v4.1.272026-11-21
Added
- 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.
Changed
- With no saved model, the Cline provider now starts on the first free model instead of Claude Sonnet 5.5. Saved selections are untouched.
- 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, and 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.
- Refreshed the model catalog. 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.
Fixed
- Picking "None" for reasoning effort is now saved and sent. Before, 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. Before, gpt-5.6 and gpt-6 chat-completions endpoints rejected every tool call.
- On the native OpenAI provider, the reasoning level is fitted to what each model supports, so 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.
- Compaction no longer falls back to the basic summary on models the catalog doesn't describe.
- LiteLLM: the legacy maxtokens field is no longer used as the input budget when the model reports maxoutput_tokens.
- Reasoning text from a turn that was cancelled or failed no longer shows up at the start of the next turn's reasoning.
- Commit search passes your query to git as arguments instead of shell text, so a quoted search matches the exact phrase.
- Security: undici raised to 7.30.1, which fixes two process-crash issues.
Full Changelog: https://github.com/cline/cline/compare/v4.1.26...v4.1.27
Desktop v0.0.502026-11-21
- The thinking picker offers Max reasoning effort, for models whose top level is max
- A manual /compact can be stopped. Before, it could sit on "Thinking…" with no Stop button when the summarizer stalled
- With no saved model, the Cline provider now starts on the first free model instead of Claude Sonnet 5.5. Saved selections are untouched
- The Account view has a Privacy settings link to the Cline web app's data-use preferences
- Settings → About has a Third-party notices entry with the licenses for the open source software included in Cline
- On macOS, Check for Updates from the menu runs once instead of twice
- Refreshed colours and typography across the app
- 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
- Reflection Beam no longer fails every turn with 400 Missing required parameter: 'messages[n].reasoning_content' after a compaction or a model switch
- Compaction no longer falls back to the basic summary on models the catalog doesn't describe. The summarizer was sending reasoning_effort: "none", which some endpoints reject
- Turning reasoning off now reaches custom models that have saved Model Configuration overrides. Before, gpt-5.6 and gpt-6 chat-completions endpoints rejected every tool call
- On the native OpenAI provider, the reasoning level is fitted to what each model supports, so max and Extra High are no longer rejected by GPT-5.x and o-series models
- LiteLLM: the legacy maxtokens field is no longer used as the input budget when the model reports maxoutput_tokens
- Refreshed the model catalog. 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
Full Changelog: https://github.com/cline/cline/compare/desktop-v0.0.49...desktop-v0.0.50
CLI v3.0.752026-11-21
- One-shot runs (a prompt argument or piped stdin) now start in yolo mode instead of act mode, since nobody is at the keyboard to approve tools. Act mode is kept when auto-approve is off (--auto-approve false or the saved setting), when a mode flag is passed, when plan mode is saved, or when the prompt is a /team prompt. Yolo mode has no team tools, so pass --act when you need them
- --timeout and Ctrl+C now stop the first turn of a one-shot run. Before, neither could interrupt it
- The CLI no longer reads .env or bunfig.toml from the directory it is launched in
- With no saved model, the Cline provider now starts on the first free model instead of Claude Sonnet 5.5. Saved selections are untouched
- /account shows a Privacy settings link to the Cline web app's data-use preferences
- 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
- Reflection Beam no longer fails every turn with 400 Missing required parameter: 'messages[n].reasoning_content' after a compaction or a model switch
- Compaction no longer falls back to the basic summary on models the catalog doesn't describe. The summarizer was sending reasoning_effort: "none", which some endpoints reject
- Turning reasoning off now reaches custom models that have saved Model Configuration overrides. Before, gpt-5.6 and gpt-6 chat-completions endpoints rejected every tool call
- On the native OpenAI provider, the reasoning level is fitted to what each model supports, so max and Extra High are no longer rejected by GPT-5.x and o-series models
- LiteLLM: the legacy maxtokens field is no longer used as the input budget when the model reports maxoutput_tokens
- Removed the CLINECAPTURE debug variables, which wrote full provider request bodies to disk
- Experimental: set CLINECLOUDSESSIONS=1 to run tasks in Cline Cloud from the TUI. Ctrl+O toggles Local/Cloud, Ctrl+K picks the repo and branch, and /model picks the cloud model
- Each platform package now ships a THIRDPARTYNOTICES.txt
- Security: undici raised to 7.30.1, which fixes two process-crash issues
