# JetBrains ships Mellum 2.1, a reasoning MoE for code

> The updated Mellum packs 12B total parameters but activates just 2.5B, pairing code specialization with a new thinking mode under Apache-2.0.

Published by The Open Weights on Sep 20, 2026. Canonical: https://theopenweights.com/news/mellum2-1-12b-a2-5b-thinking-e9kt

## Key facts

- Company: JetBrains
- Model: Mellum2.1-12B-A2.5B-Thinking
- Version: 2.1
- Category: Code
- Modalities: Code, Reasoning
- License: Apache 2.0 (Open weights, commercial use allowed)
- Parameters: 12B (mixture of experts)
- Active parameters: 2.5B
- Context window: 131K tokens
- Architecture: MellumForCausalLM
- Precision: BF16
- Weights size: 24.3 GB
- Significance: notable
- Published: 2026-09-20
- Last verified: 2026-10-10
- Hugging Face: https://huggingface.co/JetBrains/Mellum2.1-12B-A2.5B-Thinking
- Canonical URL: https://theopenweights.com/news/mellum2-1-12b-a2-5b-thinking-e9kt

JetBrains has released [Mellum 2.1](https://huggingface.co/JetBrains/Mellum2.1-12B-A2.5B-Thinking), the latest version of its code-focused language model line. The new build is a mixture-of-experts design with 12B total parameters but only 2.5B active per token, and it adds a reasoning-oriented "Thinking" variant aimed at harder programming tasks.

The sparse layout is the headline here. By routing each token through a small fraction of the network, an MoE model can offer the capacity of a larger system while keeping inference costs closer to a 2.5B dense model. For a company whose core business is developer tooling, that efficiency matters: code assistance is latency-sensitive and often runs at scale across IDEs.

## Why it matters

- **Permissive licensing.** Mellum 2.1 ships under Apache-2.0, making it straightforward to fine-tune, self-host, or embed in commercial products.
- **Reasoning for code.** The Thinking configuration signals a push toward multi-step problem solving rather than pure autocompletion.
- **Lean active footprint.** At 2.5B active parameters, it is designed to be practical to run outside of heavyweight datacenter setups.

JetBrains has positioned Mellum as a purpose-built family rather than a general-purpose chatbot, and this update continues that focus on software development workloads. Teams evaluating open models for coding can find the weights and details on the [Hugging Face repository](https://huggingface.co/JetBrains/Mellum2.1-12B-A2.5B-Thinking).

## Get the model

- [Hugging Face](https://huggingface.co/JetBrains/Mellum2.1-12B-A2.5B-Thinking)

## Sources

- [JetBrains/Mellum2.1-12B-A2.5B-Thinking](https://huggingface.co/JetBrains/Mellum2.1-12B-A2.5B-Thinking) — Hugging Face, Sep 20, 2026

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "JetBrains ships Mellum 2.1, a reasoning MoE for code", The Open Weights, Sep 20, 2026, https://theopenweights.com/news/mellum2-1-12b-a2-5b-thinking-e9kt