Installing additional Frameworks/en - Versionsgeschichte

Cg: /* AI (Tensorflow) */

2026-07-12T17:00:37Z

AI (Tensorflow)

← Nächstältere Version		Version vom 12. Juli 2026, 17:00 Uhr
Zeile 423:		Zeile 423:

	== AI (Tensorflow) ==		== AI (Tensorflow) ==
	These are typically used via Python actions, and installed using a Python package manager (eg. "pip install tensorflow").<br>Please refer to Tensorflow web pages for details.		These are typically used via Python actions, and installed using a Python package manager (eg. "<code>pip install tensorflow</code>").<br>Please refer to Tensorflow web pages for details.

	== AI ==		== AI ==

Cg: /* llama.cpp */

2026-07-12T16:59:48Z

llama.cpp

← Nächstältere Version		Version vom 12. Juli 2026, 16:59 Uhr
Zeile 435:		Zeile 435:

	=== llama.cpp ===		=== llama.cpp ===
	Goto [https://github.com/ggml-org/llama.cpp/releases https://github.com/ggml-org/llama.cpp/releases] and download a version for your system. Run "llama-server pull <modelName>" or "llama-server --help" for instructions.		Goto [https://github.com/ggml-org/llama.cpp/releases https://github.com/ggml-org/llama.cpp/releases] and download a version for your system. Run "<code>llama-server pull <modelName></code>" or "<code>llama-server --help</code>" for instructions.

	=== Coding Assistants ===		=== Coding Assistants ===

Cg: /* AI */

2026-07-12T16:58:42Z

← Nächstältere Version		Version vom 12. Juli 2026, 16:58 Uhr
Zeile 436:		Zeile 436:
	=== llama.cpp ===		=== llama.cpp ===
	Goto [https://github.com/ggml-org/llama.cpp/releases https://github.com/ggml-org/llama.cpp/releases] and download a version for your system. Run "llama-server pull <modelName>" or "llama-server --help" for instructions.		Goto [https://github.com/ggml-org/llama.cpp/releases https://github.com/ggml-org/llama.cpp/releases] and download a version for your system. Run "llama-server pull <modelName>" or "llama-server --help" for instructions.

			=== Coding Assistants ===
			You can use any AI assistant (ClaudeCode, OpenCode, Zen, ...). In order for the AI to work with expecco, its "mcp" service must be configured. For the most wellknown assistants, this is done by expecco. For others, google for how to enable mcp for your assistant. Make sure that the server named "mcp" is enabled as http service at "localhost:9876/mcp".

Cg: /* llama.cpp */

2026-07-12T16:54:20Z

llama.cpp

← Nächstältere Version		Version vom 12. Juli 2026, 16:54 Uhr
Zeile 435:		Zeile 435:

	=== llama.cpp ===		=== llama.cpp ===
	Goto [https://github.com/ggml-org/llama.cpp/releases] and download a version for your system. Run "llama-server pull <modelName>" or "llama-server --help" for instructions.		Goto [https://github.com/ggml-org/llama.cpp/releases https://github.com/ggml-org/llama.cpp/releases] and download a version for your system. Run "llama-server pull <modelName>" or "llama-server --help" for instructions.

Cg: /* AI (Ollama) */

2026-07-12T16:53:32Z

AI (Ollama)

← Nächstältere Version		Version vom 12. Juli 2026, 16:53 Uhr
Zeile 425:		Zeile 425:
	These are typically used via Python actions, and installed using a Python package manager (eg. "pip install tensorflow").<br>Please refer to Tensorflow web pages for details.		These are typically used via Python actions, and installed using a Python package manager (eg. "pip install tensorflow").<br>Please refer to Tensorflow web pages for details.