- Refreshed the model catalog. 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
Full Changelog: https://github.com/cline/cline/compare/cli-v3.0.74...cli-v3.0.75
SDK v0.2.12026-11-21
- read_files results are no longer re-cut to the 8k tool-result cap on the way to the provider. They use the 50k file-content cap, so a read the tool reported in full arrives in full. Command, search, MCP and custom tool results keep the 8k cap, and an explicit tool-result cap still applies
- Models that need interleaved reasoning get a placeholder reasoningcontent on assistant messages that have none. This fixes Reflection Beam's 400 Missing required parameter: 'messages[n].reasoningcontent' after compaction or a model switch. The fact is carried as ModelInfo.metadata.interleavedReasoningField
- The summarizer now states a preference to go without reasoning instead of disabling it outright. The gateway disables it only for models whose catalog entry advertises an off option. Before, uncatalogued models got reasoning_effort: "none", which some endpoints reject, and compaction fell back to basic
- A custom model with saved overrides now has reasoning seeded into its capability list unless it sets supportsReasoning: false, so an explicit reasoning disable reaches the provider. Before, gpt-5.6 and gpt-6 chat-completions endpoints rejected every tool call
- The native OpenAI adapter fits the reasoning level to the model's advertised effort ladder, so max and Extra High are no longer sent verbatim to GPT-5.x and o-series models that reject them
- LiteLLM: the legacy maxtokens is no longer used as the input budget when maxoutput_tokens is set. Null limits are treated as missing
- The cline provider's bundled default is now the first bundled free model. resolveClineDefaultModelId() in @cline/core returns the first free model from the live recommended feed, then the bundled one, then CLINEDEFAULTMODEL_ID
- CLINEDEFAULTMODELID is now anthropic/claude-sonnet-5.5, and the remaining hardcoded default-model literals use it. @cline/shared adds ANTHROPICDEFAULTMODELID for Anthropic's bare id format
- A running compaction can be stopped, and cancellations are counted in telemetry
- Session config and the hub RPC schema accept the max reasoning effort
- With tools and spawning enabled, core registers uxverifier, which delegates an ordered checklist to a subagent through the host's spawnagent executor. Disabling spawn_agent disables it too
- createUxVerificationReportTool() creates the opt-in reportuxverification tool for structured reports with captions, pass/fail/blocked results and screenshot references
- createPublishArtifactsTool({ publish }) creates the opt-in publish_artifacts tool. createArtifactPublisher from @cline/core/cloud publishes to cloud sessions through signed upload policies. Artifact IDs default to the source filename
- McpToolDescriptor.annotations carries a server's tool hints (readOnlyHint and friends) through the stdio and URL clients and the manager cache. task.tool_used telemetry gains mode
- session.resume starts a runtime for a saved session ID and reads history on the hub, instead of requiring initialMessages
- Cloud task creation accepts initialMessage to create and send in one call, and localGit to launch from a captured Git snapshot
- HubCloudSessionClient.getArtifactDownloadUrl(sessionId, artifactId) returns a signed URL without starting the sandbox
- Provider, model, MCP and MCP OAuth management are available as hub commands, with token authentication required for provider management
- The hub can return session history in bounded chunks and archived cloud history
- The hub validates caller accounts for cloud commands, and no longer re-sends a rejected Cline refresh token on every cloud poll
- Hub cloud requests no longer duplicate the API prefix
- Hub session handlers are retained when a session deletion fails
- Confirmed cloud handoff publication is exported from @cline/core/cloud for direct hosts
- @cline/shared adds resolveClineAppUrl, and the account response carries dataPrivacyPath for the web app's privacy settings page
- @cline/ui adds SectionHeading, Banner, ItemCard, DetailList, controlled Dialog components and a UX verification report carousel with screenshot downloads. Buttons, badges and switches are refreshed (loading and pending states, a small switch size), along with the colour tokens and a new text-xxs step
- Removed the CLINECAPTURE provider request capture, which wrote full request bodies to disk
- Every SDK package now includes its license file
- Refreshed the model catalog (6,457 → 7,621 models). 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, so the cline provider's bundled default is Nemotron 3.5 Super VL. 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. Qwen models now report tiered families such as qwen-plus and qwen-max
Full Changelog: https://github.com/cline/cline/compare/sdk/sdk/v0.2.0...sdk/sdk/v0.2.1
v4.1.262026-10-26
Added
- Cloud tasks: Settings > API Configuration has a Local | Cloud toggle when cloud sessions are on. Cloud picks the Cline model new cloud tasks run on, saved separately from your local models. Clicking the model label in a cloud session's composer opens it there.
