whatmodel.sh / Mellum2 12B-A2.5B Instruct

Mellum2 12B-A2.5B Instruct

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Mellum2 12B-A2.5B Instruct is an open-weights language model from JetBrains, released May 2026. Its published size is 12.15B parameters (2.5B active per token) and a 131k-token context window. Licence: apache-2.0.

Catalogue identity notes

JetBrains/Mellum2-12B-A2.5B-Instruct @273d9c1 (created 2026-05-26, licence apache-2.0). JetBrains Mellum2 coding-focused MoE (instruct, May 2026; 64 experts, 8 active, 12B total / 2.5B active per the name). Rows from its own card table with JetBrains' runs of catalogued Qwen3.5 4B/9B, OLMo 3 7B and Ministral 3 14B.

Model information sources

Release date ↗ · Weights licence ↗

HF safetensors metadata 12,149,923,072 parameters; 2.5B active per the maker card.

50.2±1.7#340 of 719 LLMsfrom 8 benchmark results across 1 source

$ how it compares

The LLMs ranked either side of Mellum2 12B-A2.5B Instruct, out of 828 with a score, then more from JetBrains.

#modelscorewhereRAMreleased
408Llama-3-8B-ProLong-512k-Instruct Princeton NLP50±5local~35GB2024-08
409EXAONE Deep 2.4B LG AI Research50±2local~3.2GB2025-03
410Mellum2 12B-A2.5B Instruct JetBrains50±2local~8.6GB2026-05
411OpenCodeInterpreter-DS-33B M-A-P50±1local~22GB2024-02
412Llama 3.2 90B Vision Instruct Meta50±1local~58GB2024-09
more from JetBrains
228Mellum2.1 12B-A2.5B Thinking JetBrains60±2local~8.3GB2026-09
340Mellum2 12B-A2.5B Thinking JetBrains54±1local~8.6GB2026-05

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #177 of 300 * · 10% of its score · 2 tests

Beats o1 (2024-12-17), o3-mini, DeepSeek R1 0528 here, though they rank higher overall. Trails SmolLM3 3B, Jamba 1.5 Mini here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
BFCL V4 (December 2025 archive) lab44.2% ✎▁▁▁▂▇▆▆▆█▂▄▂ beats 60/111medium · avg 36%▇▇▇ 6.7%at its level
BFCL v3 lab66.3% ✎▁▁▁▁▁▃▅█▆▅▄▁ beats 129/170medium · avg 56%▇ 3.6%blunt · at its level
coding▁▁▁▃▆██▇▆▄▃▁ #315 of 458 * · 26% of its score · 2 tests

Beats GLM-4.7-Flash, Qwen3-VL 8B Instruct, Qwen3-VL 4B Instruct here, though they rank higher overall. Trails DeepSeek LLM 67B Chat, WizardLM 70B V1.0, Phi-4-mini 3.8B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
LiveCodeBench v6 (window unspecified) lab37.2% ✎▂▃▄▄▅██▅▄▇▃▆ beats 6/65hard · avg 34%▇▇▇▇▇▇▇▇ 20%at its level
MultiPL-E lab67.1% ✎▃··▄·▅▅▇▄█▃▄ beats 17/31medium · avg 55%▇▇ 5.1%at its level
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #242 of 393 * · 4.0% of its score · 1 test

Beats Mistral Small 4 (119B-A6.5B), Ministral 3 14B Reasoning 2512, Qwen2.5 Coder 32B Instruct here, though they rank higher overall. Trails Flan-T5 XXL, Granite 4.0 H 1B, LFM2.5 1.2B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
IFEval (aggregation unspecified) lab75.8% ✎▁▁▃▁▂▄▄▆█▃▂▁ beats 45/141easy · avg 76%▇▇ 4.0%typical
other tests3 not in a task · 59% of its score
testresultvs the others in its rundifficultycountswhy
GPQA Diamond lab40.9% ✎▁▁▁▁▂▄▅▅███▄ beats 53/310hard · avg 47%▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 34%sharp · at its level
MMLU-Redux lab78.1% ✎▁▁▃▁▃▃▄▅▇▅█▄ beats 8/63easy · avg 78%▇▇▇▇▇▇▇▇ 21%at its level
GSM-Plus lab80.5% ✎▁··▁▁▁·▃▁▄█▁ beats 5/16easy · avg 82%▇▇ 4.0%blunt

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run Mellum2 12B-A2.5B Instruct · needs ~8.2 GB · from huggingface.co/bartowski
    chat in the terminal
    llama cli -hf bartowski/Mellum2-12B-A2.5B-Instruct-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf bartowski/Mellum2-12B-A2.5B-Instruct-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run Mellum2 12B-A2.5B Instruct · needs ~8.2 GB · from huggingface.co/bartowski

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/bartowski/Mellum2-12B-A2.5B-Instruct-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run Mellum2 12B-A2.5B Instruct · needs ~8.2 GB · from huggingface.co/bartowski
    chat in the terminal
    ollama run hf.co/bartowski/Mellum2-12B-A2.5B-Instruct-GGUF:Q4_K_M

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Mellum2 12B-A2.5B Instruct · needs ~6.8 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/Mellum2-12B-A2.5B-Instruct-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/Mellum2-12B-A2.5B-Instruct-4bit

Other ways to run it

  1. Run it as published · needs ~24.3 GB, NVIDIA only · from huggingface.co/JetBrains (official)
    vllm serve JetBrains/Mellum2-12B-A2.5B-Instruct

Official weights and documentation: huggingface.co/JetBrains ↗

$ how the score was worked out · JetBrains/Mellum2-12B-A2.5B-I… 99%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
JetBrains/Mellum2-12B-A2.5B-Instruct model card · lab-reported8▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 99%Apache-2.0
prior · pull toward the middle▇ 1%
weighted average= 50.2

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json