Repository navigation
fix(models): mark syn:* aliases as floating, correct stale served-model labels (#26) - #27
Merged
Merged
Conversation
…el labels (#26) Synthetic re-points syn:* aliases server-side without renaming. Verified against the live API 2026-09-29: - synthetic (syn:large:text): GLM-5.2 -> deepseek-ai/DeepSeek-V4.1-Flash - syn-kimi (syn:large:vision): Kimi-K2.7-Code -> moonshotai/Kimi-K3 - syn-qwen (syn:small:vision): Qwen3.6-27B -> Qwen/Qwen3.8-27B - syn-flash (syn:small:text): correct (GLM-4.7-Flash), marked floating All four entries now carry floating-alias documentation, the generated tables point users to the hf:...-pinned shorthands for stable identity, and the Synthetic note documents the chat-completion probe as the re-verification path. Regenerated markdown; manage-models.py check green. Bump version to 0.40.13.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Fixes #26 via the durable option:
syn:*entries are now documented as floating aliases with the currently-verified binding, instead of hard labels that rot the next time Synthetic re-points them.Verified against the live API 2026-09-29 (2 calls per alias, stable):
synthetic(syn:large:text)deepseek-ai/DeepSeek-V4.1-Flashsyn-kimi(syn:large:vision)moonshotai/Kimi-K3syn-qwen(syn:small:vision)Qwen/Qwen3.8-27Bsyn-flash(syn:small:text)Changes
scripts/models.json: all foursyn:*entries now declare floating-alias semantics indocs.actual_model+ descriptions, direct users to thehf:...-pinned shorthands (syn-ds41-flash,syn-kimi3,syn-glm53-flash) where model identity matters, and note the probe (chat-completion responsemodelfield) as the re-verification path.scripts/manage-models.py: the generated Synthetic note explains floating aliases and the re-verify probe (was still showing "Synthetic GLM-5.2" in its example).CLAUDE.md,README.md,commands/*.md,skills/prompt-executor/SKILL.md);manage-models.py checkgreen.Tests
pytest skills/cli-detector/tests/test_router.py skills/prompt-executor/tests/— 421 passed.python3 scripts/manage-models.py check— in sync.