- Starting or resuming a cloud sandbox shows a progress card with the repo, branch and steps (start sandbox, clone repository, start agent). It collapses once the agent starts.
Changed
- Cloud sessions hide the Plan/Act toggle and the auto-approve bar instead of showing them disabled.
- With no model chosen, the Cline provider uses Claude Sonnet 5.5 again, which is back at the top of Cline's recommended list.
- Model requests retry at most 3 times instead of 5.
- Kimi K3 now gets max reasoning for Extra High and Max, instead of an xhigh setting it rejects.
- Refreshed the model catalog. Adds the CommandCode, Reflection, Liquid Inference and Munito AI providers and drops CrofAI, iFlow and Llama. The Cline free list adds Reflection Beam, Nemotron 3.5 Lightning and Laguna S 2.1 and drops Mimo V2.6 Flash and Muse Spark 1.3 Contributor. ClinePass now defaults to Kimi K3.1 Pro and adds MiniMax M3.1 Flash and DeepSeek V4.1 Flash 1017. Default models change for most other providers, mostly to Kimi K3.1 Pro, MiniMax M3.1 Flash or DeepSeek V4.1 Flash 1017. If you use a provider without pinning a model, expect a different default.
Fixed
- GPT-6 models (such as gpt-6-sol and gpt-6.1-sol) work on OpenAI and Azure OpenAI-compatible endpoints, and SAP AI Core no longer rejects requests with "Unsupported parameter: 'maxtokens'". Both now get maxcompletion_tokens.
- Gemini API keys (the gemini provider) no longer send native Google Search. Free-tier keys failed every turn with a 429 quota error on Gemini 3+, and on Gemini 2.x every Cline tool was dropped.
- Toggling Plan/Act while a task is starting no longer drops your first prompt, and the session no longer keeps running with the other mode's tools.
- Disabled rules stay disabled.
- Starting a command no longer moves keyboard focus into the terminal, so text you're typing in the chat no longer lands in the shell.
- Cancelling a task or opening another one while Cline is asking a question no longer lets your next message get taken as the answer to the old question.
- A turn whose connection drops after the model has streamed only reasoning is now retried instead of ending the task.
- Auto-compaction no longer runs for models with an unknown context limit, and it no longer runs again every few tool calls after the first compaction.
- ask_question: the model keeps going after you pick an option, and empty answers are rejected.
- run_commands with an empty command list now returns an error instead of a silent success. On PowerShell, the tool tells the model how quote escaping works and to confirm recursive deletes first.
- Task header badges wrap instead of hiding the close button, the GitHub connection check no longer flashes a loading indicator, and the Local/Cloud toggle's padding is fixed.
