Mirendil Releases Fern 0.1 with 10M Token Code Context

1 reports archived1 independent sourcesupdated 10/27/2026, 21:00:01

Synthesis & Latest Updates

1 Sources Cross-Validated

Mirendil has released its first model, Fern 0.1, designed to handle a 10 million token context window (approx. 500k lines of code). While it achieves perfect single-needle retrieval, it lags significantly behind Claude Opus 5.5 in multi-hop reasoning and terminal benchmarks, currently available only via early access.

LATEST/Mirendil has released its first model, Fern 0.1, designed to handle a 10 million token context window (approx. 500k lines of code). While it achieves perfect single-needle retrieval, it lags significantly behind Claude Opus 5.5 in multi-hop reasoning and terminal benchmarks, currently available only via early access.

Hourly heat curve

3 fully observed hours

Now

15.6

Peak

16.322:00

24h change

–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first

Marktechpost (@Marktechpost)T2Oct 27, 21:00·78 pts
Mirendil Releases Fern 0.1 with 10M Token Code Context
- Fern 0.1 natively supports a 10 million token context window, equivalent to ~23 MB or 500k lines of code.
- Maintains 100% accuracy in RULER single-needle retrieval but drops to 45% in multi-hop reasoning (HotpotQA).
- Built on Mirendil-Base, trained by one researcher using agents in 3.5 months for under $10M.
View Related Reports & Analysis Original Source