# Mirendil announces Fern v0.1 with 10M-token context window

Digest AI · Generative AI & Models · published 2026-10-26T23:36:24Z

Canonical: https://digestai.news/story/mirendil-announces-fern-v0-1-with-10m-token-context-window

## Summary

Mirendil announced Fern 0.1, its first model pretrained from scratch, featuring a context window of up to 10 million tokens. The company claims the model can hold about 23 MB of text and code at once, equivalent to roughly half a million lines of code or dozens to hundreds of complete agent traces.

Mirendil states that access to the model is coming soon, and it plans to release larger base versions, including Mirendil-Base-Large and Mirendil-Base-XL, in the future.

## Key points

- Fern 0.1 features a context window of up to 10M tokens, handling about 23 MB of text and code
- Access is marked as coming soon, with no public release or API currently available

## Why it matters

A 10 million token context window allows AI agents to analyze entire large codebases and extensive execution logs in a single prompt without complex retrieval systems.

## Sources

1. [Mirendil announces Fern v0.1: Their first model with a 10M context window](https://mirendil.com/news/introducing-fern) (mirendil.com, 2026-10-26)

## Cite

Digest AI, "Mirendil announces Fern v0.1 with 10M-token context window", 26 October 2026, https://digestai.news/story/mirendil-announces-fern-v0-1-with-10m-token-context-window

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/mirendil-announces-fern-v0-1-with-10m-token-context-window.json