This release includes 3 breaking changes for platform teams planning a safe upgrade.
✓ No known CVEs patched in this version
Topics
ReleasePort's take
Light signalClaude Fable 5 is removed from the Anthropic picker due to a US export ban. Z.AI drops glm‑5.1 and glm‑5, retaining only glm‑5.2 and glm‑5‑turbo. OpenAI removes gpt‑5.4‑nano while keeping 5.5, 5.4, and 5.4‑mini. Editor clients now display API retry/backoff messages instead of an endless "Thinking…" spinner.
Why it matters: Deprecations affect model selection UI for Anthropic (Claude Fable 5), Z.AI (glm‑5.1, glm‑5), and OpenAI (gpt‑5.4‑nano); update integration configs accordingly. The bug fix improves user experience by showing retry/backoff status rather than a stuck spinner.
Summary
AI summaryModel picker cleanup removes Claude Fable 5, several older variants, and fixes ACP retry visibility to show backoff messages instead of an endless spinner.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Deprecation | Medium |
Claude Fable 5 removed from Anthropic picker (US export ban). Claude Fable 5 removed from Anthropic picker (US export ban). Source: llm_adapter@2026-06-15 Confidence: high |
— |
| Deprecation | Medium |
Z.AI drops models glm‑5.1 and glm‑5 (keeps glm‑5.2 + glm‑5‑turbo). Z.AI drops models glm‑5.1 and glm‑5 (keeps glm‑5.2 + glm‑5‑turbo). Source: llm_adapter@2026-06-15 Confidence: high |
— |
| Deprecation | Medium |
OpenAI drops gpt‑5.4‑nano (keeps 5.5, 5.4, 5.4‑mini). OpenAI drops gpt‑5.4‑nano (keeps 5.5, 5.4, 5.4‑mini). Source: llm_adapter@2026-06-15 Confidence: high |
— |
| Bugfix | Medium |
Editor clients now show API retry/backoff messages instead of endless "Thinking…" spinner. Editor clients now show API retry/backoff messages instead of endless "Thinking…" spinner. Source: llm_adapter@2026-06-15 Confidence: high |
— |
Full changelog
Trimmed the model pickers (Claude Fable 5 is de-listed — unavailable under the US export ban — and a few older variants drop off), and editor clients (VS Code, Zed) now see API retry/backoff instead of an endless "Thinking…" spinner.
Changed
- Model picker cleanup across every provider. Claude Fable 5 is removed
from the Anthropic picker (unavailable under the US export ban; Opus 4.8 stays
the default). Z.AI dropsglm-5.1andglm-5(keepsglm-5.2+glm-5-turbo);
OpenAI dropsgpt-5.4-nano(keeps 5.5 / 5.4 / 5.4-mini). DeepSeek, Google, and
MiniMax are unchanged. All ids remain valid if set by hand — they're just no
longer offered. Context/cost tables updated to match.
Fixed
- ACP retry visibility. When a request hit a transient API error and the
agent retried with backoff, the ACP path dropped the notice (only the bare
iteration counter was suppressed, but the retry message went with it) — so
editor clients showed an indefinite "Thinking…" while the CLI was actually
retrying. Retry/backoff notices ("API 429 … retrying in 10s (1/3)") and
context warnings (⚠) are now forwarded as agent thoughts; the plain
iteration counter stays internal.
Breaking Changes
- Claude Fable 5 removed from Anthropic picker (US export ban)
- Z.AI drops glm-5.1 and glm-5
- OpenAI drops gpt-5.4-nano
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About Codeep
All releases →Related context
Related tools
Earlier breaking changes
- v2.8.0 `codeep account push` and `account sync` no longer transfer API keys unless `/keysync on` is enabled
- v2.4.1 MiniMax M3 replaces MiniMax-M2.7 as default model across all providers.
- v2.0.0 McpServer protocol now optional fields `command`, `args`, plus new `url` and `headers`; version bumped to 2.0.0.
Beta — feedback welcome: [email protected]