From da68521d5397ed588789e2439f72ba8f49e0df21 Mon Sep 17 00:00:00 2001 From: Classic298 <27028174+Classic298@users.noreply.github.com> Date: Tue, 30 Jun 2026 00:46:42 +0200 Subject: [PATCH] docs: correct phantom 0.9.7 references to 0.10.0 across Computer, Open Terminal and tool-calling docs Native tool calling became the default in Open WebUI 0.10.0 (not 0.9.7), and the {{USER_MESSAGE_ID}} / {{USER_MESSAGE_PARENT_ID}} / {{TASK}} header variables also landed in 0.10.0. There was no 0.9.7 release (0.9.6 was followed by 0.10.0), so every doc claim attributing these to 0.9.7 referenced a version that never shipped. Verified against the upstream CHANGELOG: native-default and the header variables are both 0.10.0 entries. Checked the current tool versions while here (Computer 0.7.1, Open Terminal 0.11.34, Terminals 0.0.4); the docs do not pin tool versions, so no tool-version changes were needed. --- .../chat-features/code-execution/python.md | 2 +- .../chat-conversations/web-search/agentic-search.mdx | 2 +- docs/features/computer/open-webui.md | 4 ++-- docs/features/extensibility/plugin/tools/development.mdx | 2 +- docs/features/extensibility/plugin/tools/index.mdx | 8 ++++---- docs/features/open-terminal/index.md | 2 +- docs/features/open-terminal/setup/connecting.md | 2 +- 7 files changed, 11 insertions(+), 11 deletions(-) diff --git a/docs/features/chat-conversations/chat-features/code-execution/python.md b/docs/features/chat-conversations/chat-features/code-execution/python.md index 84d3597f4..93914cd52 100644 --- a/docs/features/chat-conversations/chat-features/code-execution/python.md +++ b/docs/features/chat-conversations/chat-features/code-execution/python.md @@ -68,7 +68,7 @@ When using **Native function calling mode** with a capable model (e.g., GPT-5, C **Requirements:** 1. `ENABLE_CODE_INTERPRETER` must be enabled globally 2. Model must have `code_interpreter` capability enabled -3. Model must be in **Native** function calling mode (the default as of v0.9.7; only an issue if it was switched to **Legacy** in the model's advanced params) +3. Model must be in **Native** function calling mode (the default as of v0.10.0; only an issue if it was switched to **Legacy** in the model's advanced params) For more details on builtin tools and native mode, see the [Tool Development Guide](/features/extensibility/plugin/tools#built-in-system-tools-nativeagentic-mode). diff --git a/docs/features/chat-conversations/web-search/agentic-search.mdx b/docs/features/chat-conversations/web-search/agentic-search.mdx index d90c17326..338633909 100644 --- a/docs/features/chat-conversations/web-search/agentic-search.mdx +++ b/docs/features/chat-conversations/web-search/agentic-search.mdx @@ -33,7 +33,7 @@ To unlock these features, your model must support native tool calling and have s 2. **Enable Model Capability**: In **Admin Panel โ†’ Settings โ†’ Models**, select your model and enable the **Web Search** capability. 3. **Enable Default Feature**: In the same model settings, under **Default Features**, check **Web Search**. This controls whether the `search_web` and `fetch_url` tools are available by default in new chat sessions. 4. **Confirm Native Mode (Agentic Mode)**: - * Native is the default tool-calling mode as of v0.9.7, so this is already active unless the model was switched to Legacy. In the same model settings, under **Advanced Parameters**, make sure **Function Calling** is set to `Native` (not `Legacy`). + * Native is the default tool-calling mode as of v0.10.0, so this is already active unless the model was switched to Legacy. In the same model settings, under **Advanced Parameters**, make sure **Function Calling** is set to `Native` (not `Legacy`). 5. **Use a Quality Model**: Ensure you're using a frontier model with strong reasoning capabilities for best results. :::info Configuring via environment variables instead diff --git a/docs/features/computer/open-webui.md b/docs/features/computer/open-webui.md index a8bba826b..7c68250c4 100644 --- a/docs/features/computer/open-webui.md +++ b/docs/features/computer/open-webui.md @@ -37,6 +37,6 @@ The `{{...}}` values are Open WebUI's [custom-header placeholders](/reference/en - **Branch correctly.** Regenerating or editing a message in Open WebUI creates a matching branch in the cptr sidebar (via `X-OpenWebUI-User-Message-Id` and `X-OpenWebUI-User-Message-Parent-Id`), instead of appending in the wrong place. - **Ignore background work.