---
title: JetBrains/Mellum2.1-12B-A2.5B-Thinking-MTP
url: https://recentmodel.com/models/JetBrains/Mellum2.1-12B-A2.5B-Thinking-MTP/
status: just landed
published: 2026-10-30T17:20:32Z
updated: 2026-10-30T23:18:11Z
source: https://huggingface.co/JetBrains/Mellum2.1-12B-A2.5B-Thinking-MTP
---

# JetBrains/Mellum2.1-12B-A2.5B-Thinking-MTP

Mellum2.1 is the next version of Mellum2 Thinking. Mixture-of-experts, 12B total, 2.5B active (64 experts, 8 per token). Context: 128K tokens (config). License: apache-2.0.


- **Org:** JetBrains
- **Released:** 2026-10-30
- **Status:** just landed
- **License:** apache-2.0 (permissive)
- **Parameters:** 12.1B total, 2.5B active
- **Architecture:** MoE · 64 experts, 8/token · MellumForCausalLM
- **Context:** 131,072 tokens
- **Task:** text-generation · text
- **Signal:** 2 HF likes · 0 downloads

## Runs on

real quant files; sizing calculation, not tested.

- Q4 ~10.5 GB: 12GB GPU (RTX 3060), Mac 16GB
- Q8 ~14.9 GB: 16GB GPU (5060 Ti / 4060 Ti / 4080 / 5080), 24GB GPU (3090 / 4090), 16GB GPU (7800 XT / 9070 XT), 24GB GPU (7900 XTX), Arc B60 24GB, Mac 24GB, CPU only, 32GB RAM (slow)
- FP16 ~26.7 GB: 32GB GPU (5090), 48GB GPU (RTX 6000 Ada / A6000), 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), Strix Halo 128GB unified (~96GB usable), Arc B70 32GB, Mac 36GB, Mac 48GB, Mac 64GB (M1 Max), Mac 96GB, Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 64GB RAM (slow), CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)

## Links

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