Mellum2.1-12B-A2.5B-Thinking-GGUF
This repository contains GGUF builds of JetBrains/Mellum2.1-12B-A2.5B-Thinking in every quantization we publish, ready to run with llama.cpp, Ollama, LM Studio, and other GGUF-compatible runtimes. Each variant is one file; download the one you need. The repository also ships the model's multi-token prediction (MTP) head for speculative decoding in llama.cpp; see Speculative decoding with the MTP head.
Params
12.0 B
Context
—
Downloads 30d
215 K
Likes
145
Download history
tracking started — chart appears after 7 days of snapshots (5 recorded)215 K downloads in the last 30 days
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| Mellum2.1-12B-A2.5B-Thinking-MTP-Q4_K_M.gguf | Q4_K_M | 8.3 GB | 11.5 GB | ✅ Runs comfortably |
| Mellum2.1-12B-A2.5B-Thinking-MTP-Q6_K.gguf | Q6_K | 11.3 GB | 14.7 GB | ✅ Runs comfortably |
| Mellum2.1-12B-A2.5B-Thinking-MTP-Q8_0.gguf | Q8_0 | 13.4 GB | 17.0 GB | ✅ Runs comfortably |
| Mellum2.1-12B-A2.5B-Thinking-MTP-BF16.gguf | GGUF | 25.2 GB | 30.0 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ ollama run mellum2-1-12b-a2-5b-thinking-gguf # pin the quantization explicitly $ ollama run mellum2-1-12b-a2-5b-thinking-gguf-q4_k_m
$ huggingface-cli download JetBrains/Mellum2.1-12B-A2.5B-Thinking-GGUF-GGUF \
Mellum2.1-12B-A2.5B-Thinking-MTP-Q4_K_M.gguf --local-dir .
$ llama-cli -m Mellum2.1-12B-A2.5B-Thinking-MTP-Q4_K_M.gguf \
-c 8192 -ngl 99 -t 8 --color
$ curl -s https://aimodelscomparison.com/api/v1/models/mellum2-1-12b-a2-5b-thinking-gguf
{
"hf_id": "JetBrains/Mellum2.1-12B-A2.5B-Thinking-GGUF",
"params_b": 12.00,
"context_length": null,
"license": { "id": "apache-2.0", "commercial": "yes" },
"downloads_30d": 214663,
"vram_estimates": [
{ "quant": "GGUF", "gb": 30.0 },
{ "quant": "Q4_K_M", "gb": 11.5 }
],
"updated_at": "2026-10-22T04:00:11Z"
}
Specifications
- Parameters
- 12.0 B
- Licence
- apache-2.0
- First seen on the Hub
- 2026-10-07
- Training datasets
- undisclosed
- Added to our catalog
- 2026-10-21
Compare with any text-generation model