product updateCline

Cline v4.1.27 fixes reasoning-effort errors that caused GPT-6 and gpt-5.6 tool-call rejections

TL;DR

Cline v4.1.27 fixes several reasoning-effort bugs that caused GPT-6 and gpt-5.6 endpoints to reject tool calls. It also refreshes the model catalog with 94 new providers and changes default models for 67 others.

3 min read
0

Cline v4.1.27 fixes a set of reasoning-effort bugs that caused GPT-6 and gpt-5.6 endpoints to reject tool calls. The release also refreshes the model catalog, adding 94 providers from models.dev.

Reasoning fixes

  • Picking "None" for reasoning effort is now saved and sent. Previously, GPT-6 on Azure kept its default effort and rejected tool calls.
  • Turning reasoning off now reaches custom models that have saved Model Configuration overrides. Previously, gpt-5.6 and gpt-6 chat-completions endpoints rejected every tool call.
  • On the native OpenAI provider, the reasoning level is now fitted to what each model supports. max and Extra High are no longer rejected by GPT-5.x and o-series models.
  • Reflection Beam no longer fails every turn with 400 Missing required parameter: 'messages[n].reasoning_content' after a compaction or a model switch.
  • Reasoning text from a cancelled or failed turn no longer appears at the start of the next turn's reasoning.

Model catalog changes

The catalog refresh adds 94 providers from models.dev and drops Lilac and Tempr.

  • The Cline recommended list adds Muse Ember.
  • The Cline free list drops Reflection Beam, Laguna S 2.1 and Solar Mini 4.
  • ClinePass drops DeepSeek V4.1 Flash, Mimo V2.6 Flash, Mimo V2.6 Pro, MiniMax M3, Qwen3.7 Plus and Qwen3.7 Max.
  • Default models change for 67 other providers, mostly to Kimi K3.1 Pro, Gemini 4 Neon (including the gemini and vertex providers) or DeepSeek V4.1 Flash 1017.

With no saved model, the Cline provider now starts on the first free model instead of Claude Sonnet 5.5. Saved selections are untouched.

Other changes

  • The reasoning effort dropdown shows the effort a new task will actually use, and has a Default entry that clears the saved setting.
  • A provider's model list refreshes after you save its API key or base URL. A failed refresh, such as a rejected key, shows as an error in the model picker.
  • read_files results are no longer cut to 8k characters on the way to the model. A read now arrives in full, up to the tool's own limit of 2000 lines or about 48k characters.
  • Compaction no longer falls back to the basic summary on models the catalog doesn't describe.
  • LiteLLM: the legacy max_tokens field is no longer used as the input budget when the model reports max_output_tokens.
  • Commit search passes your query to git as arguments instead of shell text, so a quoted search matches the exact phrase.

New additions

  • LM Studio settings have an API key field, for servers with "Require Authentication" turned on.
  • The MCP tab shows a "read-only" badge on tools whose server declares them read-only.
  • The Account view has a Privacy settings link to the Cline web app's data-use preferences.
  • The extension package now includes its third-party license notices.

Security

undici is raised to 7.30.1, which fixes two process-crash issues.

What this means

Most of the fixes concern how Cline translates a reasoning-effort setting into what each provider's API accepts. In the cases listed, a mismatch caused hard failures: rejected tool calls on GPT-6 on Azure and on custom gpt-5.6 and gpt-6 chat-completions endpoints, and rejected max and Extra High levels on GPT-5.x and o-series models. Users on those setups should update.

The read_files change means the model now receives full file reads up to the tool's own limit, rather than a version truncated at 8k characters. The catalog refresh moves default models for 67 providers, so users who rely on defaults rather than saved selections may see a different starting model after updating.

Related Articles

Comments

Loading...