v4-flash and base_url to bare domain (no /v1).
Clean up deepseek-chat-v3-0324 and deepseek-reasoner from all model lists — these are deprecated since 2026-07-24. Keep only v4-flash and v4-pro.