# EmEditor v26.2.0 — EmEditor v26.2.0 Released — Enhanced AI Features, Introduction of MSIX Technology, and Faster Performance

- Product: EmEditor (https://whatsnew.fyi/product/emeditor)
- Vendor: Emurasoft
- Date: 2026-06-30
- Version: v26.2.0
- Original notes: https://www.emeditor.com/emeditor-core/emeditor-v26-2-0-released-enhanced-ai-features-introduction-of-msix-technology-and-faster-performance/
- Permalink: https://whatsnew.fyi/product/emeditor/releases/v26.2.0
- Labels: Platforms: Windows

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'.

---

Today, we released EmEditor v26.2.0. 1. Enhanced AI Features In this version, we enhanced the highly requested AI-related features. As in previous versions, the AI-related features fall into two main…

- **added** — Support for additional AI providers including Anthropic, DeepSeek, Google, and LM Studio/OpenAI-compatible providers in AI Writing Assistance
- **added** — Option to show additional information on status bar when prompts are sent to AI or responses are received
- **added** — Option to show AI suggestions only when confident with adjustable confidence level slider
- **added** — Temperature control for AI text generation to adjust randomness and creativity of suggestions
- **added** — Option to prevent AI suggestions after specified end-of-sentence characters
- **added** — Adjustable delay time for automatic AI suggestions in milliseconds
- **added** — Additional instructions field to customize AI suggestion behavior and tone
- **added** — Ability to specify two or more AI providers or models for each AI Prompt to compare responses
- **changed** — AI Writing Assistance now supports multiple AI providers instead of only OpenAI

Today, we released EmEditor v26.2.0.
##### 1. Enhanced AI Features
In this version, we enhanced the highly requested AI-related features. As in previous versions, the AI-related features fall into two main categories.
###### 1-1. AI Writing Assistance
AI Writing Assistance, first introduced in v24.1, has been further enhanced with the following additional options.
- Customize dialog > AI Options
- AI Provider and Model: Previously, AI-assisted writing could use only OpenAI, but starting with this version, you can also select Anthropic, DeepSeek, Google, and LM Studio/OpenAI-compatible providers. For example, by using a local LLM through LM Studio, you can enjoy the benefits of AI-assisted writing without worrying about AI usage costs. When using a paid AI provider that requires an API key, you must specify the API key in advance under Settings in Chat with AI.
- Show additional information on status bar: When this option is turned on, information is shown on the status bar whenever a prompt is sent to the AI or a response is received.
- Configuration Properties > AI Assist
- Show suggestions only when confident: If you do not want to receive AI suggestions too frequently, you can check this option to show suggestions only when the AI is confident. When this is checked, you can use the Confidence level slider below it to set the minimum confidence level required to display a suggestion. Because judging confidence is difficult for AI, and some AI models may not interpret it correctly, we recommend turning this off if you do not need it.
- Set the temperature: In AI, temperature is a parameter that controls the “randomness (creativity)” used when the LLM chooses the next word while generating text. The lower the temperature, the lower the randomness, and the more deterministic the suggestions you receive. The higher the temperature, the higher the randomness, and the more creative the suggestions you receive. When the Set the temperature option is off, the model’s default temperature is always used. When it is on, a temperature of 0.0 is used initially so that you receive deterministic suggestions; however, when you press Ctrl + Space to request a different suggestion, the temperature is raised to produce more creative suggestions. Many recent reasoning models do not accept a temperature parameter, in which case this option is ignored. If an error appears on the status bar because the temperature cannot be set for a particular model, please turn this option off. For AI-assisted writing, we recommend specifying a lightweight model that returns responses quickly, rather than a reasoning model.
- Don’t show suggestions after any of the following: If you check this and enter characters that indicate the end of a sentence in the field below—for example, “.。”—suggestions will no longer appear after symbols such as “.” or “。”. This lets you tailor the feature for easy use according to the type of text you are editing.
- Delay Time: AI suggestions appear automatically after the specified amount of time has elapsed once typing stops and the editor becomes idle. You can adjust this delay time in milliseconds using the slider.
- Additional instructions: Here, you can add instructions such as what kind of suggestions you want to receive, which language to use, and the desired tone. For example, when writing code in JavaScript, you can specify “Use JavaScript,” and if you want to use polite language, you can specify “Write in a polite style.” However, depending on the AI model, these additional instructions may not always be fully understood or produce an appropriate response, and entering too many instructions may slow down the response. Even so, now that more customization is possible, we believe you will be able to fine-tune the behavior to your liking through trial and error.
###### 1-2. AI Prompts
This version enhances the AI Prompts feature, which was first introduced in v24.3. With AI Prompts, you can write a question to the AI in the

_[Truncated at 4000 characters — full notes: https://www.emeditor.com/emeditor-core/emeditor-v26-2-0-released-enhanced-ai-features-introduction-of-msix-technology-and-faster-performance/]_