	== AI ~~(Ollama)~~ ==		== AI ==
			=== Ollama ===
	Goto [https://ollama.com/download ollama.com/download] and get the installer for your operating system. The loaded Setup/Install procedure should install ollama on your local machine as a background service.<br>It will also open an Ollama interactive chat and configuration window, which is not needed in the future, but will be useful to verify that the installation worked and especially to download new models.		Goto [https://ollama.com/download ollama.com/download] and get the installer for your operating system. The loaded Setup/Install procedure should install ollama on your local machine as a background service.<br>It will also open an Ollama interactive chat and configuration window, which is not needed in the future, but will be useful to verify that the installation worked and especially to download new models.
	<br>Models are to be placed into the ''Model Location'' folder, which is found in the Ollama window under ''Settings''.		<br>Models are to be placed into the ''Model Location'' folder, which is found in the Ollama window under ''Settings''.
Zeile 432:		Zeile 433:
	Models to consider (as of May 2026): we tried qwen3.6, gemma4 and [https://ollama.com/library/codellama:7b codellama]. We are currently investigating which model is best suited for our AI tasks, but tend to codellama (if you prefer a free model running on your local machine).		Models to consider (as of May 2026): we tried qwen3.6, gemma4 and [https://ollama.com/library/codellama:7b codellama]. We are currently investigating which model is best suited for our AI tasks, but tend to codellama (if you prefer a free model running on your local machine).
	<br>Be aware that models are huge, typically in the order of tens of gigabytes.		<br>Be aware that models are huge, typically in the order of tens of gigabytes.

			=== llama.cpp ===
			Goto [https://github.com/ggml-org/llama.cpp/releases] and download a version for your system. Run "llama-server pull <modelName>" or "llama-server --help" for instructions.

Cg: /* GNU Octave Installation (Matlab compatible Scripting) */

2026-07-12T16:46:37Z

GNU Octave Installation (Matlab compatible Scripting)

← Nächstältere Version		Version vom 12. Juli 2026, 16:46 Uhr
Zeile 385:		Zeile 385:
	If you are already using Matlab, set the path to the matlab executable in the "''Execution''" → [[Settings ExternalToolsSettings/en#GnuOctave_Path \| "''External Script Interpreter''"]] settings.		If you are already using Matlab, set the path to the matlab executable in the "''Execution''" → [[Settings ExternalToolsSettings/en#GnuOctave_Path \| "''External Script Interpreter''"]] settings.

	New in 26.2: now bridged octave/matlab actions are also supported (see [[Expecco_API/en#Bridged_Octave_Elementary_Blocks \| Bridged Octave Elementary Blocks]]. You have to install the octave socket package (<code>pkg install -forge socket</code>)		New in 26.2: now bridged octave/matlab actions are also supported (see [[Expecco_API/en#Bridged_Octave_Elementary_Blocks \| Bridged Octave Elementary Blocks]].<br>You have to install the octave socket package (<code>pkg install -forge socket</code>) for octave blocks.

	==Dart Installation==		==Dart Installation==

Cg: /* GNU Octave Installation (Matlab compatible Scripting) */

2026-07-12T16:45:34Z

GNU Octave Installation (Matlab compatible Scripting)

← Nächstältere Version		Version vom 12. Juli 2026, 16:45 Uhr
Zeile 384:		Zeile 384:

	If you are already using Matlab, set the path to the matlab executable in the "''Execution''" → [[Settings ExternalToolsSettings/en#GnuOctave_Path \| "''External Script Interpreter''"]] settings.		If you are already using Matlab, set the path to the matlab executable in the "''Execution''" → [[Settings ExternalToolsSettings/en#GnuOctave_Path \| "''External Script Interpreter''"]] settings.

			New in 26.2: now bridged octave/matlab actions are also supported (see [[Expecco_API/en#Bridged_Octave_Elementary_Blocks \| Bridged Octave Elementary Blocks]]. You have to install the octave socket package (<code>pkg install -forge socket</code>)

	==Dart Installation==		==Dart Installation==

Cg: /* AI (Tensorflow) */

2026-05-25T14:45:12Z

AI (Tensorflow)

← Nächstältere Version		Version vom 25. Mai 2026, 14:45 Uhr
Zeile 421:		Zeile 421:

	== AI (Tensorflow) ==		== AI (Tensorflow) ==
	These are typically used via Python actions, and installed using a Python package manager (eg. "pip install tensorflow"). Please refer to Tensorflow web pages for details.		These are typically used via Python actions, and installed using a Python package manager (eg. "pip install tensorflow").<br>Please refer to Tensorflow web pages for details.