Full Changelog: https://github.com/cline/cline/compare/v4.1.25...v4.1.26
Desktop v0.0.492026-10-26
- A new Changes button in the session header shows the session's line counts and opens a pane listing the files changed in the last turn or across the whole session, with a diff of each. When a turn finishes, a summary under it lists the files it changed; click it to open the pane on that turn
- A project files panel (folder button in the session header) shows a file tree beside the conversation, with a read-only tabbed viewer, git status letters, and a marker on files the session changed. Files the agent edited get a File/Diff toggle. Works on local and SSH environments
- An integrated terminal panel opens at the bottom of the window (toggle with Ctrl+backtick)
- Pull request status moved from the bar above the composer into a header button. Its menu shows the PR title, merge status and checks (failing checks first), and lets you open the PR on GitHub, create a PR, copy the branch name, or refresh
- A new Local Models settings page manages Ollama and LM Studio models: setup guidance, model details and storage, and a composer chip showing how much memory the loaded model uses
- Keybindings are configurable in Settings (also reachable from the app and Help menus). Quit can require holding the shortcut, 1 second by default on macOS, so it's harder to quit by accident
- Slash commands and @file mentions show as chips in the composer and in sent messages
- If the Hub fails to start, the app now shows a startup screen with Retry and Export diagnostics, instead of failing every action with "desktop backend endpoint not ready"
- General settings shows the Hub's status and has a Restart Hub button
- Opening a task that's already running in another Cline app now attaches to that run instead of starting a second one that couldn't be stopped. Requests that were waiting on an app that closed mid-task now fail with a clear message instead of hanging
- Sessions with a very large transcript (over about 8 MB) can be resumed again. Before, the transcript was sent to the Hub twice, which went over its message size limit
- Long lines in the file and diff viewers wrap, and rows always span the full width when scrolled
- New Cloud chats start on your local Cline or ClinePass model. If Cloud doesn't offer that model, ClinePass subscribers get a ClinePass model and everyone else gets Claude Sonnet 5.5
- A pasted Cline API key now replaces an old browser sign-in instead of being ignored. Provider settings fields stack properly in the narrow side panel
- The Sessions/Projects sort choice in the sidebar is remembered across restarts
- On Linux desktops without a tray (stock GNOME on Fedora or Debian), closing the window quits the app instead of leaving it running with no way to quit
- Internal messages meant only for the model no longer show in the transcript, and live session titles no longer show the raw user_input wrapper
- Schedule status stays current and failed runs show the full error
- Checking for changes no longer spawns thousands of git processes in workspaces with many untracked files
- SSH environments behind a ProxyJump alias now connect
- The update toast's Restart button sits under the message text
- A turn whose connection drops after the model has streamed only reasoning is now retried instead of ending the run
- Model requests retry at most 3 times instead of 5
- Kimi K3 now gets max reasoning for Extra High and Max, instead of an xhigh setting it rejects
- Gemini API keys (the gemini provider) no longer send native Google Search. Free-tier keys failed every turn with a 429 quota error on Gemini 3+, and on Gemini 2.x every Cline tool was dropped
- Auto-compaction no longer runs for models with an unknown context limit, and it no longer runs again every few tool calls after the first compaction
- ask_question: the model keeps going after you pick an option, empty answers are rejected, and an answer with only an attachment counts
- run_commands with an empty command list now returns an error instead of a silent success. On PowerShell, the tool tells the model how quote escaping works and to confirm recursive deletes first