** Open WebUI also calls the gateway for title generation, tag extraction and follow-up suggestions. `X-OpenWebUI-Task` tells cptr which requests are background tasks, so it skips them instead of spawning ghost chats or running a full agent loop. -:::caution Requires Open WebUI 0.9.7 or newer -The `{{USER_MESSAGE_ID}}`, `{{USER_MESSAGE_PARENT_ID}}` and `{{TASK}}` placeholders were added in Open WebUI 0.9.7. On 0.9.6 the basic gateway connection still works, you can chat with a workspace as a model, but those three headers are not filled in, so branching and background-task filtering do not work yet. +:::caution Requires Open WebUI 0.10.0 or newer +The `{{USER_MESSAGE_ID}}`, `{{USER_MESSAGE_PARENT_ID}}` and `{{TASK}}` placeholders were added in Open WebUI 0.10.0. On 0.9.6 the basic gateway connection still works, you can chat with a workspace as a model, but those three headers are not filled in, so branching and background-task filtering do not work yet. ::: diff --git a/docs/features/extensibility/plugin/tools/development.mdx b/docs/features/extensibility/plugin/tools/development.mdx index d6c089128..dd543edd5 100644 --- a/docs/features/extensibility/plugin/tools/development.mdx +++ b/docs/features/extensibility/plugin/tools/development.mdx @@ -173,7 +173,7 @@ else: ``` :::caution Do not check for `"native"` or `"default"` -Before v0.9.7 an unset model carried `"default"`, and tools detected Native with `function_calling == "native"`. That is now wrong. An unset model runs **Native** but does not carry the string `"native"`, so a `== "native"` test (or treating a `"default"` fallback as non-native) misclassifies the now-default case as Legacy and feeds it Legacy-only events that silently break. Test for `== "legacy"` and treat everything else as Native. +Before v0.10.0 an unset model carried `"default"`, and tools detected Native with `function_calling == "native"`. That is now wrong. An unset model runs **Native** but does not carry the string `"native"`, so a `== "native"` test (or treating a `"default"` fallback as non-native) misclassifies the now-default case as Legacy and feeds it Legacy-only events that silently break. Test for `== "legacy"` and treat everything else as Native. ::: #### Complete Event Type Compatibility Matrix diff --git a/docs/features/extensibility/plugin/tools/index.mdx b/docs/features/extensibility/plugin/tools/index.mdx index 52e519b61..0f84041f7 100644 --- a/docs/features/extensibility/plugin/tools/index.mdx +++ b/docs/features/extensibility/plugin/tools/index.mdx @@ -104,8 +104,8 @@ Attaching a workspace tool to a model does **not** bypass access control. When a ## Tool Calling Modes: Native vs. Legacy {#tool-calling-modes-default-vs-native} -:::danger Native is the default as of v0.9.7; "Legacy" (formerly "Default") is the unsupported opt-out -**Native (Agentic) Mode is now the default.** As of **v0.9.7**, every chat and model that has not explicitly chosen a tool-calling mode runs **Native**, which relies on the model's built-in function-calling support. The old prompt-injection approach, previously called **Default**, is now renamed **Legacy**. It is the explicit opt-out and remains **unsupported**: no feature work, no bug fixes, no built-in system tools, and incompatible with modern Open WebUI features (Agentic Research, Interleaved Thinking, the built-in Memory/Notes/Knowledge/Channels tools, and web-search/image-gen/code-interpreter tool injection). +:::danger Native is the default as of v0.10.0; "Legacy" (formerly "Default") is the unsupported opt-out +**Native (Agentic) Mode is now the default.** As of **v0.10.0**, every chat and model that has not explicitly chosen a tool-calling mode runs **Native**, which relies on the model's built-in function-calling support. The old prompt-injection approach, previously called **Default**, is now renamed **Legacy**. It is the explicit opt-out and remains **unsupported**: no feature work, no bug fixes, no built-in system tools, and incompatible with modern Open WebUI features (Agentic Research, Interleaved Thinking, the built-in Memory/Notes/Knowledge/Channels tools, and web-search/image-gen/code-interpreter tool injection). **Breaking change:** if any of your models depended on the old prompt-based behavior, they now run Native unless you switch them back to **Legacy**, per chat, per model, or globally in your default model parameters. If a model struggles with Native, the right fix is a stronger tool-calling model, not falling back to Legacy. ::: @@ -128,7 +128,7 @@ Why it is legacy: - **Does not support modern capabilities.