	== AI (Ollama) ==		== AI (Ollama) ==

Cg: /* AI (Ollama) */

2026-05-25T14:44:47Z

AI (Ollama)

← Nächstältere Version		Version vom 25. Mai 2026, 14:44 Uhr
Zeile 425:		Zeile 425:
	== AI (Ollama) ==		== AI (Ollama) ==
	Goto [https://ollama.com/download ollama.com/download] and get the installer for your operating system. The loaded Setup/Install procedure should install ollama on your local machine as a background service.<br>It will also open an Ollama interactive chat and configuration window, which is not needed in the future, but will be useful to verify that the installation worked and especially to download new models.		Goto [https://ollama.com/download ollama.com/download] and get the installer for your operating system. The loaded Setup/Install procedure should install ollama on your local machine as a background service.<br>It will also open an Ollama interactive chat and configuration window, which is not needed in the future, but will be useful to verify that the installation worked and especially to download new models.
	<br>Models to consider: we tried qwen3.6, gemma4 and [https://ollama.com/library/codellama:7b codellama]. We are currently investigating which model is best suited for our AI tasks, but tend to codellama (if you prefer a free model running on your local machine).<br>Models are to be placed into the ''Model Location'' folder, which is found in the Ollama window under ''Settings''.		<br>Models are to be placed into the ''Model Location'' folder, which is found in the Ollama window under ''Settings''.
	<br>You can also pull models with "<CODE>ollama pull <modelName></CODE>" or "<CODE>ollama run <modelName></CODE>". On Windows, the ollama command is typically found in your personal "<CODE>AppData\Local\Programs\Ollama</CODE>" folder.		<br>You can also pull models with "<CODE>ollama pull <modelName></CODE>" or "<CODE>ollama run <modelName></CODE>". On Windows, the ollama command is typically found in your personal "<CODE>AppData\Local\Programs\Ollama</CODE>" folder.

			Models to consider (as of May 2026): we tried qwen3.6, gemma4 and [https://ollama.com/library/codellama:7b codellama]. We are currently investigating which model is best suited for our AI tasks, but tend to codellama (if you prefer a free model running on your local machine).
	<br>Be aware that models are huge, typically in the order of tens of gigabytes.		<br>Be aware that models are huge, typically in the order of tens of gigabytes.

Cg: /* AI (Ollama) */

2026-05-25T14:33:12Z

AI (Ollama)

← Nächstältere Version		Version vom 25. Mai 2026, 14:33 Uhr
Zeile 427:		Zeile 427:
	<br>Models to consider: we tried qwen3.6, gemma4 and [https://ollama.com/library/codellama:7b codellama]. We are currently investigating which model is best suited for our AI tasks, but tend to codellama (if you prefer a free model running on your local machine).<br>Models are to be placed into the ''Model Location'' folder, which is found in the Ollama window under ''Settings''.		<br>Models to consider: we tried qwen3.6, gemma4 and [https://ollama.com/library/codellama:7b codellama]. We are currently investigating which model is best suited for our AI tasks, but tend to codellama (if you prefer a free model running on your local machine).<br>Models are to be placed into the ''Model Location'' folder, which is found in the Ollama window under ''Settings''.
	<br>You can also pull models with "<CODE>ollama pull <modelName></CODE>" or "<CODE>ollama run <modelName></CODE>". On Windows, the ollama command is typically found in your personal "<CODE>AppData\Local\Programs\Ollama</CODE>" folder.		<br>You can also pull models with "<CODE>ollama pull <modelName></CODE>" or "<CODE>ollama run <modelName></CODE>". On Windows, the ollama command is typically found in your personal "<CODE>AppData\Local\Programs\Ollama</CODE>" folder.
			<br>Be aware that models are huge, typically in the order of tens of gigabytes.