- Refreshed the model catalog. Adds the Liquid Inference and Munito AI providers and drops CrofAI, iFlow and Llama. The Cline free list adds Reflection Beam, Nemotron 3.5 Lightning and Laguna S 2.1 and drops Mimo V2.6 Flash and Muse Spark 1.3 Contributor. ClinePass now defaults to Kimi K3.1 Pro and adds MiniMax M3.1 Flash and DeepSeek V4.1 Flash 1017. Default models change for 56 other providers, mostly to Kimi K3.1 Pro, MiniMax M3.1 Flash or DeepSeek V4.1 Flash 1017
Full Changelog: https://github.com/cline/cline/compare/desktop-v0.0.48...desktop-v0.0.49
CLI v3.0.742026-10-26
- Opening a task that's already running in another Cline app now attaches to that run instead of starting a second one that couldn't be stopped. If an app closes mid-task, requests waiting on it now fail with a clear message instead of hanging
- cline hub status and cline doctor no longer hang on a Hub that has frozen. Its health check now times out after 3s and the Hub is reported as not running
- A turn whose connection drops after the model has streamed only reasoning (for example, "The socket connection was closed unexpectedly") is now retried instead of ending the run
- Model requests retry at most 3 times instead of 5
- Kimi K3 now gets max reasoning for Extra High and Max, instead of an xhigh setting it rejects
- Gemini API keys (the gemini provider) no longer send native Google Search. Free-tier keys failed every turn with a 429 quota error on Gemini 3+, and on Gemini 2.x every Cline tool was dropped
- Auto-compaction no longer runs for models with an unknown context limit, and it no longer runs again every few tool calls after the first compaction
- ask_question: the model keeps going after you pick an option, empty answers are rejected, and an answer with only an attachment counts
- run_commands with an empty command list now returns an error instead of a silent success. On PowerShell, the tool tells the model how quote escaping works and to confirm recursive deletes first
- Plugin slash commands run in the active session
- Compiled CLI binaries load plugins and serve the dashboard from resources embedded in the binary
- ACP: the provider and account options now appear under the right config categories
- The skills tool call label no longer shows null arguments
- Refreshed the model catalog. Adds the Liquid Inference and Munito AI providers and drops CrofAI, iFlow and Llama. The Cline free list adds Reflection Beam, Nemotron 3.5 Lightning and Laguna S 2.1 and drops Mimo V2.6 Flash and Muse Spark 1.3 Contributor. ClinePass now defaults to Kimi K3.1 Pro and adds MiniMax M3.1 Flash and DeepSeek V4.1 Flash 1017. Default models change for 56 other providers, mostly to Kimi K3.1 Pro, MiniMax M3.1 Flash or DeepSeek V4.1 Flash 1017
Full Changelog: https://github.com/cline/cline/compare/cli-v3.0.73...cli-v3.0.74
SDK v0.2.02026-10-26
- Starting a session that already has a run in progress now reattaches to the live session instead of replacing its runtime. Before, the original run kept going but couldn't be reached, and its transcript writes landed on a stale snapshot. A start that includes a prompt while a run is in progress is rejected
- The hub records every client that registered a capability for a session, including clients that join a task that's already running, and sends each request to the first one still connected. If none is left, the request fails quickly with a message the user can act on, instead of waiting forever
- A turn whose connection drops after the model has streamed only reasoning is now retried, and the partial reasoning is discarded. Turns with text, media or tool calls are still never retried
- Model requests retry at most 3 times after the first attempt (down from 5). Failed attempts are recorded in Langfuse traces, and cancelled requests are no longer reported as Langfuse errors
- AI SDK updated to ai 7.0.136 / @ai-sdk/provider-utils 5.0.58. max reasoning now reaches the provider: the gateway no longer turns it into xhigh, and Kimi K3 gets max for Extra High and Max. Empty Anthropic turns are still retried despite the new anthropic.* custom stream parts
- The gemini provider no longer sends native Google Search. On Gemini 3+, free-tier keys failed every turn with a 429 quota error, and on Gemini 2.x it dropped every Cline tool. Vertex keeps native search