** Every new feature shipped since 2024 targets Native Mode. ### ๐ŸŸข Native Mode (Agentic Mode / System Function Calling): The Only Supported Mode -Native Mode (also called **Agentic Mode**) leverages the model's built-in capability to handle tool definitions and return structured tool calls (JSON). It is the **default** as of v0.9.7 and the **recommended mode** for all models that support it, which includes the vast majority of modern models (2024+). +Native Mode (also called **Agentic Mode**) leverages the model's built-in capability to handle tool definitions and return structured tool calls (JSON). It is the **default** as of v0.10.0 and the **recommended mode** for all models that support it, which includes the vast majority of modern models (2024+). :::warning Model Quality Matters **Agentic tool calling requires high-quality models to work reliably.** While small local models may technically support function calling, they often struggle with the complex reasoning required for multi-step tool usage. For best results, use frontier models like **GPT-5**, **Claude 4.5 Sonnet**, **Gemini 3 Flash**, or **MiniMax M2.5**. Small local models may produce malformed JSON or fail to follow the strict state management required for agentic behavior. @@ -143,7 +143,7 @@ Native Mode (also called **Agentic Mode**) leverages the model's built-in capabi - **System Tools**: Only Native Mode unlocks the [built-in system tools](#built-in-system-tools-nativeagentic-mode) (memory, notes, knowledge, channels, etc.). #### How to Enable Native Mode (Agentic Mode) -Native Mode is the **default** as of v0.9.7, so new and existing models use it unless they are explicitly set to Legacy. You can still set `Function Calling` explicitly (to pin Native, or to switch a model to Legacy) at these levels: +Native Mode is the **default** as of v0.10.0, so new and existing models use it unless they are explicitly set to Legacy. You can still set `Function Calling` explicitly (to pin Native, or to switch a model to Legacy) at these levels: 1. **Universal Default for Every Model (Fastest, Recommended)**: * Navigate to **Admin Panel โ†’ Settings โ†’ Models**. diff --git a/docs/features/open-terminal/index.md b/docs/features/open-terminal/index.md index cbc6fddef..af545bbf3 100644 --- a/docs/features/open-terminal/index.md +++ b/docs/features/open-terminal/index.md @@ -80,7 +80,7 @@ Bulk rename, sort, deduplicate, convert, compress, and organize files. Manage di **[Installation โ†’](./setup/installation)** ยท **[Connect to Open WebUI โ†’](./setup/connecting)** :::info Model requirements -Open Terminal needs a model that does tool calling at agentic quality, not just one that technically supports it. Driving a terminal is a multi-step loop: call a tool, read the output, decide the next command, repeat, often for many turns. Small models cannot sustain that. A 4B to 9B model will not cut it here, even if it ticks the "supports tools" box. Frontier models (GPT-5.4, Claude Sonnet 4.6, Gemini 3.1 Pro) handle these complex multi-step workflows well. Native is the default tool-calling mode as of v0.9.7, so it works out of the box unless the model was switched to Legacy; [check the model's tool-calling mode](./setup/connecting#8-enable-native-function-calling) if tools are not firing. +Open Terminal needs a model that does tool calling at agentic quality, not just one that technically supports it. Driving a terminal is a multi-step loop: call a tool, read the output, decide the next command, repeat, often for many turns. Small models cannot sustain that. A 4B to 9B model will not cut it here, even if it ticks the "supports tools" box. Frontier models (GPT-5.4, Claude Sonnet 4.6, Gemini 3.1 Pro) handle these complex multi-step workflows well. Native is the default tool-calling mode as of v0.10.0, so it works out of the box unless the model was switched to Legacy; [check the model's tool-calling mode](./setup/connecting#8-enable-native-function-calling) if tools are not firing. ::: --- diff --git a/docs/features/open-terminal/setup/connecting.md b/docs/features/open-terminal/setup/connecting.md index 0feb89ed1..0b7096577 100644 --- a/docs/features/open-terminal/setup/connecting.md +++ b/docs/features/open-terminal/setup/connecting.md @@ -69,7 +69,7 @@ The selected terminal name appears next to the cloud icon. The AI can now execut ### 8. Enable native function calling -Native function calling is the default tool-calling mode as of v0.9.7, so it is already active for new models. The only thing to check is that the model was not switched to **Legacy**: +Native function calling is the default tool-calling mode as of v0.10.0, so it is already active for new models. The only thing to check is that the model was not switched to **Legacy**: 1. Go to **Workspace โ†’ Models** 2. Click the edit button on the model you're using