- Auto-compaction no longer runs when the model's input limit is unknown. Manual and overflow-recovery compaction still work
- After the first compaction, the size estimate now uses the messages as they are actually sent, with tool results truncated. Before, it overestimated about 7x and compaction ran again every few tool calls
- Checkpoint diffs read file contents with a single git cat-file --batch process instead of up to two git show processes per file. Workspaces with many untracked files used to spawn thousands of git processes
- SSH remotes work behind a ProxyJump alias. The connect budget goes from 10s to 30s, the per-command limit to 60s, and tunnel readiness to 40s. When ssh exits mid-upload, its real error is reported instead of EPIPE, and a LocalForward on the alias no longer collides with the tunnel
- ask_question answers are framed so the model keeps going after a bare choice. Empty answers are rejected, and an answer with only an attachment counts as answered
- run_commands rejects an empty commands list with a tool error, and its schema now advertises minItems: 1
- On PowerShell, the run_commands description explains backtick escaping, single-quoting literal paths, preferring cmdlets over cmd /c, and confirming recursive deletes first
- Plugin commands run in the active session runtime and the hub. Plugin watches start working again when discovery directories appear
- Compiled binaries load embedded plugin and dashboard resources straight from the binary, and every declared SDK runtime export is packaged
- A configured CLINEPLUGINRUNTIME_DIR is resolved to an absolute path
- When OpenRouter has no match for a Cline recommended model, its capabilities are filled in from the Vercel catalog
- Desktop cloud session orchestration moved into the hub. session.list reports activeCloudCommands, and the hub isn't replaced while cloud commands are still running
- Core adds conversation summaries and cloud continuation support
- Run stall telemetry: agent.run-finished and agent.run-failed carry stall signals from a new RunHealthObserver
- @cline/ui now ships with the SDK packages on the shared version. It adds a Kbd component and token chips for slash commands and file mentions in AgentComposer
- Refreshed the model catalog (7,041 → 6,457 models). Adds the Liquid Inference, Munito AI and Munito AI (EU) providers and drops CrofAI, iFlow and Llama. The Cline free list adds Reflection Beam, Nemotron 3.5 Lightning and Laguna S 2.1 and drops Mimo V2.6 Flash and Muse Spark 1.3 Contributor. ClinePass adds Kimi K3.1 Pro (now its default), MiniMax M3.1 Flash and DeepSeek V4.1 Flash 1017, and drops Mimo V2.5 and V2.5 Pro. Default models change for 56 other providers, mostly to Kimi K3.1 Pro, MiniMax M3.1 Flash or DeepSeek V4.1 Flash 1017
Full Changelog: https://github.com/cline/cline/compare/sdk/sdk/v0.0.95...sdk/sdk/v0.2.0
Desktop v0.0.482026-10-17
- GPT-6 models (such as gpt-6-sol and gpt-6.1-sol) work on OpenAI and Azure OpenAI-compatible endpoints. Before, every request was rejected because Cline sent maxtokens instead of maxcompletion_tokens
- The Cline provider's default model is Claude Sonnet 5.5 again, which is back at the top of Cline's recommended list
- Refreshed the model catalog. Adds the CommandCode and Reflection providers. Default models change for 302.AI (Gemini 4 Argon), Eden AI and Prime Inference (Muse Spark Open), CoreWeave (Reflection Beam 1.0), NanoGPT, Pioneer (Nemotron 3.5 Super VL), and Scaleway (DeepSeek V4.1 Flash)
Full Changelog: https://github.com/cline/cline/compare/desktop-v0.0.47...desktop-v0.0.48
CLI v3.0.732026-10-17
- GPT-6 models (such as gpt-6-sol and gpt-6.1-sol) work on OpenAI and Azure OpenAI-compatible endpoints. Before, every request was rejected because Cline sent maxtokens instead of maxcompletion_tokens
- With no model chosen, the Cline provider uses Claude Sonnet 5.5 again, which is back at the top of Cline's recommended list
- Refreshed the model catalog. Adds the CommandCode and Reflection providers. Default models change for 302.AI (Gemini 4 Argon), Eden AI and Prime Inference (Muse Spark Open), CoreWeave (Reflection Beam 1.0), NanoGPT, Pioneer (Nemotron 3.5 Super VL), and Scaleway (DeepSeek V4.1 Flash)
Full Changelog: https://github.com/cline/cline/compare/cli-v3.0.72...cli-v3.0.73